Primary market: after a busy 1H, we have seen a significant slowdown in 2H
After nearly EUR15bn of EUR benchmark covered bond supply in September (up by EUR4bn vs. September 2023 but EUR6bn less than in 2022) and EUR 9.4bn in October, YtD supply now stands at c.EUR154bn, lagging last year's figures at this point in the year by EUR25bn (14% YoY), with net supply YtD inching up to EUR53bn driven by low redemption volumes between August and October vs. EUR64bn of net supply as of end-October 2023.
Issuance is likely to remain muted in the coming weeks with some banks still in blackout periods and the US elections. November EUR supply in the past seven years has averaged EUR10bn but we think this year, given possible volatility surrounding US elections, supply could be more erratic. After EUR7bn of EUR CB redemptions in October, covered bond redemptions will increase in November (EUR11.25bn).
The negative funding gap in 2024 has helped limit issuance activity. Deposit and lending flows continue to produce net liquidity surpluses rather than leading to additional funding needs.
When looking at YtD supply, French banks have been the most frequent issuers pricing EUR36.25bn in 37 transactions, followed by Germany (EUR28.5bn), Canada (EUR13.5bn, significantly below the EUR17.75bn issued in 2023 and the EUR30.25bn issued in 2022), Italian issuers (EUR10.2bn) and Dutch banks (EUR9.65bn). Supply from Spain (EUR2.85bn) has been underwhelming. Hence, the YtD supply has been driven by French and German issuers. Canadian issuers account for 9%, while the shares of Italy, the Netherlands, Norway and Austria are in a range of between 4% and 7%.
Non-EUR covered bond issuance below the historical average: the GBP covered bond market now stands at GBP13.3bn YtD. In terms of countries, UK issuers account for half of the year-to-date GBP covered bond supply, followed by banks from Canada (24%), Singapore (15%) and Australia (11%). On the other hand, covered bond issuance in USD has remained lacklustre this year and there have only been three covered bond transactions. At USD3.7bn, YtD issuance represents only 22% of last year’s supply volume of USD17.2bn and only 13% of the exceptionally high supply volume of USD29.5bn registered in 2022.
While the first half of 2024 saw the re-opening of the long end, with 20% of deals priced in tenors longer than 10Y (with the longest issued CB this year a 15Y) and with 49% of the year-to-date covered bond supply in EUR benchmark format having maturities of seven years or more (vs. 26% in 2023), we have seen a much more conservative approach in the second half of the year due to volatile rates markets.
The current sweet spot of investor demand seems to be new issues with maturities of three to six years. Notably, almost 30% of new issuance since the beginning of June 2024 had short maturities of up to four years. There has been no supply with tenors beyond seven years in October.
Rising loan demand in 2025 could be a factor driving increased issuance. The 3Q24 bank lending survey shows further signs of recovery. The majority of banks not only report an increase in demand for housing loans, but also expect it to increase further in the coming months as lending conditions ease. The faster-than-expected decline in ECB interest rates and lower inflation should support the trend heading into next year, despite the weakness of the economy, and should provide covered bond issuers with fresh cover assets and potentially increasing funding needs in 2025.
We expect 2025 issuance activity to be moderately up from the 2024 figures, as covered bond redemptions in 2025 will rise by EUR15bn to c.EUR130bn. After a low point in 2024, redemptions will increase again next year, and will be even higher in 2026 at EUR150bn and EUR183bn in 2027. As such, our initial estimate for next year is EUR170-190bn of covered bond supply in EUR benchmark format, buoyed by improved mortgage origination dynamics, a growing issuer base and 2026 pre-funding.
2024 has been the year of the comeback of real-money investors and particularly AMs to the asset class
Bank treasuries remain the most active investors in covered bonds, buying on average 44% of new issues (vs. 51% in 2023), but their share is declining as they become more selective given relative value considerations and their already large covered bond holdings. The pick-up that covered bonds offer over SSAs (particularly beyond the 7Y tenor) no longer looks attractive.
The comeback of the credit/AM investors has been one of the most noticeable trends in 2024, enticed by improving liquidity, high yields and the higher spreads on offer. The increase in yields this year and the more marked differentiation between the various issuers and market segments have resulted in much higher demand from asset managers, rising from 24% in 2023 to 31% YtD in 2024.
The extent of asset manager (and hedge funds) participation this year given the (still) attractive RV of covered bonds versus senior preferred was rightly questioned in terms of its sustainability, given it was instrumental in the tightening of covered bond spreads earlier this year after two years of monotonic wider repricing. The general perception is that covered bonds now have better structural demand and even credit accounts are willing to consider the asset class on a more ‘permanent’ basis given its improved liquidity and valuation post-QE, as well as its distinctively lower volatility.
Secondary spread dynamics: where to look? Attractive RV vs. senior financial debt? Multi-year tights in the Bund-swap spread? SSAs wider than ASW spreads?
The YtD spread dynamics in covered bonds have in general terms followed the traditional seasonal pattern of a better primary market and secondary spread performance trends in the first half of a year than the second, with the period between February and April particularly benign. The significant tightening of Bund swap spreads is negative for covered bond swap spreads, and CBs now look expensive vs. SSAs but cheap vs. credit.
Senior preferred bonds' premium to covereds has been narrowing since September 2023 and reached a low of c.35bp in April, but after a slight increase to c.47bp in August is now back at c.36bp.
Covered bonds spreads have been widening since early June, driven mostly, but not exclusively, by France. Covered bond swap spreads have generally come under pressure in recent weeks, but in no country have spreads widened as much as in France. Although the French sector had already significantly underperformed following the EU and national elections, it has remained vulnerable. This said, and despite the overall weakness, we have seen Canadian, Norwegian and UK EUR-denominated covered bonds outperforming.
As measured by the iBoxx country index, French covered bonds are now trading around 13bp cheaper on average than four months ago. This represents a spread widening of more than 40%, which is significantly higher than in other countries. The reason for this distortion is, of course, not to be found in the covered bond market or in the banks, but in the government/political arena.
SSAs such as the EU have underperformed YtD vs. covered bonds, although the biggest moves have been in the long part of the curve, where CBs are more scarce (only one 15Y transaction so far this year from CAFFIL). Moreover, and as a result of the widening of French OATs, French covered bonds have outperformed OATs and the spread pick-up of French CBs to OATs has declined.
The Bund-swap spread has tightened significantly in recent weeks, meaning that German Pfandbriefe now trade at a lower premium to Bunds than in early July, with a pickup vs. Bunds of in the 30-35bp range in the short-tenors and 40-50bp in longer maturities. These levels are at the lower bound of the Bund spread range since 2023.
The rapid tightening of Bund swap spreads is negative for covered bond swap spreads, as SSA bonds are likely to remain under pressure versus swaps amid a declining pick-up over government bonds. As such, the gradual widening of EUR covered bonds versus swaps that has characterised recent months is, therefore, unlikely to be over, although the correlation between Bund swap spreads and covered bonds is a complex one and variable in time and depends, among other factors, on how expensive covered bonds are perceived relative to SSAs, EGBs and senior preferred financial debt.
Covered bond spreads still look good value versus credit, but with EGBs and EUR SSAs widening, covered bonds are generally less attractive from a relative value perspective. As such, although covered bond spreads are wide compared to historical levels, they are tighter than usual relative to EUR SSAs, which have been on a widening trend, particularly since August.
At the moment, covered bonds show a significant dislocation vs. senior preferred debt, which, frankly speaking, have been extremely resilient to the elevated country-risk premium.
Such a performance anomaly is likely possible because of the not so significant overlap of investors engaging in relative value opportunities between covereds and seniors, compared to those who look at covereds through a comparative lens of SSAs. However, for those who might consider covered bonds as a ‘credit’ alternative, we find it convincing enough to look at the current, much wider spreads of covered bonds, as senior preferreds offer an asymmetric risk reward at this point, in our view, with little room to rally further on any positive news but sizable room to reprice wider on any negative macropolitical development.
As such, anecdotal evidence suggests that, asset managers (AMs) are naturally a lot heavier in their participation in credit products, including banks’ senior preferred, thereby bringing the element of broader credit outperformance into the sector. In contrast, covered bonds typically flow to more rates-heavy investors (such as bank treasuries) and the spread performance in rates products has been challenged.
The sweet spot of covered bonds now is up to 5Y, but investors prefer SSAs beyond 5Y. While covered bond spreads have stabilised recently, and they have widened much less month-to-date compared to the significant weakness seen in September, covered bond valuation is still challenged by SSAs (such as the EU), which have continued to gap wider. However, we think it is notably more relevant at the long-end, while short- to-intermediate maturity covereds (i.e., 5-6Y and under) have spread support from the still-positive pickup versus SSAs and the very attractive RV versus senior preferred.
Headwinds and tailwinds in 2025 for covered bond spreads
(+) Supply-demand balance for covered bonds is much better now versus the past two years, with both a moderation in supply pace in 2024 YtD and an increase in asset manager participation this year.
(+) We believe asset manager participation, including from the traditionally credit-focused investors who were instrumental earlier this year in turning around covered bond spreads, will be sustained, given the relative value (RV) of covered bonds versus senior preferred.
(-) On the other hand, given the current weakness in covered bonds has been induced by SSAs, with the latter themselves repriced meaningfully wider versus swaps, to the extent that we believe a lot of wider repricing is likely behind us. Going into 2025, we see the relative pricing vs. EGBs and SSAs (effectively the EU) as the main risk for covered bonds.
(+) Covered bonds rating stability vs. sovereign downgrades: Fitch’s recent outlook change of French sovereign rating has a potentially very limited adverse impact on covered bond ratings, which are protected by ample buffers (3 to 6 notches, as of earlier this year) against any IDR downgrades. Regarding the senior ratings, the French banks IDRs possess a one-notch buffer against a French sovereign downgrade (to ‘A+’ from ‘AA-’), but further downgrade of the sovereign might lead to a downgrade of the IDRs.
(+) Less headline risk at the issuer level: Amidst an otherwise resilient backdrop for European bank fundamentals, one of the only real asset quality ‘themes’ or stress-points to emerge in the recent past has been commercial real estate risk. Nordic lenders took the first ‘blow’, suffering the read-across from investor focus on SBB and other Swedish real estate borrowers’ financial distress in 2022. For a brief period, Austrian banks then followed, owing to the collapse of Austrian property company Signa in November 2023, as well as the more direct spotlight thrown on Bawag’s CRE exposures earlier in the year by activist investor reports. Finally, and most acutely, German banks suffered the fallout from the materialisation of US commercial property risks earlier in 1Q24, and with it, heightened investor caution around the read-across to German commercial real estate.
As rates decline, pressure on CRE-exposed borrowers will gradually fade away. Issuer rhetoric does now appear to point to an inflection point being reached. Deutsche Bank’s management sought to assure investors that CRE risk provisioning headwinds were seen as having peaked in the third quarter, to normalise here onward. Closer to the maelstrom, specialist lender Deutsche Pfandbriefbank’s bespoke pbbIX index report also pointed to a stabilisation in its index after six consecutive quarters of decline, suggestive of a potential ‘bottoming out’ in valuation. In the Nordics, unsecured debt market access for CRE borrowers (the likes of Castellum, SBB, etc.) has also notably increased, signalling some rehabilitation in underlying fundamentals.
CRE exposure, less of an issue than in 2024: a lot has already been said with little in the way of real news coming to light. Provisioning for US offices seems to have peaked, what is still missing to boost market confidence is a pick-up in transactions. As such, issuer event risk was not seen as leading to contagion and a wider exodus of investors from covered bond markets.
The market seems to have digested the CRE qualms better over the last few months, with the focus now squarely on US Office assets rather than adopting an ‘any-and-all CRE’ approach.
(-) Tighter Bund-swap spreads, a negative for covered bonds: with 10Y swap spreads dropping to levels inside 10bp as of late October and approaching the zero line, levels not seen since 2004, it is unclear whether we are starting to get to the end of the tightening trend in the Bund-swap spread we have seen since late 2022, as some analysts are already pencilling in negative swap spreads. The 10Y Bund-swap spread has only reached such low levels on two occasions since 2007, for a relatively short period at the beginning of 2010 and in the middle of 2014. Beyond 10Y, we are firmly negative and while 30Y had been negative for some time, 15Y swap-spreads are now also well inside -10bp. While negative spreads at the long end (20Y and beyond) do not really matter much in terms of the pricing of covered bonds, they matter much more for 5-10Y CBs.
Although there have been episodes when the performance of covered bond spreads has temporarily decoupled from tight Bund-swap spreads, SSAs do usually have a harder time decoupling from the Bund-swap spreads than covered bonds, and if SSAs were to remain under widening pressure in order to maintain an attractive pick-up over Bunds, this would in the end have an impact on covered bonds via the relative value channel.
To sum up, we see wider levels in early 2025 as an opportunity rather than the start of a gap wider. Bank liquidity is still strong, the rotation into fixed income is intact and for covered bonds, it is above all about the relative pricing to competing asset classes. The current valuation of covered bonds is at post-QE wides, together with other factors such as strong relative value versus senior preferred, significantly improved valuation of SSAs (i.e., widening pressure may be limited in SSAs, and consequently covereds), and little chance of 2025 being as supply heavy as the transformational years of 2022/23 (which saw TLTRO repayments, rates volatility, higher redemptions etc.).
French covered bonds vs. the OAT-Bund spread gap
The first thought is that there is a considerable difference between French covered bonds and the OATs depending on area the curve. At or below the 5Y tenor, French covered bonds still offer some pickup, around 5bp, vs. the sovereign, as the latter trades at MS+27bp in the 5 Y. On the other hand, at the long end, French covered bonds trade clearly tighter vs. OATs (-15/-20bp in the 10Y area and c-35bp in the 37s), as in the absence of supply French covered bonds have continued to outperform the country’s sovereign.
However, French long supply will be back in 1Q next year, and this will lead to more curve steepening, as even in the current context, a new 10Y French covered bond would have to price around if not above ASW+60bp in our view. It is worth bearing in mind the levels where we are right now in two critical references: the EU 10Y trades around MS+52bp pressured by France’s long-term budget concerns, while the 10Y OAT has reached 62bp over the swaps level.
Could long-tenor French covered bonds price tighter than the sovereign? For this to happen, we would need to see French covered bonds reaching a level on ASW basis wide enough for international accounts to step in and offset the drop-in activity from French buyers.
In our view, and in a bearish scenario, French covered bonds at ASW+70bp would attract enough traction from the traditional credit investor base that should cover the gap left by domestic investors (mostly bank treasuries), as at these levels covered bonds are almost a no-brainer for credit investors due to the relative value vs. senior preferred debt. In the event of severe stress at the sovereign level, covered bonds – particularly those backed by home loans rather than public sector collateral – could trade inside their sovereign.
As such, in our view, for French covered bonds to trade tighter than the sovereign we would need to see a scenario where the 10Y OAT-Bund spread widens beyond the 90bp mark, which would imply the 10Y OAT at or above 80bp on an ASW basis.
We are going to talk a lot about SSAs in 2025
SSA markets have come under renewed pressure with swap spreads collapsing even more. It is hard to put a floor underneath swap spreads at present. Similarly, covered bonds is also a tough environment for any asset class priced against swaps. SSA spread levels against swaps have now reached new highs, with the widening mostly visible in the EU curve to date.
Needless to say, the EU continues to be the main focus, given the heavy issuance volumes and long duration, which have had a big impact on the broader SSA market.
Forecasting EU issuance volumes has never been an easy task. However, with the EU trying to smooth issuance volumes over time by decoupling funding operations from disbursements, this has in a way become even trickier, although also slightly easier (i.e., with an annual target of EUR140-150bn irrespective of actual disbursements).
We expect the EU to communicate its 1H25 funding plan in mid-December. The back-loading of disbursements from the Recovery and Resilience Facility (RRF) over 2025-26 combined with more payments to Ukraine as well as the first NGEU redemptions could lead to EU disbursements (including refinancing needs) almost doubling over the next two years, with the biggest jump to come from RRF loan disbursements and to a lesser extent from RRF grant payments. While the EU will be able to rely on substantial cash reserves and net bills issuance to face this jump, our figures still suggest that EU-bond gross issuance would have to be c.EUR150-165bn per year over the next two years, with around EUR80bn raised in 1H25, the highest six-month volume and slightly less in 2H25, which is still quite a jump from the EUR140bn raised this year, if we assume full disbursement of NGEU funds by YE2026.
As such, gross supply will continue to be a drag on EU valuations in the coming months, but there is a case for stress levels to fall beyond this. In addition, the start of SURE and NGEU redemptions next year will also mean that the net supply will begin to stabilise at EUR130bn in 2025 with the number then beginning to drop from 2026.