Covered bonds reach for a new standard
Buyers, sellers and arrangers gathered together in March, in the wake of the European Commission proposing its Covered Bond Directive, for the GlobalCapital/Crédit Agricole CIB covered bond roundtable to debate how the market would fare as the European Central Bank withdraws its quantitative easing programme.
Though asset purchases have already been halved, the influence on covered bonds will take time to be felt, as net purchases potentially extend into 2019 and the reinvestment of redemptions will follow for many years thereafter. Higher spreads should encourage an improvement in structural demand from regulated bank investors and make up for the partial loss of demand from the Eurosystem.
Though interest rates will be anchored down at the short end, the improving economic outlook is likely to take its toll at the long end which will potentially cause a loss in value of covered bonds relative to the sovereign, supranational and agency sector. Conversely, as the crushing impact of cheap central bank purchases slowly diminishes, the credit curve will steepen and spreads across the capital structure will widen, notably in the senior non-preferred asset class.
As issuers strive to meet regulatory ratios and seek the smoothest route to replacing liquidity borrowed under the ECB’s refinancing operations, they will have to become used to competing for investors’ attention again.
But fortified with the European Commission’s newly proposed directive, which should enhance the market’s quality even further, covered bonds will continue to provide a reliable and cost-effective funding option. The directive offers a high level path towards greater excellence, raising minimum standards across the board and will be implemented in an orderly and non-disruptive manner.
Though timing is still a long way off, the European Secured Note could yet come of age and, with the right regulatory treatment and a more normalised credit curve, it is a market that has the potential to help fuel European growth, forming a lynchpin of the Capital Markets Union.
But the regulatory backdrop is not entirely benign: specialist mortgage lenders are anxiously looking on at the tussle between the European Commission and Parliament over the Net Stable Funding Ratio in the hope that their business model is not dislocated.
Green bonds are here to stay, but whether green covered bonds can become anything more than a niche product remains to be seen. The lack of standardisation and incentives, for issuers and investors, is holding the market back.
Last but not least, global housing market hotspots, which have been inflamed by monetary policy, are not yet posing a big problem for covered bond investors. This is not just because of the protection afforded by low loan-to-value ratios and high collateral cushions, but also because the problem has so far been largely confined to the best rated banks in the best rated countries.
Participants in the roundtable were:
Andreas Kraft, executive director, credit trading,
Crédit Agricole CIB
Daniel Herdt, portfolio manager, Lazard Asset
Florian Eichert, head of covered bond and SSA research, Crédit Agricole CIB
Götz Michl, head of funding and debt investor relations, Deutsche Pfandbriefbank
Henrik Stille, portfolio manager, Nordea Investment Management
Kristion Mierau, head of European covered bond portfolio management, Pimco
Michael Sittrop, treasury associate, long-term funding and capital issuance, ABN Amro
Sami Gotrane, financial markets manager,
Société de Financement Local
Vincent Hoarau, head of FIG syndicate,
Crédit Agricole CIB
Bill Thornhill, moderator, GlobalCapital
End of CBPP3
Henrik Stille, Nordea Investment Management: I don’t think that net purchases will end in September, they’ll continue for at least another three months and maybe six, but at a reduced rate.
Andreas Kraft, Crédit Agricole CIB: I can’t see the programme going from €30bn to zero completely. The ECB has been very supportive for the markets and they are well aware of the impact they’ve had on spreads. With that in mind, I would expect a more gradual exit from CBPP3 with a reduction down to €15bn a month going into year-end and a further reduction in 2019. Coupled with the commitment to reinvest redemptions, spreads should still be supported, offering only limited widening potential for 2018.
Daniel Herdt, Lazard Asset Management: We learned one thing from the ECB over the last couple of years: it was that they try to be as supportive as possible, and when it comes to tapering they want the end of QE to be as gradual as possible. So I imagine they’ll try to taper until year end. They could go from €30bn per month to €20bn, and from €20bn to €10bn, or something like that.
Götz Michl, Deutsche Pfandbriefbank: I think the ECB will reduce buying from September but will still give some kind of a support. If you have a smaller portion of primary issuance placed into CBPP3 then there is a better chance the ECB will buy in the secondary market. And if we see order books get a little bit smaller then maybe we have a slight repricing on the covered bonds.
Reinvestment of redemptions will become a greater force driving the market dynamics. But for me the big question is around the envisaged balance between primary and secondary covered bond purchases. I would expect the ECB to reduce its tacit support post-bookbuilding of covered bond new issues.
Kristion Mierau, Pimco: They’ve already committed to reinvestments so redemptions in 2019 coming in around €2bn a month will have a material impact, particularly if net supply is less than that — which will be difficult to forecast given uncertainties around funding strategies and potential early refinancing of upcoming TLTRO-II redemptions.
Florian Eichert, Crédit Agricole CIB: This year it’s about €18bn and, though next year is not yet public, we estimate annual redemptions at about €22bn-€23bn, rising to €26bn-€27bn in 2020. That is certainly less than the gross purchases we have now and it will be more seasonal, though the ECB won’t try to match purchases with redemptions each month.
But based on these numbers you won’t see the Eurosystem putting in an order for half an issue size from 2018. They might come in with 20%-30% which will not be anywhere near enough to support current spread levels. So you will certainly have a primary driven repricing, with issuers being a bit more cautious, and that might eventually bring up secondary spread levels.
If you were to then have a more volatile backdrop causing issuers to stop primary activity altogether, purchases of €2bn a month would, however, certainly be enough to push spreads tighter again. The ECB already holds about 40% of the market, liquidity is really poor and for the market to normalise again it would take many years, as the existing stock would need to mature and be replaced with publicly placed issuance.
Herdt, Lazard: The reinvestment of redemptions will have a big impact for the euro covered bond market, at least for the next two to three years, and should support spreads.
Stille, Nordea: Gross QE purchases will start to become more important than net purchases given that the reinvestment of redemptions is really going to pick up. In April I believe PSPP redemptions will jump to around €25bn which means gross purchases will also probably move higher and become a very significant supporting factor for all the purchase programmes, including covered bonds.
Hoarau, CA-CIB: Redemptions will be supportive per se and the amounts are significant, but it is very difficult to read how this will impact different jurisdictions. The TLTRO refinancing will add a substantial element of complication when it comes to an assessment of how CBPP3 redemptions will affect spreads. CBPP3 redemptions may help offset pressure on spreads in jurisdictions and names where covered bonds are likely to be used for TLTRO refinancing purposes.
But collateral is not unlimited, so I expect senior spreads to be impacted the most — particularly in southern Europe where TLTRO take-up has been most important. In that respect, covered versus senior spreads will be scrutinised over the coming months and this will drive issuance patterns, and subsequently spread pressure on one or the other asset class.
So there’s nothing to stop the ECB continuing to buy covered bonds at the same pace as it does now in the event primary issuance maintains momentum into the second quarter and beyond. After all, some banks will be looking to refinance liquidity borrowed under the TLTRO into covered bonds and, if this occurs, net covered bond supply could increase quite sharply. Faced with this, the ECB may choose to soften the supply surge by reallocating out of the public sector into the private sector.
Stille, Nordea IM: Exactly, I think it might very well end up like that.
Impact on spreads of peripheral revival
Florian Eichert, Crédit Agricole CIB: Peripheral spreads essentially raced ahead because of QE and I’m starting to see a little more interest in, for example, Spanish covered bonds on the back of better growth, which helps relative value. So to some extent growth is helping offset some of the required widening that will happen due to less QE buying. But, is it going to fully offset that? No way. But the better fundamental economic outlook is certainly going to be a support.
Mierau, Pimco: Some significant things have changed for periphery covereds. For one, fundamentals are stronger with banking sectors having been, or are being, detoxed, and balance sheets are being shored up with more and better quality capital. And covered bonds sit on top of that with bail-in exemption. And, if we’re talking about idiosyncratic risks, there are Spanish and Italian covered bonds with arguably stronger credit metrics than some German counterparts.
Ratings agencies have acknowledged this with revised covered bond ratings methodologies and the subsequent migration upwards has played into the ratings-based regulatory treatment of covered bonds for institutional investors, which has broadened the investor base of natural buyers. Evidence that peripheral covered bonds still offer enough spread pick-up to keep regulatory driven investors involved can be seen in their strong participation in recent primary deals.
Kraft, CA-CIB: Eurosystem buying and a lack of supply have provided the main drivers for the peripheral spread tightening to current levels. Coupled with the low interest rate environment, investors have ignored the negative impact on liquidity and relative value in their hunt for yield.
Assuming higher yields and more volatility going forwards, I would expect a rather moderate widening in peripheral covered bonds versus their underlying government bond markets, driven by higher supply, leading to a return of relative value opportunities and spread normalisation. At the same time, I expect to see the correlation between sovereign and covered bonds increase, as in my opinion, this has pretty much been ignored lately.
Herdt, Lazard: At least in the case of Italy, OBGs are trading far through their own govvies. It doesn’t necessarily make that much sense to buy Italian covered bonds when you get more spread and liquidity in the BTP, it’s not sustainable.
Hoarau, CA-CIB: In the long run, compression and convergence will dominate. But in my opinion, credit spread differentials are still too tight in some specific situations. The risk of default has disappeared, volatility is edging lower and it seems likely that growth acceleration and rating migration will support the trend.
Nevertheless, spreads in higher beta covered bonds may widen before they converge again post-QE. Spread decompression, or taper tantrum, is likely when the ECB steps out of the game. But, at present, the Eurosystem has delivered an unlimited backstop bid in the secondary market.
Given that the ECB has the ability to buy up to 70% of a given outstanding bond it is probable the ECB will bid for bonds in the event there is a resurgence in volatility. But when the ECB finally disappears, traction from opportunistic buyers will evaporate until a new pricing paradigm is established.
Eichert, CA-CIB: Greece has not in reality been part of QE and, being sub-investment grade, it’s in a completely different ballpark from the investor perspective. Greek covered bonds have tightened quite a bit, but that’s a market that is controlled by four or five investors. These guys that have bought into Greece have done so because of the pick-up. But in general it’s hard to put Greece in the same basket as the other peripherals. Even if you do see upgrades you’re still probably looking at single-B or double-B rated covered bonds depending on the agency.
Stille, Nordea IM: Not necessarily, relative to the core countries at least. If you have a general widening, then the peripherals should also widen, of course, but I don’t see Spain, for example, as more at risk than the core countries. Many Spanish covered bonds are now trading with a pick-up to the sovereign, and I believe the sovereign would need to widen in order for Cédulas to widen.
Mierau, Pimco: This has potential to be a bigger problem for covered bonds than CBPP3 ending given the relative size of potential TLTRO refinancing. Even if banks only refinance a small proportion of their TLTRO liquidity into covered bonds via public markets, it could be material relative to net supply — which has been flat to just marginally positive in recent years. I wouldn’t be surprised to see some issuers begin to prefund this June when the window opens for early redemptions of 2019 TLTRO-II and before CBPP3 closes for business as planned in September.
Herdt, Lazard: I don’t think issuers want to come to the primary market around the same time as the TLTRO matures. So I would expect them to prefund some months before that as we just recently saw from Bank of Ireland, which also participated in the BoE’s Term Funding Scheme. So yes, I think we will see more supply coming due to the TLTRO this year and next, but not necessarily in the third quarter when the first tranche matures.
Sami Gotrane, SFIL: My feeling is that in the end only a small portion of TLTRO will be refinanced by covered bonds. Covered bonds face a lot of constraints for the largest banks as they need to keep their programmes fuelled with assets on a regular basis and they need to finance the over-collateralisation. Also, with funding on the senior preferred and non-preferred being rather cheap right now issuers will probably want to make better use of those markets. So, I don’t think covered bonds will be used extensively.
Michl, PBB: We took TLTRO liquidity firstly because it’s cheap and secondly because it is rather flexible, as you have repayment options. This liquidity was accounted against covered bond issuance in 2017, so we could refinance it with covered bonds. We either keep collateral in the public sector pool, or put it in the ECB account for the TLTRO.
Michl, PBB: We have an internal schedule of when we expect to repay depending on new business, balance sheet, and the composition of overall funding.
Michael Sittrop, ABN Amro: I doubt we will fully re-invest back in covered bonds, or at all. TLTRO is four year funding and we are primarily active on the very long part of the curve in covered bonds. If we were to replace TLTRO as early as this year then that would be for us the fourth quarter, but it would then make more sense to replace it with a similar type of tenor and that would for us is more likely be in senior unsecured format.
Gotrane, SFIL: And, in terms of limiting covered bond issuance, we also need to keep in mind the fact that a lot of banks are monitoring their encumbrance ratio.
Eichert, CA-CIB: A number of issuers a bit further down south have used retained covered bonds as collateral for the TLTRO, so as far as encumbrance goes, if they refinance that in the market it doesn’t increase their encumbrance. I think some of them will certainly do that at the first opportunity which is the end of June this year because at that point they could still expect the Bank of Spain or Bank of Italy to buy up to half their deals. In that sense the refinancing decision depends on the country and the balance sheet structure that they have.
Sittrop, ABN Amro: If you look at our needs and our growth path towards the MREL target then we’re not really in a hurry. Therefore, we are quite comfortable. In the Netherlands we are expecting to be able to issue in the second half of this year. At that point we’ll take a closer look as to whether we will issue at the earliest possible moment, or later on — because we don’t have that such a need.
Eichert, CA-CIB: In the current environment, where some of the stronger banks have their senior preferred trading only 15bp-20bp back of covered bonds, senior might actually be the cheaper way of raising liquidity to replace TLTRO out to the five or seven year part of that curve.
Sittrop, ABN AMRO: Senior to covered spreads are relatively tight but, given the long duration of our mortgage portfolio, we have a relatively large long term funding need. And from that perspective covered bonds are our preferred instrument to use. In that sense, for us there is a greater play-off between tight senior preferred versus non-preferred spreads, but not, per se, from covered bonds towards preferred or non-preferred.
Herdt, Lazard: I expect some banks will have more to do in the senior non-preferred space than in the covered bond space, even though covered bonds are a cheaper funding tool.
Stille, Nordea IM: A part of the TLTRO will probably be refinanced with senior non-preferred because banks still need to comply with MREL. A large TLTRO tranche matures in 2020 and, though issuers are allowed to pay back from this year, I don’t see any strong reason for them to do that just yet. But I do expect more covered bond supply than we expected a few months ago as issuers take advantage of good funding conditions.
Spreads to senior preferred
As for senior non-preferred, roughly 80% of our RWAs are at MREL capacity. I don’t expect there’ll be any MREL funding need for us in the near term. But of course, we will need to assess our additional MREL requirement in future years. So, if the product is ready to use, we are likely to issue.
Eichert, CA-CIB: My working assumption is that we’ll have a pretty active second quarter as banks refinance some of their TLTRO and make use of CBPP3 while it’s still there. If you get a bit of widening in the covered space on the back of this, you could have a temporary period where the covered bonds widen by 5bp-7bp while senior unsecured stays where it is thus compressing the product pair even more.
A few investors, such as UK asset managers, are monitoring that relationship a bit more closely again and, although they believe covereds are too expensive, they might actually start looking back into the market again if there’s 5bp-7bp tightening to senior.
I realise ABN uses senior for shorter maturities and covereds for longer, but there are also a number of banks where the asset side is nowhere near as long as it is in the Netherlands, and where funding teams arbitrage much more between covereds and senior. So, if we see another 5bp-7bp tightening in the senior/covered spread then I would expect more senior preferred supply, which should normalise the relationship. SEB, for example, has a senior/covered differential of 20bp and I’m pretty sure it issued senior recently because it was cheaper versus covereds after including hedging costs, which add 7bp-8bp.
Stille, Nordea IM: Our view is that covered bonds look attractive at current levels versus both senior preferred and non-preferred. So from an issuer’s point of view, it seems logical to issue senior versus covereds, which is why I also think that it’s very difficult to see covered bond spreads widen without senior widening more.
Mierau, Pimco: We have the flexibility to invest anywhere in the capital structure but covereds pretty much have to trade inside any unsecured instrument or you have an arbitrage due to the priority claim on collateral and additional layer of protection provided by statutory covered bond legal frameworks.
Herdt, Lazard: There is some linkage between spreads in senior non-preferred, seniors and the covereds. I do expect covered bonds to widen, but then there should also be a spread buffer to senior unsecured given the fact covered bonds are not bail-inable.
Stille, Nordea IM: A few weeks ago we had this stock market correction and iTraxx followed with a 10bp-12bp widening. So if we get more uncertainty regarding interest rates that could cause a widening. And then of course there is always the possibility of more supply particularly on the senior side because there issuers need to fulfil their regulatory ratios. This could push spreads wider, especially in conjunction with less buying from the ECB.
Bund-swap spread impact
Eichert, CA-CIB: Macro-wise to be honest, as far as covereds are concerned, you can pretty much throw whatever you want at that market and it isn’t going to move. If you look at Italy as one example, we recently saw some pretty volatile moves in the BTP but OBGs did not budge. Our rates analysts forecast Bund yields at 90bp at the end of the year which is reasonably close to where we are now. We’ve had periods in the past when Bunds went from virtually zero to 100bp in a few weeks and the covered/swap spread didn’t move.
Eichert, CA-CIB: The only thing that could lead to some pressure, apart from a supply related widening, would be if the Bund/swap collapses. Since early 2015 the 10 year Bund/swap spread has traded a 30bp-55bp range and we’re currently in the bottom third percentile. The spread fell to the mid-30bp area a few weeks ago and at that point we really did see a lot of investors moving from covereds into the SSA space. But now we’ve moved back 10bp wider that pressure has gone away. But if the spread were to tighten back again then you could see pressure on the tightest core covered bonds and I don’t think issuers would be able to defend levels that are currently up to 20bp through mid-swaps.
Stille, Nordea IM: Even though swap spreads have tightened, making covered bonds look more expensive versus Bunds, we still haven’t seen a covered bond widening, but of course there is a limit. As long as the Bund/swap spread stays in its established range then not much will happen. But if it moves below 30bp then covered bonds could sell off.
We recently saw this happening when Bund yields rose at the beginning of this year and the Bund/swap compressed by nearly 10bp. But this was reversed very quickly and did not have much impact on covered bonds against swaps. But if we return to the former Bund/swap levels in the 30bp area for a longer period of time, I expect we’ll see selling pressure increase and covered bond spreads should widen to swaps. This would most likely be caused by another generally bearish rising yield trend.
Herdt, Lazard: At the current Bund/swap level there is little tightening opportunity left in covereds. It doesn’t necessarily make sense to buy covered bonds in a euro aggregate mandate, especially from a risk-weighted point of view. Until the summer, spreads could stay where they currently are, but with the TLTRO maturing, supply perhaps intensifying and lower central bank purchases from September I expect covered bond spreads to widen against swaps.
Mierau, Pimco: To be less arbitrary, we can look across the credit and liquidity spectrum for more high quality liquid alternatives that trade versus governments instead of swaps like some explicitly guaranteed agencies. There are covered bonds trading only a few basis points to names like KfW for example, thus I would expect further tightening of this subset of covered bonds — whether driven by swap or credit spreads — will likely be met with resistance.
to bring bank investors
Stille, Nordea IM: It’s difficult to answer in isolation because I think it depends a lot on the other factors that we’ve discussed, in terms of how senior bonds and swaps perform. You probably need to have a general widening across asset classes to bring back private investors, but I don’t think 5bp will be enough. I think the investors that left probably compared covered bonds to something other than senior bonds, such as additional tier one. And to get those investors back it will probably take more than just a few basis points of widening. To get back to levels that attract these real money buyers, we may need to see something like a 25bp-30bp widening.
Herdt, Lazard: If you look at the recent years, like 2015 and 2016, where we were in the heat of QE, we had some volatility and spread widening. That provided a good reference point of where the spreads could go, so potentially a 10bp-20bp widening in the core. In the periphery, it should be some more because those markets experienced a bigger tightening. Italian OBGs are trading 50bp-60bp through BTPs, which is too much. I wouldn’t necessarily expect them to trade over, but I would say a 20bp-30bp widening in periphery is possible.
Eichert, CA-CIB: Different investors look at different things. German insurers are already back as 15-20 year yields are close to their internal targets. Dutch and French covered bonds are now positive to govvies at the long end, so that’s also something that is likely to interest French insurers. And because corporates and other higher risk, higher capital charge instruments have compressed, these insurance investors are locating back into covereds, SSA and govvies, because some of these corporate bonds are not compensating enough for the additional capital they need to hold.
Sittrop, ABN AMRO: Yes, given the fact that we have only issued 15 years plus for the last 2.5 years. A few basis points of spread widening or tightening is less relevant in these tenors.
The additional demand from bank treasuries coming in at these slightly wider levels would, in my view, go a long way to replacing the slowdown in the CBPP3. And it is always important to stress that for some years to come we are talking about replacing some, clearly not all, CBPP3 demand.
On the asset management side we’ve lost quite a few investors and for them to come back, 7bp is not going to do it. Otherwise you would need to see a lot more compression in other sectors which would cause them to move into a proper, conservative [covered bond] asset allocation and wait for other markets to widen.
Kraft, CA-CIB: There are many motivations to invest in certain asset classes for the different groups of investors. If we will see a widening of 7bp-10bp against swaps, we will arrive at positive levels against three month Euribor for a number of core covered bonds and this will trigger the first wall of demand.
Mierau, Pimco: I don’t think we need much widening to replace CBPP3 demand. We’re not in the same environment compared to when the first programmes were put in place. And let’s not forget CBPP2 stopped before the scheduled end, with no discernible impact. Although I do think it would be a healthy development if spreads were to slightly decompress and attract more value-driven investors.
Maybe more important than simply the technical impact of CBPP3 demand vanishing would be the lower propensity for secondary market makers to warehouse risk. I think they’ve become accustomed to being able to flip paper to the CBPP3, so when the programme ends their storage costs will rise. Scarcity will decrease and float of paper will increase, which means end-accounts will be less price-takers when looking to buy and market makers will be more price-takers when looking to sell.
This could translate into liquidity providing opportunities in the secondary market for investors with large balance sheets. It should also bring higher new issue premiums in the primary market. So CBPP3 ending isn’t necessarily a bad thing for investors.
Michl, PBB: The last tap we did was at swaps minus 18bp, so it was quite favourable. If we see a little bit of widening, this is not likely to worry us too much. In the senior non-preferred and senior preferred markets, of course, we could also see spreads widen by more. But on the other hand, with the better ratings for the new senior preferred product, and the assumption that we don’t really need senior non-preferred over the next few years, we are also quite confident that we would be able to issue at reasonable levels. In any case, spreads should be below senior non-preferred issues.
Gotrane, SFIL: We think that the potential repricing for core covered bonds with a long tenor is, say, between 10bp-20bp. We believe that would be enough to make the relative value of, let’s say a 10 year French covered bond, more aligned with a non-CBPP3 eligible covered bond. The reinvestment of CBPP3 redemptions will continue to support spreads after September.
Eichert, CA-CIB: The Commission tried to be very, very principles based, high level and to not kill off any national market or product. So they’ve tried to balance whatever they’ve put in there against what the various member states already have to avoid a situation where they’re shutting down the bond product because it’s so far from being aligned. They actually want to encourage other countries to join, and support the covered market.
Having said that, you do have some countries that will need bigger adjustments than others. France and the Netherlands, for example, are among those that have the least to do. But, for pretty much every country, there is at least something that needs to be changed.
On the other hand, Spain will have to change quite a few items in their law. They started working on an adjustment years ago but it got constantly postponed, and in the end they probably decided to wait for the directive to come out. They’ll have to change quite a few things, like introducing liquid asset buffers, putting in place independent cover pool administrators as well as considering a different approach to special public supervision.
Despite that I don’t believe it’s going to be disruptive because all the existing outstanding bonds can be grandfathered and because there’ll be a phase-in period to give people time to adjust their frameworks.
Gotrane, SFIL: I think that the Commission wants to make sure they really do not disrupt the market, that’s very important. It’s why I have no real major concerns.
Michl, PBB: Harmonisation comes at the risk of limited room for diversification. If we have some external impact, there is a risk that every market is affected. I would expect the directive to focus on raising minimum standards, which would be a good thing.
Sittrop, ABN Amro: I think Dutch issuers in general are relatively comfortable about the directive. I appreciate that there’s now some clarity around the extendable maturity structure so from that perspective I think the directive is going to be helpful.
Eichert, CA-CIB: The French were indeed quite vocal to avoid having an actual issuer default as the trigger for a maturity extension. The final text now says maturity extension triggers should be based on contract or statute law and the maturity should not be triggered at the discretion of the issuing institution.
They are basically saying there has to be an objective set of criteria that need to be met. You can’t just have some funding guy get up in the morning, look out the window and say ‘It’s raining today, I’m feeling bad, let’s extend!’ You have to have a specific set of circumstances that need to be met.
But on top of that, information needs to be provided to investors that shows clearly what the actual triggers and mechanisms are that need to be met, as well as the parties involved. The European Parliament, for example, proposed having supervisory authorities involved in the extension trigger, something the directive allows for.
Even though the issuer default is one of the triggers in the Netherlands, as well as some other countries, the directive only says there needs to be an objective set of criteria that must be met and transparently communicated to investors. This is not going to be a disruptive item, though it is likely some language might need to be changed in the prospectuses.
Sittrop, ABN Amro: I’m not so sure this is something ABN would use. I mean I get the concept. But from what I understand an ESN is yet another asset class and is going to get a completely different regulatory treatment. I doubt whether that will work.
Herdt, Lazard: ESNs could potentially be quite interesting but I guess much will depend on their regulatory treatment. This is definitely a space where you could get some pick-up, and they could still be LCR eligible, for example. That’s something we wouldn’t buy nowadays in a pure covered bond mandate, nor in an LCR mandate. But that could potentially change depending how they are treated in the directive.
Eichert, CA-CIB: As part of the Capital Markets Union, supporting the funding of the European economy is a central part. To help the Commission to reach that goal, the European Covered Bond Council proposed the ESN as one solution. At the moment, the EBA is working on ideas and proposals around an ESN product and an independent consultant is working on the impact assessment, but these are preliminary works. The Commission at some point thought they could come up with an ESN framework together with the covered bond directive, but then decided against that to concentrate purely on the covered bond directive.
So, timing wise, you’re now looking at possible implementation around 2020-2021 if the Commission — that gets into office at that point, and the European Parliament, which is also up for re-election next year — still sees a need in having an ESN asset class. So on the regulatory side it’s been pushed back a few years, which means there’s not going to be a regulatory framework in place that issuers can use or that investors could benefit from for a while.
And from the issuer side, as was just mentioned by Sami, I don’t see the need for banks to use something like this. Maybe, as you said, in a few years’ time if credit curves are steeper, if there’s a big-time spread differential between senior preferred and covered, then maybe it has a use. At that point it might make economic sense to add an extra layer in between senior and covereds.
But at the moment, where in some cases you have a 20bp difference between senior preferred and covereds, there’s no advantage to issuing an ESN which is going to price somewhere between those two. Even if ESNs price closer to the covered bond there’s still only a very marginal spread differential between the two and running an ESN programme is going to cost money as well. So from an issuer’s perspective it’s better to just raise senior unsecured money, at this point and I don’t really see a big market for the ESN.
Eichert, CA-CIB: Maybe with renewable energy there’s an angle than can come into play. But in theory that could form a part of the directive, because the directive doesn’t yet go into a lot of detail on the asset side. It does say CRR Article 129 compliant assets are automatically compliant with the Covered Bond Directive. But they also say high quality assets that need to have a market value or mortgage lending value can be eligible as well, without going into further detail.
If some member states, and I’m thinking Luxembourg for example, deem it necessary to consider that as a way forward, it could maybe be possible that green infrastructure assets fall under the covered bond umbrella.
Eichert, CA-CIB: There would then be a question of how well you can you value renewable energy projects, from wind parks to solar parks, or similar types of renewable infrastructure assets? I’m not an expert, but if the Luxembourg authorities think there’s a valuation scheme that is robust enough, they could implement the directive in their own country with these green infrastructure assets being a part of it.
Eichert, CA-CIB: I don’t see the need from the issuer side, and Sami mirrored that as well.
Eichert, CA-CIB: No, I do not think so. The Commission will merely change the references in other regulatory texts from Ucits Article 52.4 to the Covered Bond Directive. But that’s a very technical matter that doesn’t yet touch on the actual substance of the LCR. Whereas, at the moment, the LCR says bonds should comply with Ucits 52.5 or CRR Article 129, it will in future say they must comply with the directive or CRR. The questions that you have about the eligibility of Luxembourg Lettres de Gage in the LCR are not going to be cleared up by that. They will, as far as I’m aware, look into the LCR delegated act in a bit more in depth once the directive is in place and as part of a second step.
So for the next one or two years it’s likely there will still be questions around Luxembourg and the interpretation on EBA, which I think is wrong but it is out there. And the same goes for some of the Danish markets. In Denmark at least the Danish FSA came out about a week after EBA saying it considers Realkreditobligationer eligible for the LCR which clears the issue up for Denmark. But the Luxembourg supervisor has not yet said anything.
Sittrop, ABN Amro: Not really, it doesn’t really affect our issuance plans. I don’t know how they will treat OC for encumbered assets, I think that’s one of the questions, along with the RSF factor. At the end of the day, if you look at the ratio itself I think the covered bonds have a relatively minor impact on the overall ratio.
Eichert, CA-CIB: The ECBC is trying to achieve a few things, one being lower required stable funding on encumbered mortgages. For a bank like ABN it’s not a crucial thing, I suppose, because it’s a big bank with different businesses and different funding programmes and products that generate NSFR surpluses. So if your covered bond business is producing an NSFR shortfall it’s not the end of the world.
But if you are a specialised mortgage lender having the encumbered assets in the covered pool all with an RSF of 100, while at the same time the covered bonds below one year get an available stable of below 100, it ultimately means you run an NSFR shortfall unless you start building up a big liquidity book outside the cover pool which is funded via unsecured funding, thus not encumbered with lower RSF factors and which therefore produces an NSFR surplus.
On the mortgage side I think the Commission could be happy to support an 85% RSF factor, but Parliament doesn’t really seem to want to move away from the Basel rules. So it’s now down to the European Council and thus member states to propose something that can then be adopted by the Commission and hopefully they can convince Parliament to play ball.
On the OC, it makes no sense to consider all OC encumbered if an issuer has 60% OC in their covered pool. Not all of that should be getting a 100% RSF. The problem is to find a reference to use instead. After all, the NSFR is a going concern metric and issuers are very unlikely to bring OC down to the bare legal minimum just like that. And linking or referencing to rating agency requirements is something European regulation should avoid to do if possible. So in my view, a lower RSF on encumbered mortgages is clearly possible — on the other items it looks slightly less rosy.
ECB on CPT
Stille, Nordea IM: I don’t think so. But in general, we do think pass-through covered bonds are trading too tight versus hard bullets and soft bullets. I don’t think that’s because of the collateral rules, it has more to do with the fact that the QE programme has compressed spreads.
Herdt, Lazard: We are not against CPT structures per se, but the point is that from the beginning investors have worn the risk of maturity extension largely because of the effect of QE compressing credit spreads. But now the ECB has given a clear signal that there are definitely some additional risks around CPT structures, which have to be analysed more intensively compared to hard and soft bullets and which investors need to be compensated for. Perhaps this will help markets to re-evaluate CPT covered spreads.
Green covered bonds
Stille, Nordea IM: It seems like this is a very big trend at the moment, but actually I believe the green agenda fits much better on the equity side where you get a bigger reward for buying a green product. We buy green covered bonds, but we don’t have any dedicated green fund, we just buy them for our normal fund.
Herdt, Lazard: It’s difficult for an investor. If you don’t have dedicated green bond mandates, it doesn’t necessarily make sense currently to buy green bonds because usually they are more expensive.
Stille, Nordea IM: Yes there is demand particularly from our teams in Sweden, but we have not seen much demand from our own clients. It’s clearly a growing trend, though I believe at this stage the investment decision seems to be driven more by politics than a return on investment.
And, to a large degree, borrowers issue green deals because it gives them a lot of good publicity. But I cannot tell my clients that we are buying green bonds for the portfolio because they will be safer, because that’s not the case. The security of a green covered bond is identical to a vanilla covered bond. That’s a fact. The cover pool is the same, the issuer is the same. If the issuer defaults investors in the green bonds will not be any better off than other investors.
Michl, PBB: We have green commercial real estate loans on the balance sheet that we’ve been earmarking for three years now. So in principal we would be able to rather quickly issue a green bond. But so far we have not used it partly because the capital market has been very favourable. But this is something we could do in the future depending on the economics.
Gotrane, SFIL: We have decided to potentially price a transaction this year, so yes we’re positive. The first deal is going to be most likely a social or healthcare bond backed by loans to public hospitals.
And the next step could be part of a global new offer of green loans to local authorities in France that we would like to launch with La Banque Postale next year. We’re financing mainly green and social projects because we are financing the investments of local authorities and we would like clearly to target new investors. So here we have two different steps which shows it’s a priority for us. The motivation for us is to diversify the investor base which is also the same motivation as the French government which has also started to issue green bonds.
Eichert, CA-CIB: At the moment there is none. You have standardisation within Germany, for example with Deutsche Hypo being essentially the same as Berlin Hyp. They are both secured on commercial mortgages with an external certification from Oekom research.
The moment you start talking about residential mortgages, there is no standardisation whatsoever and no commonly accepted approach either, which is why most issuers that do not have a viable commercial mortgage portfolio go for social or sustainable covered bonds backed by, for example, social housing assets like Kutxa, Caja Rural de Navarra and Münchener Hypothekenbank. They go down that route because it’s relatively easier to put together a programme.
You have discussions in Brussels at the moment on a green supporting factor for green mortgages. The moment you have something from the regulator, adding the green covered to that in a standardised form should be no big deal. But until then it’s always going to be one guy using one approach, another guy using another approach.
Spabol used the Norwegian building code as a short cut. They said that current building regulations are so strict that houses built after the code came into force by definition are green. The deal went well but I am sure there are investors out there who would not deem this approach to be sufficient. So there is no standardisation at the moment whatsoever. The easiest approach would be to have the European Commission set a definition of green mortgages and give a set of rules and then you can construct a covered bond that corresponds to that.
Herdt, Lazard: When I talk to other investors, there is still a huge discussion about whether green bonds are really green. As matters currently stand issuers use only one programme to issue both green bonds and regular bonds. A separate programme should be used for green issuance but that’s costly, so perhaps there is another point where the government could step in and try to give incentives. There should also be more standardisation by regulatory law which would help improve investor demand.
Stille, Nordea IM: The green component does not really have any impact on investment risk. Financial regulation is supposed to safeguard investors and the financial industry and to differentiate between different types of risks. It should not be used as a tool for political imperatives. Politicians have any number of ways to incentivise a reduction in carbon emissions and they don’t need to rely on lower risk weights to achieve those aims.
Sittrop, ABN Amro: You see that newly built houses are generally considered green. But I don’t think you can really say that the credit risk of a mortgage decreases in line with the year a building was constructed. So from that perspective I doubt more recently constructed greener residential properties deserve a lower capital charge.
Eichert, CA-CIB: We’re talking about arguments that are really well received in Brussels because it fits the overall political environment and the push that they have to bring down carbon emissions and be in line with COP21 targets. So you’re running into open doors already. But we’re not yet at the point where we have sufficient data to really prove that the probability of default is lower because the mortgage holder is paying less in terms of utility bills, and that the value of his house is substantially higher than it would otherwise be after taking into account the higher mortgage repayments.
Stille, Nordea IM: Better to do it like that. Issuers should then also benefit from financing properties that emit less CO2.
Eichert, CA-CIB: If you add a whole lot of extra taxes on some brown mortgages, budgets are going to be more strained and by definition brown mortgage risk is going to be higher relative to green. But I don’t think that governments want to put people at a disadvantage and increase the risk for owning houses that emit more CO2. At this stage it’s about incentivising and improving the situation for homeowners that invest to reduce CO2.
Stille, Nordea IM: Exactly. But also, if you really want to incentivise green investors, then you should have a separate cover pool for the green assets. An investor that really believes these loans are safer could genuinely get exposure to different collateral. But for now, all covered bonds are issued from the same pool, irrespective of whether they are green or brown.
Stille, Nordea IM: Yes, in this case the green pool is separated. This is a better way.
Stille, Nordea IM: Sweden, Canada, Norway and Australia have all experienced house price inflation. But these countries are actually some of the few AAA rated sovereigns that are left. Higher capital cushions and a higher proportion of bail-in-able debt also gives much more investor protection than 10 years ago. So if real estate prices were to correct, the spill-over risk into the real economy should be smaller than, for example, what we saw in Spain back in 2007/08.
Michl, PBB: We are more careful in the UK market and we have originated less new business volume in the UK in 2017. We continue to be active in the London real estate market and we expect the UK to remain a core market for PBB.
Mierau, Pimco: Sure. There are regions where housing markets are showing signs of overheating which is a reason to monitor them closely. But rapidly rising house prices in isolation doesn’t necessarily warrant concern. Actually, rising house prices is a good thing for covered bond investors as it means higher equity values on the loans, which are held as collateral assets in covered bond pools. We need to look more at the propensity to default and recovery rates for foreclosures, which are driven by several variables not just home price depreciation.
It seems, anytime we have a housing hotspot, there are parallels made to the US experience which simply isn’t comparing apples with apples.
For one, most countries with covered bond legislative frameworks have full recourse to borrowers which means even high LTVs following a bursting housing bubble are not necessarily a strong indicator of default probability.
If we take Sweden or Denmark, for example, where house prices have been on a steep ascent, from a covered bond investor standpoint, the collateral of most issuers’ covered bond programmes could absorb a housing correction in excess of 30% without incurring any loss severities due to their low LTVs and near zero NPLs. But before you worry about recoveries on your covered bond collateral, you first need the issuers to be insolvent. And in Denmark and Sweden we’re talking about AA rated issuers in an AAA country macro backdrop.
Stille, Nordea IM: We look to see how much OC is available and how easy it is for issuers to remove it. Even with OC that is not contractual and is easy to remove, it’s still better to have it there than not because it should provide a buffer if real estate prices start to fall.
Herdt, Lazard: There is some investor complacency, though I expect that to change over the next couple of years because, as spreads widen, it’s likely investors will want to take a closer look between different issuers and their programmes. My expectation is that a spread widening will put pressure on issuers to take care of their cover pools, especially nowadays as transparency is much greater in some legislations compared to a few years ago.
Mierau, Pimco: When it comes to OC, remember covered pools are dynamic so you have to assume that if an institution is getting into trouble, it’s going to do its best to survive, which may include monetising its balance sheet. In practice, we take a conservative approach and do not assume we will have claim to any excess assets above and beyond the contractual minimum. A jump to default scenario like Lehman, as opposed to a slow death of an institution, could actually be a positive for covered bond investors in terms of loss severities as it is more likely excess collateralisation in the covered bond programme would be crystallised and segregated post-insolvency.