The basic idea of Basel 3.1 / IV/ End Game is to close the gap between banks with their own capital models (IRB banks) and banks using off-the-shelf regulatory stipulations (standardised approach), using the “output floor” to limit the benefits of IRB.

The regulatory worry is that letting banks mark their own homework on capital leads to undercapitalised banks — though IRB is more risk-sensitive, and avoids the cliff effects generated by banks optimising their balance sheets under a set of standard credit risk weights.

The PRA document matters because it amends the risk weights applicable to different categories of lending, with a particularly large potential impact in mortgages. The big high street lenders are all on IRB, and can therefore support vast amounts of prime mortgage lending with minimal equity capital.

As a random example, here’s Nationwide, running a £168bn portfolio with £24bn of risk-weighted assets.

Nationwide actually has a very healthy 26.5% CET1 ratio, but its regulatory minimum is 11.5%. So it could, in theory, support its £168bn prime portfolio with £2.76bn of equity capital — £1.60 of capital for every £100 of mortgages.

This sounds like not very much, but the last Silverstone Master Issuer investor book shows annual losses of 0.007% since April 2014, on a presumably representative sample of the Nationwide prime book.

But the point is, it’s very hard for a bank on standardised risk weights to compete with this kind of return on regulatory capital. OneSavings Bank, to take a challenger example, has an average risk density of 37.3% on its portfolio.

If a bank is going to allocate 35% risk-weight to its mortgage book, then it might as well try to get paid for it — and hence most of the challenger institutions on standardised risk weights concentrate on better-yielding parts of the mortgage universe, such as buy-to-let.

For similar yield-related reasons, this is the same market where the UK’s securitisation-funded non-banks are active, and the rapidly rising rates cycle of the last two years (plus Liz Truss etc) has tilted the balance of power in favour of the deposit-funded balance sheets of the challenger banks, and against the securitisation-funders.

So the part of the new UK rules which really matters for securitisation is the border between challenger bank products and those which might be better funded in the market. The PRA’s initial proposals, which we discuss here, proposed different risk treatment for properties which were “materially dependent on cash flows from property”, the intuition being that these look much more like commercial properties than residential. So single property buy to let with a guarantee from the owner looks like residential, while multi-property limited company, holiday lets and portfolio landlords look more like small companies or CRE. Student housing, too, is residential in that people live in it, but has a different risk profile from owner-occupied resi.

All else being equal, this will make it harder for regulated banks to compete in these products, opening the door to more market-based securitisation funding on these products, at the same time as the challenger banks see a levelling of the playing field in prime owner-occupied or low LTV prime buy-to-let.

The final rules don’t change much on this score; indeed, the PRA felt the need to clarify that student accommodation, care homes and holiday lets were not to be considered “residential real estate” exposures. It did, however, bow to pressure on self-build mortgages, exempting these from an onerous requirement that mortgages be backed by “finished properties” to qualify for the best treatment. The properties will be subject to a valuation haircut, but otherwise considered resi.

The PRA has kept the proposed three property limit for portfolio landlords in place (though tweaked the details), but will no longer including Houses in Multiple Occupation in the list of lending which is “materially dependent on cash flows from the property”. Lenders will need to treat this case by case.

For all exposures which are dependent on cashflows from the property, the PRA has introduced new risk weighting bands compared to the draft proposals, but this is unlikely to change much; instead of a 45% risk weight for 60%-80% LTV, it’s 40% for 60%-70% and 50% for 70%-80%.

Impacting all lenders, though, will be changes to the valuation assumptions on housing (and hence on the LTV, which feeds through to risk weight). Lenders will be required to revalue on a five-year time horizon, or following a major market downturn. As most of the UK mortgage stock is on two or five-year fixed periods, this won’t have a huge impact on regular mortgages, but could be a big benefit for legacy pre-crisis collateral, as well as lifetime / equity release product. House prices have risen, so these products could receive better risk treatment on bank balance sheets.

Now the rules are final-ish, the commercial decisions can get underway. This is not likely to blow up anyone’s business, but it will redraw the competitive landscape in the specialist end of UK housing finance…. hopefully in favour of securitisation.

Greener pastures

Last week’s storming specialist lender triple-whammy (Foundation Home Loans, Together and Belmont Green all printing within a couple of days, and all scoring strong executions) was a fine thing to behold, but may be tinged with a little regret for the mezz buyers in Tower Bridge Funding 2024-3, as this could be their last chance for allocation in the shelf.

Belmont Green has been on a long journey towards a banking licence, and it’s very nearly there. Three outings this year is a record securitisation slate for the lender, but once it scores the banking licence, expect a different style of RMBS execution, focused on senior issuance for funding.

Charter Court / OneSavings Bank has already been on this journey. Old Charter Court / Precise deals were full-stack fully levered deals with residuals that were actively traded. Now it funds in the market via CMF (owner-occupied prime) and PMF (buy-to-let) RMBS programmes, offering single tranches only (and buys RMBS for its treasury book).

At least there are deals. Paragon, icon, legend of the pre-2008 market, first UK company to issue RMBS in 1987, more than 50 deals before the financial crisis….. and it’s basically absent today. It last issued a deal at the back end of 2023, which was retained, and positioned as Bank of England collateral. One of the mightiest BTL issuers is dormant.

Once Belmont Green fires up the deposit machine, there’s a good chance this funding will be cheaper than securitisation, and more flexible as well; it won’t need to originate loans fitting neatly into warehouse criteria. There’s still a value, especially for a new bank, in showing supervisors access to wholesale markets, and in terming out funding, so expect some RMBS deals. But 2024 is likely to be the high water market of Tower Bridge supply, especially down the capital structure.

We should also congratulate Together Financial Services, partly on last week’s solid trade, but also on being 50 years young today, and on naming a new CEO, Richard Rowntree, most recently of Paragon Bank. Founder and majority shareholder Henry Moser will be executive vice chairman.

One of the great questions in the UK specialist lender landscape has been “will anyone buy Together?”, though it should really be rephrased as “will Henry Moser ever sell Together?”. There’s been plenty of interest, but persuading a founder-owner to part with their beloved company isn’t easy.

If anything, the Basel changes above should help Together gain further share, so here’s to a few more years in securitisation.

Let’s talk (more) about mortgages

I’ll be at DealCatalyst’s Annual UK Mortgage Finance conference on Wednesday, at the Landmark Hotel in Marylebone, where once again 9fin is lead media partner. As ever it should be an excellent event, especially as I’m not moderating anything this year, and I’m assured they’re still welcoming registrations.

If you want to whet your appetite for some mortgage-related discussion, you can check out a couple of interviews I’ve done with speakers at the event — Scott Robertson, head of mortgage solutions at Phoenix Group, giving the perspective of one of the biggest players in equity release, and Graham McClelland, CFO at Generation Home, who’s seeking to broaden access to housing finance through a range of mortgage innovations. We’ll shortly be sending out an interview with Doug Charleston, portfolio manager at TwentyFour Asset Management as well.

Find the interviews here or on the DealCatalyst and 9fin LinkedIn profiles.

Whatever it takes (to get securitisation back)

Another week, and another European bigwig backs a real reform of securitisation rules. They don’t come much bigger than Mario Draghi, saviour of the eurozone and former prime minister of Italy. One could argue the toss about whether it was the Draghi-era wall of cheap money during his 2011-2019 tenure as ECB president which supressed bank-issued securitisation when it was supposed to be bouncing back, but the important thing is, he’s on board now.

Per Draghi’s EU Competitiveness Report, circulated on Monday (9 September), “Securitisation makes banks’ balance sheets more flexible by allowing them to transfer some risk to investors, release capital and unlock additional lending. In the EU context, it could also act as a substitute for lack of capital market integration by allowing banks to package loans originating in different Member States into standardised and tradeable assets that can be purchased also by non-bank investors”.

Great, great, inject it into my veins.

He also gets into some specifics: “This report recommends that the Commission makes a proposal to adjust prudential requirements for securitised assets. Capital charges must be reduced for certain simple, transparent and standardised categories for which charges do not reflect actual risks.

In parallel, the EU should review transparency and due diligence rules for securitised assets, which are relatively high compared to other asset classes and reduce their attractiveness. Setting up a dedicated securitisation platform, as other economies have done, would help to deepen the securitisation market, especially if backed by targeted public support (for example, well-designed public guarantees for the first-loss tranche).”

The vibes here are definitely good. Over at PCS, Ian Bell writes: “This long-awaited report cannot but be a welcome addition to the now long series of such analyses invariably concluding in the need for Europe to retrieve a stong securitisation market.”

But some of it is worth discussing a little more. There’s a more detailed “in-depth analysis” companion document, which also suggests “securitisation products with ships as collateral assets”, and also notes “The EU lacks the equivalent of US government-sponsored enterprises (GSEs). GSEs have been crucial in fostering the standardisation of mortgage products across American banks and States, reducing transactions costs, lowering credit risks for both banks and buyers, and building a large and deep market.”

This must be roughly what’s meant by “a dedicated securitisation platform”, and echoes proposals in the Noyer Report, which even comes with a diagram:

If the United States did not already have the GSEs, would it feel the need to invent them? They’ve been in conservatorship since the financial crisis, suggesting there aren’t any easy answers about what to do with them in future. They do allow a product that doesn’t exist in the wild, the 30-year prepayable mortgage, to flourish, but at the cost of a massive taxpayer backstop for more or less the entire prime mortgage market. US banks have fewer mortgages around, so do proportionally more corporate lending, and appear healthier and more profitable. But the Fannie and Freddie balance sheets are north of $3trn each.

If the EU is willing to write €6trn in guarantees, for about three quarters of the outstanding mortgage stock (or even the €1.5trn discussed by Noyer) then it could indeed achieve quite a lot, but this is using a sledgehammer to crack a nut. How many potential mortgage borrowers are being shut out of the market for lack of a government guarantee? How many bankable, creditworthy SMEs cannot borrow because banks have too many mortgages? The EU’s emergency Covid scheme was worth some €800bn, and that was a response to a disease that shut down huge sectors of the economy…. I’ve no doubt it would work, but if you can get agreement on a multi-trillion guarantee package, isn’t there something more worthwhile one could do with it?

This is slightly unfair to the GSEs, which have remodelled themselves over the past decade with a massive programme of credit risk transfer under the STACR and CAS shelves. That makes them look a lot more like the Noyer schematic above, in which private investors take junior credit risk and the state guarantees the senior. The Draghi report mentions “well-designed public guarantees for the first loss”, so perhaps there’s some disagreement to iron out there. Either way, it still looks like a solution in search of a problem.

Much as I hate to praise covered bonds to this audience, if the goal is to turn mortgages into a standardised rates product (without a massive taxpayer backstop), the Danish system seems to work ok?

I’d also quibble a bit with the figures. Draghi states that EU securitisation issuance is 0.3% of GDP, vs 4% in the US. I make that €47.4bn, vs the €15.8trn EU GDP, while the companion piece says it uses Afme data, which gives €94.7bn placed in 2023.

But this ignores the large and healthy SRT market, where data is harder to come by. The International Association of Credit Portfolio Managers estimated €203bn of notional in 2023, of which about half is EU banks. Conservatively, that more than doubles the issuance figure for the year, and folding in private cash transactions would take it another leg higher, probably not to 4% of GDP, but to a less anaemic base at least.

So the focus should be: what’s so attractive about the SRT market vs cash? Cash securitisation is bopping along and SRT is galloping, growing at double digits year-on-year. Reviving “securitisation” broadly considered should involve looking at the characteristics of the healthy market, and seeing which aspects could be useful for the less healthy part.

Horsing around

In the spirit of examining healthy parts of the securitisation market looking for clues, consider Lloyds.

The UK bank already went through a heavy period of derisking in the immediate post-crisis years under government ownership, cleaning up the mess of the HBOS merger, but in the past couple of years it has really fired up its “balance sheet optimisation” initiatives, with securitisation as the centrepiece, selling nearly £10bn of risk across legacy mortgages, consumer loans and auto finance.

Now, we understand, it’s back in the market with a portfolio of reperforming assets, working with Alantra to sell the book.

Admittedly, this programme has not resulted in very much widely placed securitisation supply, and the list of winning counterparties for Lloyds disposals is very short.

But this is exactly the sort of thing that Draghi means when he says “Securitisation makes banks’ balance sheets more flexible by allowing them to transfer some risk to investors, release capital and unlock additional lending.”

Lloyds has a pretty healthy balance sheet, so it’s not like it has to scratch around for any crumb of incremental capital. But it’s decided to get ahead of the Basel changes, staff up properly and proactively manage the balance sheet using the tools available to it.

Some of the trades focus on regulatory capital, some optimise for stress testing, some optimise for provisioning, some for accounting deconsolidation. Sometimes all at once. For all that securitisation regulation might be kind of annoying, it’s clearly not blocking banks from using it in precisely the way that the great and the good appear to want.

One could say the same for Santander, with a massive risk transfer programme across cash and synthetic structures from multiple subsidiaries. Or BNP Paribas, or Societe Generale for that matter.

All that it would take for the “revival” of European securitisation is for these institutions to become the rule, rather than the exception, and to see similar active disposals and optimisations in similar proportions from other European or UK banks.

So what’s special about Lloyds? Being big helps. The fixed costs of a £100m disposal aren’t that different to the fixed costs of a £1bn disposal, and securitisation regulation probably makes them higher than is strictly necessary. Doing yards at a time helps support a big and sophisticated treasury function, which was, until recently, run by a banker of impeccable securitisation pedigree, Cecile Hillary (she’s now going to be CFO of Danske Bank, so I guess, watch for more optimisation on that front?).

Having an active and engaged securitised products group in the investment bank also helps. Bankers want to do deals, and are happy to do them for internal or external clients. That applies to the likes of Santander and BNP Paribas as well; driving SRT programmes out of the investment bank turns them from an occasional nice to have into active programmes scouring the balance sheets of subsidiaries for suitable portfolios.

So while amending risk weights and P-factors and matching adjustments and all sorts of regulatory tinkering helps, the biggest factor in securitisation supply, it seems to me, is that banks have to want to do it, which is as much a question of culture and organisation as it is of regulation.

Enjoyed this weekly wrap? Our customers receive this content ahead of the crowd — find out more about 9fin’s news and analysis.