Bank issuers of additional tier one (AT1) capital wonder how long the sunshine can last. The market experienced a prolonged rally during 2017, with prices falling consistently. Without exception, the handful of banks that issued twice in core currencies over the year ended up paying much lower coupons on their return.
“Every time you decided to do a deal you regretted it a month later because spreads were tighter,” says Sandeep Agarwal, head of debt capital markets EMEA at Credit Suisse in London.
As well as the familiar names, smaller banks such as Jyske Bank, OneSavings Bank and NIBC Bank tapped into strong demand and offered investors some rarity value.
“People have been buying securities and not selling them — and so where there are very small deals and it’s unlikely there will be further issuance, it has just ended up driving the price of those bonds higher,” says James Macdonald, financials analyst at BlueBay Asset Management in London.
A key moment was the Julius Baer trade. The Swiss bank was about 30 times subscribed for its $300m trade in September, achieving a final spread of 4.75%, the tightest ever for a dollar AT1. “It redefined the pricing of a tier one instrument,” says Agarwal. Nordea broke that record again in November with a 3.5% coupon.
That’s not to say the market didn’t have any wobbles. Santander and ABN Amro appeared to unsettle some investors in November, attracting smaller book sizes because of the lower reset spreads on offer.
“We hit a little bit of a barrier with the ABN Amro transaction, which suggests that we are plateauing out in terms of how much tighter this market can really get,” says Agarwal.
But overall it was a year mostly free of blemishes, and that may not change in 2018. Funds specialising in bank capital have achieved healthy returns in AT1, and their success has encouraged investors to plough more cash into the asset class.
“Over the past few years returns for AT1 investors have outstripped pretty much every other credit class and that has led to those funds raising more and more money,” says Barry Donlon, global head of capital solutions at UBS in London. “That’s happening at a time when most large banks have filled their 1.5% bucket and are doing well from a leverage ratio perspective, so the regulatory need to issue AT1s is reduced.”
But how much further can the rally go? Agarwal notes that the rewards for investing in equity are starting to look much better following the huge AT1 rally in 2017, for example.
“Cost of equity today is 9%-11% on a post-tax basis,” he says. “It is possible investors will begin to question the value of tier one capital at 4.5%, less than half the cost of equity.”
But compared with the rest of the credit market, the asset class it has not lost its appeal. Other products have been squeezed even further than AT1, and banks’ credit quality has improved sharply as they have cleaned up their balance sheets.
“The AT1 class only looks historically expensive versus itself. If you compare it with European high yield, tier two or corporate bonds it still looks like very good value,” says Donlon.
Keeping your call
2018 will mark a milestone for the product, with the first batch of notes reaching their call option dates.
BBVA, Société Générale and Barclays have dollar AT1s reaching call dates, but there is unlikely to be too much drama here. The market is next to certain that the first-generation AT1s will all be called, given the pricing advantage that issuers can get from refinancing them.
But the debate may become livelier towards the end of the year. There are many more bonds reaching a call date in 2019: 13 in core currencies. And there is a greater chance that banks will have to think harder about whether or not to leave them outstanding, as many of the instruments will reset to paying much lower spreads than the AT1s up for call this year.
Unlike in the past, the market expects issuers to decide based on purely unsentimental, rational criteria rather than feeling obliged to call bonds to please investors.
But Macdonald argues that investors shouldn’t only be paying attention to the risk that issuers don’t call their bonds; they should also look at the frequency of those call options.
If issuers do not call at the first option date, some are more limited than others in when they next have the option to call. Some AT1s can be called at any time after the first date; others can be called only at certain intervals, ranging from every quarter to every five years or longer.
“When AT1s bond were first being issued, investors were asking for issuers to be able to call only every five years after the first call date. As the market has rallied, investors have lost a lot of discipline on this,” says Macdonald. “More and more deals now are callable at any time. If you are not asking issuers to pay for that, you are giving away call options for free.”
Popular shows AT1 maturity
But even if the AT1 market is losing discipline in some areas, as new money floods in, it has also gained more control in other respects.
After the market was spooked in 2016 by the prospect of Deutsche Bank missing coupon payments, an outsider would be forgiven for thinking that if AT1s were written off completely the reaction would be apocalyptic. Yet when this happened in the Popular resolution (see p26 for article on bank resolution) the market absorbed the shock with ease.
“Compare the maximum distributable amount and payment risk concern of early 2016 and the sell-off that caused with the much more real credit events of 2017 and the lack of contagion there,” says Donlon. “This shows a real maturity.”
Bondholders are now in step with regulators in accepting that AT1s will be bailed in if necessary.
“Investors have gone away from the idea that any tier one or debt issue product is never going to absorb a loss, which was potentially some of the thinking going into the crisis,” says Agarwal. “Now that people have experienced a reasonable amount of failures and we are talking about resolution capital in senior debt format, I think the penny has dropped.”
The potential for a shock is also diminished by the reactions of specialist investors looking to profit from panic.
“You have investors who are financials experts as opposed to AT1 specialists or credit specialists. They see volatile situations as an opportunity for superior credit selection,” says Donlon. “They step in and take advantage of any sell-off and that’s why these sell-offs have been so isolated.”
Rather than dulling the shine of AT1s, the Popular resolution actually boosted the perception of value they offer relative to tier two. Popular’s tier two notes were converted to equity after the AT1s, but given that the entity was sold to Santander for €1, the distinction between how the two asset classes were treated is rather academic. The resolution underlined that if a bank reaches its point of non-viability (PONV), tier two hardly offers more protection for investors than AT1.
“People are asking ‘am I not better in tier one than tier two, so that I get at least 250bp of extra yield for risk which is likely to be the same?’” says Agarwal. “The compression between tier one and tier two is probably the right trade.”
The Popular resolution also shed light on questions about the structure of AT1s. Some bonds have a trigger when the common equity tier one (CET1) ratio falls below 5.125%, and others when it falls below 7%. But the way in which Popular failed made these distinctions seem irrelevant.
“The resolution scenarios in Europe have often been liquidity-driven and occur when the reported CET1 ratio remains very far from the trigger,” says Donlon.
Another variable in AT1 structures is whether they are converted to equity or are simply written down completely.
With Popular, it was again tricky to compare the merits of the different structures. The bank was at such a critical stage when the regulator stepped in that its AT1 bondholders gained no value from being converted into equity.
“We would argue that having equity conversion more closely aligns bondholders with equity holders, in that you see more dilution of equity: as such we have a bias towards equity conversion instruments,” says Macdonald. “That being said, we see the biggest risk as the regulator declaring the point of non-viability rather than a trigger breach, so in most circumstances it’s likely to end up being largely irrelevant.”
Room for something else?
Following the Popular resolution, some market participants started to speculate that there could be room for a new type of loss-absorbing product with a higher trigger level, or which could be converted to equity at the bank’s discretion, but perhaps within a senior or tier two host. This could help banks in stress tests and optimise their cost of capital.
“I can see some uses for new forms of regulatory capital with an equity conversion at any time but I don’t think it comes at a lower cost,” says Agarwal. “If it comes close to the cost of equity, is it not better to carry additional equity?”
At the same time, any equity conversion creates problems for banks, weakening the stake existing shareholders have in the company and giving that to debtholders — something many of the latter may not want.
“Given the high trigger that banks want to put on these and the potential dilution effects on equity holders I’m not convinced that we will see a lot of banks wanting to issue them,” says Macdonald.
Indeed, AT1s are likely to remain the only game in town.
The market shrugged off every challenge that came its way in 2017. It enters this year with the confidence of a young boxer with one high profile fight under his belt.