Hybrids: Made in Germany
German banks had been notable by their absence from the AT1 market until May, when Deutsche Bank launched the largest such issue yet. The asset class is now primed for further growth, with subordinated debt also coming in more familiar guises. Here, leading German issuers and investors share their views on the market’s prospects.
The tax treatment of hybrid capital in Germany was under discussions for months and consequently this delayed the launch of the first German Additional Tier 1 (AT1) transaction. Why was this?
Florian Lechner, Linklaters: In order to have implemented something in a straightforward way it would have been beneficial to change the tax law, simply to ensure that the treatment of AT1 instruments is in line with what the market expects — which is also what the tax authorities have now confirmed. The reason why there were delays is that under the current law it is not entirely clear how to qualify these instruments and I can understand that there were some headaches for the tax authorities. They have now come to the conclusion that the desired result is in line with the existing law — but there it was not 100% black or white, it’s a grey area, and so I have some sympathy for the tax authorities. They had some issues and needed to analyse this thoroughly. The fault in my view is with the legislators — they should basically have passed specific rules for the treatment of AT1 instruments.
Norbert Dörr, Commerzbank: We were included in the discussions from the beginning — even more so when the market for AT1 really opened up and we were already seeing a lot of issuance activity outside Germany. We simply wanted to have the option for German banks to do AT1, too — it is a question of having a level playing field internationally. While outside Germany the issue is not resolved everywhere yet, I believe we are getting there.
If the tax law had needed to be changed to support the tax treatment that has now been agreed, that would have taken longer. So the approach from day one was to find appropriate arguments that the decision could be made within the current legal framework, and that is what ultimately happened.
The other difficulty at the very beginning was that the CRR wasn’t finalised and approved by the European Parliament, but was still in draft form. When you talk to the tax authorities you need to be clear about the legal form and terms and conditions of the instruments under discussion. The tax authorities nevertheless expect the regulator to confirm that the statements made with respect to potential tax treatment are consistent with the regulatory requirements from a CRR perspective for such an instrument.
So it was a big and necessary step following the finalisation of the CRR for the industry to reach the point where we are now, having agreed a sample term sheet for these instruments with the regulator confirming that it qualifies for AT1 according to CRR.
Deutsche Bank announced its AT1 plans at the end of April and issued only three weeks later. What was the process in between? How did the capital increase affect the process?
Jonathan Blake, Deutsche Bank: DB announced its intention to issue Eu5bn AT1 by end-2015 in October 2013, so we had been working on the transaction for many months prior to the announcement on 28 April. In the release, we announced our plans to conduct a roadshow to explain the transaction structure and credit story to investors. The roadshow concluded on 9 May, with very strong feedback. However, we decided to delay execution by one week to ensure that investors had full transparency regarding our capital plans before proceeding. When the news broke on 11-12 May, investors reconfirmed interest and we proceeded with execution over two days.
Jonathan Blake, Deutsche Bank
Why did Deutsche opt for sterling and US dollars on top of euros for this first appearance in AT1 format?
Blake, Deutsche: A euro tranche is a natural choice for DB. However, many investors are currency-agnostic and the markets in US dollars and, to a lesser extent, sterling are also deep. Given our Eu5bn AT1 target, this allowed us to take a very significant step to achieving this target with a Eu3.5bn issue while avoiding overloading any individual tranche and hence not jeopardising the secondary market performance.
How did you manage the size and price elements in this multi-tranche offering? Did you face a risk of cannibalisation across tranches?
Blake, Deutsche: This did pose an execution challenge as many investors had interest across a number of tranches. However, by staggering the relevant non-call periods and ensuring pricing consistency across currencies and tenors versus peer transactions, we were able to ensure that all three tranches saw similar levels of interest. The euro and US dollar tranches had total order volumes just in excess of 10bn in their respective currencies and the sterling tranche had roughly £6.5bn in orders. This left us in the comfortable position of having a number of options on how to size the three different tranches to achieve the best result.
What were the key factors behind the transaction’s success?
Blake, Deutsche: The AT1 market is still, despite a number of transactions so far, a nascent one and the requirements of CRR are such that all AT1 instruments feature considerable issuer discretion. As such, investors have a number of valid questions regarding structure, credit and management intent. We undertook a week-long roadshow, including a broad investor call to cover anyone we couldn’t physically meet. According to the feedback we received, this transparency was well received by the investor base. Combining this with the DB credit story, rarity value and good market timing resulted in the strong outcome we saw.
Were you happy to see Deutsche getting its deal done and the German AT1 market open after the delays with the tax issue?
Dörr, Commerzbank: Absolutely. To be honest, we had actually hoped that they would get it done earlier, because in the end the structure they did — meaning the 5.125% trigger and temporary write-down — would highly likely also be the target structure for us. In addition, we would have proof that the instrument has been cleared with regulators and tax authorities. From a structural perspective, the important element, really, is not necessarily the tax deductibility of the coupons; it’s the treatment of withholding tax, because the ultimate decision allows direct issuance of AT1, and that is very important.
So we had been awaiting it as a proof of concept. And, to be honest, it also serves as another data point for pricing purposes.
The Ministry of Finance confirmed the treatment of an AT1 instrument based on model terms proposed by the Association of German Banks — does that possibly restrict the potential development of different structures?
Lechner, Linklaters: It is limiting, at least with regards to timing.
If there had been a change in law, there would have been much greater certainty. But given that we don’t have specific rules, what the tax authorities have provided is probably as good as they could have. Obviously they can’t rule on all kinds of instruments because they need to analyse certain characteristics, and they have done so with the model brought by the banking association. The problem is that if you deviate from these model terms, the only way you can get certainty is to apply for a binding ruling, and there are two issues with that.
Firstly, it is time consuming, it normally takes six to eight weeks to obtain one. And given that these financial instruments are complex, it also depends on which tax office is responsible — if you go to Frankfurt they are probably more familiar with these types of instruments than if you go to a smaller tax office elsewhere in Germany.
And the other thing is that binding rulings can be quite expensive: they are subject to fees and can cost up to a little over Eu100,000.
Dörr, Commerzbank: Of course, if you were to have a legal statement saying that anything that is AT1-compliant according to CRR is tax-deductible, that would give much more certainty. But I believe that is not necessarily what was desired given the urgency of the matter. The tax authorities really wanted to embed it in the tax treatment of instruments according to their legal form independent of the actual issuer or purpose of the instrument.
And yes, if you deviate a little bit from the terms of the sample term sheet you have to convince your tax authority that you have not done so in a way that substantially changes the tax treatment — but I can’t anticipate any such cases. Where would you deviate from the model terms for an AT1? The only thing I can think of is the trigger level, and that is irrelevant for tax purposes.
Were there any significant challenges that LBBW had to overcome to ensure a successful debut in benchmark Tier 2 format?
Jörg Huber, LBBW: Actually there were no obstacles. Of course, we wanted to have the appropriate documentation in place once we decided to proceed. But as we had just updated our EMTN programme and all the new necessary language for sub debt was included, we just needed to look for the right timing.
Joerg Huber, LBBW
If you hadn’t seen such a spread tightening in the first quarter, would you have gone ahead with the benchmark Tier 2 project?
Huber, LBBW: As we have quite a comfortable CET1 ratio of 12.9% and a Total Capital Ratio of 18.5%, we were in no hurry to launch a Tier 2 transaction. We rather prefer to tap the market when conditions are very favourable.
Spreads had rallied quite dramatically in the previous months, and we saw that we had reached levels that are close to those we had before the financial crisis. It therefore seemed that there was not much room to go further. At some point it gets just too much — the difference between senior and subordinated paper won’t shrink to zero. Therefore, with this kind of demand out there and with the levels that are achievable in the market, we saw this as quite an attractive situation for us. Rather than wait for a time when we might need to issue and then just have to take the market conditions at that time, we decided to go ahead now.
What are the main takeaways of your respective inaugural transactions?
Blake, Deutsche: We took the time required to iron out all the regulatory and tax issues involved and also to prepare the investor base appropriately. This, together with a strong market backdrop, resulted in a successful transaction.
Huber, LBBW: As LBBW had not issued Tier 2 debt in public format for quite some time, we were aware that an investor update about our story was necessary. We were quite successful and the investors we spoke to were positively surprised. But as we did not tour through all of Europe, we see the need to visit our European investors more frequently.
Commerzbank tapped the 144A dollar market with a Tier 2 transaction in September — what was behind the decision to do that, the timing, and the choice of dollars?
Dörr, Commerzbank: Several factors played into our decision at that time. Firstly, we had done our equity capital increase a few months earlier, so it wasn’t that much effort to update our 144A programme. It was therefore a very lean process. Secondly, the US market is very deep and offered quite a spread advantage compared with euros or the dollar RegS market, as we saw at that time and as we still see — for example, with the Intesa Tier 2 trade just recently.
At the same time, we wanted to establish a new liquid instrument in the market, particularly in the context of the overall supportive environment for our Tier 2 levels. And it did pretty well, if you look at what happened to Tier 2 spreads after that.
Do you already have further plans for AT1 in 2014 or any intention to build up your Tier 2 curves further?
Blake, Deutsche: We plan to raise Eu5bn AT1 by the end of 2015 so have a further Eu1.5bn to raise over the next 18 months. You can expect to see us return to the market over this time period, but we haven’t decided exactly when or in which market at this point.
Huber, LBBW: We do not plan any other new issuance activities in the sub sector for the time being.
Bodo Winkler, Berlin Hyp: Berlin Hyp already covered its Tier 2 needs for 2014 in the first quarter. In total we issued Eu150m, exclusively via private placements and at favourable conditions.
Of course Berlin Hyp occupies itself with the subject of AT1, but there won’t be any new issue from our bank in the near future.
Bodo Winkler, Berlin Hyp
Why could AT1 be interesting for Berlin Hyp?
Winkler, Berlin Hyp: As a member of the savings banks’ institutional protection scheme, we see AT1 not so much as a security buffer for our unsecured investors. In its last full rating report Fitch described Berlin Hyp’s bail-in risk as low due to its ownership structure, as the mutual support scheme would come into force before unsecured creditors would be bailed in. Anyway, we feel that there are quite a few senior unsecured investors in the market that simply want to see a certain Total Capital Ratio and certain MREL (minimum requirements for own funds and eligible liabilities) even for a bank that is embedded in an institutional protection scheme like ours. Furthermore, AT1 would help us with respect to the leverage ratio.
What impression have you got about how much interest there is in doing AT1?
Lechner, Linklaters: There is much more activity in the market than a couple of months ago. That is something we see quite clearly. A couple of other players are now trying to launch their instruments.
Dörr, Commerzbank: I think it is fair to say that we are looking closely at AT1 as an option. We want to have that option available — ideally you would have a term sheet on the shelf.
The question for Commerzbank is, do we really need it at this time? We have done substantial deleveraging, with the run-down of our non-core assets portfolio, where we are ahead of the game-plan. Then it is just a question of, do we do it now while the market is in really good shape as it is now and pay the running costs, or do we do it at a later stage? This is the balance we need to strike.
The focus of Commerzbank is still on improving the CET1 fully phased-in ratio given the externally communicated target of reaching 10% by the end of 2016. However, if there is a need to do AT1 at some stage, we have the option of doing something.
Do you have a view on a management buffer to commit to on top of future CBR (combined buffer requirements) levels?
Dörr, Commerzbank: We want to be comfortably around the communicated target of 10% CET1 fully phased-in that I mentioned. The question for me is also whether down the road 10% is seen as the ultimate level by various stakeholders in the discussion — taking into account that several regulatory parts are still moving. To what extent you need an additional buffer should be seen in the then prevailing regulatory and market context.
I believe for a bank with a business model like Commerzbank and given the current information about regulatory requirements, 10% is a good number. Deutsche, for instance, as a G-SIFI with a stronger push in investment banking, as they recently announced, will obviously have to go higher in my opinion.
Do you foresee using all of the three main types of loss absorption mechanism for AT1 instruments in your ultimate capital structure?
Dörr, Commerzbank: That’s an interesting one. After Commerzbank acquired Dresdner we had a myriad of different legacy instruments, and we had to deal with all the issues such a situation might create in a crisis scenario following Lehman.
We are of the very strong opinion that the capital structure should be kept simple. That’s why I said earlier that direct issuance is important (and if you think about things like bail-in mechanisms, it’s more or less necessary). Would I deviate in loss absorption mechanisms? Why should I, unless someone tells me I have to, which I can’t think of a reason for. And could I anticipate having different instruments with different trigger levels? Not at that this time, because the simplicity argument is still for me the more important argument. It is also helpful for investors — they should see exactly where they are in the hierarchy.
Your CRD IV Leverage Ratio is already above 4% under phase-in. What is your target going forward and do you expect a shift from European authorities on the minimum requirement ahead of its Pillar 1 imposition?
Dörr, Commerzbank: It’s difficult to say, because there are two types of changes that might come.
Firstly, changes to the way relevant leverage exposures are calculated. I think that in regards to derivatives and securitised financing transactions, the latest moves are going in the right direction — it is not as conservative as before and legal netting is allowed.
Then there is the question, should we have a higher leverage ratio? I think what people have to consider is the composition and interdependence of the various regulatory requirements. We have discussion about CET1 ratios, but also concepts like MREL, or the latest GLAC (gone-concern loss absorbing capacity). They all have some similarities in concept to the leverage ratio. In addition, we have other initiatives like EMIR and Liikanen, which also address certain risk aspects that leverage has in focus. So all these things have to hang together, and have to also be seen in the context of all the supervisory mechanisms that there are. So I think people should implement the leverage ratio as it is now and not think about additional isolated adjustments increasing the requirements.
Given where we are with the leverage ratio above 4% under phase-in and the further run-down of our NCA portfolio, I also believe that going forward that it is a level where we want to be. In my opinion, the leverage ratio also should not be our constraining factor. Our focus will be on CET1 and if we fulfil our targets there we want our leverage ratio to be fine.
What has been your involvement in the AT1 space so far this year? Do you have dedicated CoCo funds?
Michael Liller, DeAW: We have been active in the AT1 space this year in dedicated funds that are able to invest in these new structures. These funds are either able to hold a high yield bucket or are dedicated hybrid funds that are structured to invest in these issues and able to hold high yield.
Michael Liller, DeAW
Stefan Sauerschell, Union Investment: Union Investment has been active in the CoCo market since 2011. In some of our funds we are allowed to invest in CoCo bonds alongside other subordinated bonds.
Our clients are interested in gaining exposure to this high yielding asset class but without having to build up individual positions themselves, so we plan to launch a new diversified CoCo fund for institutional investors in the coming months, subject to approval by the regulators.
The AT1 market is lacking a core investor base in Germany. Why is this and what could change the situation?
Rainer Gehler, DekaBank: We should see more participation proportionately in terms of general appetite for financial sub debt versus other jurisdictions. It has possibly been subdued thus far because of a lack of local name issues, which will be a logical starting point.
Liller, DeAW: The German fixed income market is mainly driven by insurance and pension funds, which are very rating sensitive.
Currently AT1 structures are mostly prohibited by investment guidelines and/or other restrictions. Also, the fact that most AT1 structures are high yield-rated limits the investor base, as does the fact that they are not included in most benchmarks.
Sauerschell, Union: The first CoCos from the UK were launched a few years ago and UK investors have therefore had a lot of experience with this asset class. However, we have not seen a similar development in Germany. Deutsche’s deal could therefore be very important as it was a high profile transaction in Germany.
More importantly, we expect a greater number of investors to consider this type of instrument because of the tight spreads and low yield environment in Europe.
Winkler, Berlin Hyp: The experiences of German investors with subordinated instruments during the crisis were not too positive. Some of them lost quite a lot of money. This experience seem to be one of the main reasons why our domestic investor base now acts cautiously concerning AT1 instruments. In addition, Germany was quite late when it came to clarifying the full terms of AT1 issuance, with Deutsche only coming in May with the first Basel III-compliant AT1. A domestic issuer is always a special incentive for investors to engage in a new sort of assets.
Furthermore, AT1 is not an easy-to-understand asset class. Each instrument is structured individually. What they have in common is the complexity of the structures, which implies a lot of analytical and credit work in advance of buying. And, as AT1 is no pure debt instrument, it is also a question concerning the mandate to buy these assets.
But we understand that large asset managers have now set up funds and begin to invest. The low overall interest and yield level should contribute to increased buying by German investors. Another supportive factor would be an increasing number of AT1 issues by German banks.
Vincent Hoarau, Crédit Agricole CIB: Germany lagged in putting in place the AT1 tax framework for the reasons covered earlier. The euro segment developed recently, while German investors on the whole are not the biggest fans of this new generation of loss-absorbing instruments. So it is not a surprise to see some resistance.
But with Deutsche Bank opening the German segment as an issuer, things are changing. Apparently, German investors bought nearly 20% of Deutsche’s euro tranche. This was by far the highest participation of German investors in an AT1 transaction. But this was not a surprise. With the help of some investor work, the domestic base should continue to grow unless an incident occurs and demonstrates they were right to be cautious. Elsewhere, the more German investors hear about buy-side accounts across Europe being in the process of launching CoCo dedicated funds, the more they will consider the asset class. I also think that the launch of AT1 dedicated indices will encourage everyone to get more heavily involved.
Amreetpal Summan, Crédit Agricole CIB: Initial supply was dollar-focused and had equity conversion, which ruled out large domestic buyers, along with the lack of domestic issuance. We expect issuance to pick up in Germany following the long-awaited approval for tax deductibility of AT1 securities and the recent issuance from Deutsche. We expect Landesbanks to take advantage of the favourable spread environment to refinance existing non-Basel III-compliant Tier 1 instruments, which we believe will be well received domestically.
Huber, LBBW: In our discussions with investors about our Tier 2 transaction we didn’t discuss Tier 1 transactions because this is not something that we are able to offer at the moment and also not for the foreseeable future. But out of other discussions I have had with investors I know that some are already investing, and others are looking more into it. Of course there are a lot of investors who are smaller investors and who would need to build up the know-how for these kinds of transactions. They used to invest in the old-style Tier 1 transactions and they got burned in the financial crisis, and now the new instruments look in a way much more dangerous than the old ones. So for these investors it will certainly take some time to adapt.
But then at some point in time they will reach a level where they say, well, on the one hand these instruments are not as secure as the old ones were, because when it comes to AT1 they have to be perpetual, but on the other hand they are issued from institutions that have to have a much higher capital base than they had in the past. So in a way you invest in a more risky instrument, but the borrowers have become much more secure, and I guess there’s a big pay-off for that. And therefore because all these investors — insurance companies, pension funds, etc — need to have some yield pick-up at some point in time they will get more and more comfortable with it.
Dörr, Commerzbank: Something you have to bear in mind is how certain investors who previously invested in those instruments have to treat these assets as part of their own regulatory perspective. That will make a difference for certain investor types. Also, these instruments with the fully discretionary coupon and these write-down mechanisms are slightly different to legacy hybrid Tier 1, so they might not fit into investment guidelines.
Nevertheless, the more issuance you see from a domestic market, the more investors from that domestic market you will see. And I am sure that you will see that in Germany, too, if a few more instruments are issued. When Deutsche did its deal you had quite a good portion of domestic investors, even if it was also highly anticipated by the international market, and we have seen this phenomenon in other countries if you look at domestic participation in French AT1s or UniCredit in Italy, for example.
And obviously the other important driver for investors right now is, where can you achieve a decent yield for your risk appetite in the current low interest rate environment?
How do you anticipate your activity for the rest of the year in the subordinated space, and how do you treat vanilla Tier 2 exposure compared with AT1?
Gehler, DekaBank: In the current low yield and spread environment, we anticipate our activity being the same as in the first half of this year.
We treat Tier 2 as Point Of Non-Viability instruments with minimal bail-in risk for strong banks versus AT1s, which can have coupon deferrals.
Liller, DeAW: We continue to be active in the subordinated space. New issues will be evaluated on the basis of our current investment process. If the issuer meets our investment standards and based on our relative value assessment we will invest in new subordinated structures.
Vanilla Tier 2 of European banks are more eligible for institutional fixed income funds, since they are mainly investment grade rated and do not have any equity conversion features. Given our book of business in the investment grade space, our demand will be higher than for AT1 structures. Dedicated funds will invest in the capital structure based on our view of risk/return and our relative value assessment.
What structural AT1 features do you prefer? Can you handle equity conversion as well as write-up/write-down?
Gehler, DekaBank: We prefer write-down/write-up. We also prefer language with static regulatory requirements with the option to convert or a liability management exercise on regulatory changes.
Summan, CACIB: From an investor’s perspective, the equity conversion provides upside optionality in the worst case scenario and therefore should be preferred over full write-down structures. However, until fund mandates are amended to allow them to buy securities with equity optionality, we expect banks to prefer issuing write-down structures. The technicals are better for write-down/write-up AT1s.
Liller, DeAW: Since AT1 structures with equity conversion are considered as a convertible bond, most institutional funds are not able to invest in these structures. In dedicated funds there are no restrictions on structural features. From an investor’s point of view, the write-up structures seem to be a good marketing instrument for issuers, since the write-up is on a discretionary basis on behalf of the issuer and linked to distributable income. Currently we view these structures as write-down structures.
Sauerschell, Union: We can handle equity conversion as well. It is very difficult for some bond funds, particularly in Germany, to handle equity conversion because it is prohibited in their prospectuses. We therefore prefer the write-down, write-up structure, because there is a broader investor base for the structure and hence more demand.
Stefan Sauerschell, Union Investment
Some observers have suggested that with equity conversion there is more of an alignment of the interests of equity and bondholders, and have discussed the hierarchy of instruments in the capital structure. Do you have any views on this?
Sauerschell, Union: Yes, with equity conversion there is a higher alignment. In a stress scenario this could be an important consideration for investors, but actually the market does not differentiate between equity conversion, permanent or temporary write-down structures. CoCos with permanent write-down loss absorption triggers are structurally subordinated to equity, and we prefer write-down/write-up structures where you may recoup some losses. This also better aligns shareholders’ and bondholders’ interests.
Hoarau, CACIB: The difference between equity conversion and temporary write-down is difficult to quantify. Some investors prefer equity structures, arguing that they are left with something in the case of a trigger event. Other can’t handle equity in their mandate. So, between the two we don’t see any impact in terms of valuation. The permanent write-down bail-in feature is used mostly in Tier 2-hosted CoCos and I am not sure that current valuations properly reflect the risks associated with this structure. I fully agree with Stefan. In a stress scenario such anomalies will be corrected.
How do you handle rating constraints? Can you consider unrated AT1?
Gehler, DekaBank: We can’t take up unrated issues, but we can invest in sub-investment grade debt.
Liller, DeAW: Currently we would not consider unrated AT1.
Based on our proprietary research, we are able to assign own ratings for these structures. The ratings of rating agencies are not a sole reliable source for assessing the risk attached to these structures, also when you take into account that some issues are structured in a certain way to receive a certain rating.
Sauerschell, Union: In principle we can invest in bonds that are unrated, and we then assign our own Union rating for the issuer or for these bonds. But when AT1s are unrated you have to question why they do not have a rating, and we believe it is because any rating would be below B-. We do not take new positions in bonds below B-, so we don’t buy unrated AT1s.
Is the investment grade/sub-investment grade threshold significant?
Sauerschell, Union: Yes, it is very important. Many funds have constraints regarding high yield and having an investment grade rating is a technically supportive factor. This is evident in the spreads of Tier 2 CoCos from Credit Suisse and UBS, which have investment grade ratings.
Spreads have compressed a lot since the beginning of the year. Do you think there is enough credit differentiation in the subordinated space in general and in AT1 in particular?
Gehler, DekaBank: No, a lot of the risk has been papered over because of the rate environment and yields. We need to see greater differentiation between names to reflect underlying credit fundamentals and the risk of coupon deferral.
Liller, DeAW: We have seen quite an impressive spread tightening this year in the AT1 space. From our point of view the main reason is the hunt for yield by most investors. We think that most investors involved in this paper do not take into account all the risks related to these structures. There are a lot of new features structured into these bonds, e.g. coupon deferral if distributable income is not sufficient. Also, we are concerned that data availability on balance sheet, income statements and capital are currently not sufficient to fully value triggers and other structural features of AT1.
Hoarau, CACIB: In the CoCo space, it has been almost one way only since the beginning of the year. And the recent ECB rate cuts combined with a series of targeted longer term refinancing operations (TLTROs) just added another round of momentum to the already phenomenal credit rally. Peripheral Tier 2 and AT1 capital benefited the most from the excess liquidity and the central bank liquidity easing measures taken over the last couple of years.
If we look at the situation in Spain, for instance, the speed of the recovery has been impressive and this has translated into better than expected financial results and much stronger capital ratios in the Spanish banking sector. So appetite in good Spanish names has picked up massively from non-domestic investors. If we look at Banco Popular Español, for instance, it placed the first euro denominated AT1 in euros at 11.5% in October 2013. The perpetual non-call October 2018 is yielding inside 6% now, less than 150bp wide of Deutsche Bank’s AT1 on a curve adjusted basis. I think convergence has gone too far, too fast, even if we know that everything out there is driven by the liquidity situation at the investor end rather than by fully rational elements.
Another example is that the credit spread differential between Deutsche Bank and Santander in euro AT1 is not wider than 50bp if we adjust the curves. In US dollars, Santander is trading very close to SG AT1.
But in the “plain vanilla” Tier 2 space, credit differentiation is much more evident. We priced LBBW’s 12NC7 50bp inside peer Aareal in the same format, although half of it was placed with international investors. Part of that has to do with the way you approach the pricing. In the “vanilla” Tier 2 segment, investors are focusing on spread rather than coupon, while in the AT1 market total outright yield is the focal point. This partially explains the flatness of most AT1 curves and the lack of credit differentiation across names and jurisdictions. In spread versus swaps, Barclays’ euro AT1 curve is inverted, for instance.
I think that investors’ attitude towards pricing will evolve, and spread focus will predominate at some point. And so anomalies or mispricing will disappear.
Sauerschell, Union: A year ago we saw new AT1 issues with coupons of 8% or 9%. Now we are at around 6%. In my opinion investing in new AT1 issues with coupons at or below 5% does not make sense.
It’s a very young market, it’s also supply and demand driven, and it is not very efficient. If you look at Credit Suisse in Swiss francs and US dollars, for example, they trade much tighter in Swiss francs, and this is simply because of the much lower supply in that currency. However, the market should become more efficient if there is more issuance.
Also, there is little differentiation in the market based on the time to call. When we modelled the impact of different factors on bond prices — such as the distance to trigger, five year CDS — we found that this maturity feature was not significant.
The market seems to care more about the prevalence of excess liquidity rather than structures, rationales and metrics. Are you differentiating much between structural features, buffer and distance to trigger?
Gehler, DekaBank: Yes, buffers and distance to trigger are key, although they are secondary to underlying credit fundamentals.
Liller, DeAW: Yes, we consider these features in our investment process and will value each AT1 structure on a standalone basis to derive our investment rationale.
Summan, CACIB: We believe the discretionary coupon element within the AT1 structure is not being priced correctly. Whilst the coupon is subject to restrictions subject to the MDA (Maximum Distributable Amount), other factors such as earnings volatility, dividend policy and the stance of the local regulator are not fully factored into spreads.
Hoarau, CACIB: Credit spread differentials between names are limited and this has been outlined many times already. It is even worse when it comes to differentiation among structures. A lot of investors continue to ignore metrics and when they do care their views can diverge a lot. MDA is a key figure, for instance, and I am not sure that it is considered properly.
Given the costs of repo operations on these instruments, running long/short strategies and arbitrages between signatures and structures is a great challenge. So it is difficult to try to correct or take advantage of market anomalies. This will not change unless oversupply appears and volatility comes back.
It is remarkable that more and more hedge funds have started hedging AT1 instruments that have equity conversion bail-in features by using equity puts. However, mainstream fixed income portfolio managers are not mandated to engage in such equity derivative instruments.
AQR is ahead of us. To what extent could this affect the market? Is there the risk of a severe correction?
Gehler, DekaBank: Any correction due to AQR is likely to be more idiosyncratic than market-wide.
Winkler, Berlin Hyp: We don’t believe that the AQR will affect the market. The stress test should be much more important. In the case that many of the monitored banks should feel the need to strengthen their capital base after the results are published this should not leave the market unaffected. For banks with poor stress test results, market access will be more difficult. Anyway, the ongoing low interest rate environment combined with the continued risk appetite of investors should support the market, we believe.
Summan, CACIB: We do not expect the AQR to have a substantial impact. The banks have had over a year to prepare for the AQR. We have seen banks aggressively deleveraging their balance sheets as well as raising equity through rights issues and selling non-core assets.
Hoarau, CACIB: AT1 valuations are excessively expensive. The massive spread compression across asset classes has been triggered by cash rich investors forced to buy higher yielding assets deeper in the capital structure, and across jurisdictions. But I can’t imagine that this is sustainable in the long run. The current demand-supply mismatch and the huge amounts of redemptions in FIG added to the recent ECB’s liquidity easing measures are certainly sensible motivations.
But again, we often get the feeling that some investors are ignoring relative value schemes and metrics because of that. Looking at current valuations and the lack of credit differentiation, I think that a correction is overdue. It would be more than healthy. It can’t be one way only for such a long time and in such proportion. Core AT1 yields are currently similar to 10 year covered bond yields in 2009 while AT1 can serve as equity! Is that justified by the development of the economic situation across Europe and the shape of the banking sector? I am not convinced, even if things have significantly improved over the last 18 months and outright yields have dropped massively since the crisis erupted.
The central bank will release the result of the health check AQR in one big announcement in October. But within banks, any capital holes will be known very soon and the results of the assessment could leak out chaotically. At the moment, only good news is priced in. So any disappointments, surprises or rumours based on the AQR stress test could easily lead investors to take profits. Since the AT1 segment is lacking a dedicated investor base, with dedicated real money investors and dedicated funds, any reversal could be severe.
To what extent do you expect the application of subordinated bail-in upon any failure of the EBA stress test baseline scenario?
Gehler, DekaBank: De minimis for now given the risk appetite and hence potential for mitigating actions via asset sales or equity raising.
Do you expect issuers to cover any stress test shortfall with higher-trigger AT1 instruments, as prescribed by the ECB?
Gehler, DekaBank: Yes, that would be a logical course of action. It could also be high trigger Tier 2.
To what extent do you value the presence of subordinated debt as a defence against senior unsecured bail-in? Do you differentiate between banks based on the level of bail-in-able subordinated buffer?
Liller, DeAW: We take this into account for our investment rationale on senior bonds.
Sauerschell, Union: The issuance of AT1 is positive for senior unsecured bonds in that it helps reduce the possibility and magnitude of a bail-in, as it increases the CET1 ratio. However, we consider the quality of banks’ business models and management to be more important factors, and also invest in banks with lower subordinated debt buffers.
Are you satisfied with the level of liquidity in the secondary market?
Sauerschell, Union: Liquidity is much lower than for other asset classes, such as industrial hybrids. The liquidity of new AT1 issues has also been lower than we would have expected in the secondary market given that the order books have been as much as 10 times oversubscribed. We are therefore aware that there is hype around some of these new issues and sometimes we pull our order when the spread moves too far between initial price thoughts and the re-offer.
As a liquidity provider, what are the main challenges you are facing?
Summan, CACIB: Liquidity and in turn volatility is improving with every new issue. With more real money mandates in the space we have seen deeper pockets of liquidity. Bond allocations have slowly changed hands from private banks and fast money to real money long-only funds who provide stability. Indexation of the CoCo market could help improve liquidity. We need the large German asset managers and other regional managers to sign on to the product. Dealer inventory still remains light and being able to manage volatility is a still a concern.