Beware of regulator discretion

Investors can’t say they haven’t been warned. Stephen Roith, a partner at Sidley Austin in London, says that it is impossible to predict what will cause the next financial crisis and, therefore, which triggers on new-style hybrid capital securities will be pulled. But one thing is certain: “the next time banks get into trouble, these are the securities that will be on the front line,” he says.

Although their methods can sometimes appear circuitous and poorly co-ordinated, what governments and regulators are aiming to achieve is clear enough. “Investors should work on the assumption that regulators will be doing everything they possibly can to ensure that taxpayers aren’t forced to put their hands into their pockets before bondholders and even possibly before shareholders,” says David Howe at Sidley Austin. 

That, as he says, is precisely what loss-absorbing securities are designed to ensure. It was their abject failure to do their job during the financial crisis, however, that has led to wholesale changes in regulation and to the wave of product innovation that has given rise to the development of Cocos and other new generation bank capital instruments. 

It is easy to see why policymakers are so single-minded about making sure that loss-absorbing instruments do what they’re supposed to do. According to the European Commission, the total amount of aid offered to banks during the financial crisis amounted to 30% of GDP. Although the amount actually used was less than half of this, at about 13%, this remains an eye-watering total. 

Small wonder that governments and regulators have left no room for doubt in their approach to creditor subordination. “There is no question that governments, regulators and, as a result, issuers are all going out of their way to communicate their message to investors,” says Roith. “Investors should be under no illusion that politicians across Europe want banks to be smaller, de-risked and more boring.”

The consequence is that although regulation governing subordinated debt may appear to be punitive for investors, it is seldom opaque. Indeed, bankers say that where structures of new generation Cocos and additional tier one securities look complicated, this is not the product of deliberate over-elaboration, but an inevitable reflection of a regulatory regime in a state of flux. 

Take the example of the $1.5bn additional tier one (AT1) transaction from BBVA at the start of May. The first of its kind to comply with the new capital requirements directive (CRD IV), the BBVA structure was regarded by many as complex because of its multiple conversion triggers.

“The structure of the BBVA AT1 has been described as complex, but frankly it wasn’t that complicated,” says Khalid Krim, head of capital solutions EMEA at Morgan Stanley in London. “There are multiple triggers in the structure, but with good reason. It is the EBA Coco instrument combined with an additional tier one. This means it is effectively two instruments rolled into one, which is necessary because we are still in a regulatory transition phase.” The apparently cumbersome structure, Krim adds, is a function of the issuer covering all bases by ensuring that the securities comply with both current and future EBA and Bank of Spain requirements. 

“For as long we remain in this transition phase, we will continue to see structural differences across jurisdictions,” says Krim. “That said, we are clearly moving towards greater regulatory harmonisation in Europe, with CRD IV and CRR ensuring that there will be reduced scope for discrepancies and differences across the EU. For the first time in the history of the European bank capital space, we will have regulation that applies in the same way in each EU state.”

Others agree. “The key point is that CRR is a regulation, and with EBA technical standards imposed on top, there will be very little scope for arbitrary or varying interpretations,” says Roith. “That is distinct from the Recovery and Resolution Directive (RRD), which as a directive will be interpreted and implemented on a country-by-country basis, which will still leave room for differences in interpretation as we have seen in the past.”

None of this is to suggest that investors are completely comfortable with some of the legal detail associated with trigger-based contingent capital instruments. Jackie Ineke, managing director at Morgan Stanley covering European financials credit, says that the concern most often expressed to her by investors about additional tier one securities is on lack of dividend stoppers. “Investors’ greatest concern is the fact that coupons can be cancelled whereas equity dividends can continue to be paid,” she says. “Rabobank has tried to address this issue by harmonising the date they pay dividends to holders of member certificates with payments on their tier one coupons. However, this doesn’t remove the possibility that still, shareholders can be paid and tier one holders not.”

Perhaps ironically, however, investors most at risk from regulatory change are more likely to be senior bondholders than those buying into Cocos and Coco-like securities. This is because, as consultant PricewaterhouseCoopers (PwC) explained in a recent briefing: “The only real distinction [between bail-in capital and Cocos] is that the conversion trigger for bail-in capital is at the discretion of the regulator, whereas for Cocos it is a fixed trigger.” 

This implies that predefined triggers can’t be subject to revision. “The issuer can’t usually change unilaterally the terms of the issue,” says Sidley Austin’s Howe. “In other words, if an issue is structured with a low trigger it wouldn’t usually be capable of being revised to a high trigger without noteholder approval.”

The distinction is important, because it suggests that while for Coco and AT1 holders conversion triggers are clearly defined, senior bondholders will continue to be hostage to regulator discretion. It is this regulatory discretion that will determine the point of non-viability (PONV), the definition of which is vague. In a recent Bank of England speech, for example, PONV was described as being reached “when the supervisory authority identifies that the institution was failing, or likely to fail, and that no other solution, absent the use of resolution tools, would restore the institution to viability.” 

It added that “resolution powers would only be used if it was necessary to employ them in order to meet clearly defined public interest objectives, for example the maintenance of financial stability and the protection of depositors.”

The problem with these definitions, for senior bondholders, is self-evident. This is that the point at which bail-in can be triggered is based purely on interpretation of highly subjective terms such as “likely to fail” and “public interest”.  “The Swiss have published more detail on PONV, but we have nothing that investors can truly analyse,” says Ineke. “From the perspective of investors, no progress has been made on pinning down a definition of PONV.”

“In the case of products that have been structured to have objectively determinable capital triggers, you have a reasonably clear picture of where you stand as an investor,” says Howe at Sidley Austen. “But if your product exposes you to what the bail-in regime happens to be in the future, then it is much more difficult for investors. You are then exposed to the more nebulous and subjective concept of PONV — and the thing to remember about that is that it looks like bail-in risk is going to apply to all bank debt.”

Although it may be difficult — and possibly dangerous — to draw decisive conclusions from the two major credit events of this year, bankers say that both have sent out reasonably conclusive, and ominous, messages to senior bondholders. Granted, investors in senior debt were spared a bail-in when SNS Bank of the Netherlands was nationalised in February, while equity holders and subordinated bondholders (up to tier two) were wiped out in what Fitch described as “the harshest burden-sharing at a rated bank since the onset of the eurozone crisis.”

But as Morgan Stanley’s Ineke says, it would be a mistake for senior bondholders to take SNS as a blueprint. “There were a number of operational difficulties with bailing-in senior SNS bondholders,” she says. “But Dutch finance minister Dijsselbloem made it very clear that in the future, when RRD is in place, senior debt will be fair game.”

The Cyprus bail-out, meanwhile, involved a sizeable bail-in dimension, with large contributions from bank debt holders at all levels and senior lenders, as well as wealthy depositors. 

Sidley Austin’s Roith says that the Cyprus event was probably not a useful precedent for bondholders, because the catalyst was a sovereign debt crisis which was dealt with by the Troika rather than a banking regulator, but it did produce a seismic shock for depositors.

Nevertheless, Roith says that Cyprus was another reminder of the value of the clarity that will be enshrined in the RRD. “Cyprus shows how important it is that everybody knows at the outset of a crisis which liabilities are going to be bailed in and which will be left untouched,” he says. “And it is to be hoped that the RRD will set out the bail-in hierarchy very clearly.”

Despite these precedents, Ineke says that many investors remain surprisingly complacent about the prospects for senior debt being subject to bail-in. “I think investors may be in for a shock when the rating agencies start downgrading senior debt in response to RRD,” she says.

The probability of senior debt being bailed in has significant implications for pricing and liquidity in the market for subordinated debt. Ineke says that Morgan Stanley advised investors to take an underweight position in senior debt last October, for several reasons. “We went underweight based on valuations and because senior debt is more vulnerable to downgrades than sub debt,” she says. “We also expect depositor preference to be included in RRD, which means that as a senior bondholder you may be subordinated to 40% of the balance sheet.”

While RRD may bring more clarity, for the time being, investors in senior debt will need to make a judgement on the likely response of national regulators to banks judged to be failing or likely to fail. As that response will differ from country to country, and will be based on a range of economic, political and social influences, it is likely to be translated into differences in trading levels.

In the subordinated debt market, many of these pricing differentials have already been ironed out. “Two or three years ago, the tier one spreads of UK banks traded wider than the rest of the market, and especially relative to the French, Dutch and German banks,” says Krim. “That was based on the market’s perception that the FSA had stronger bail-in powers under the UK’s Banking Act than many of its peers elsewhere in Europe. Those differences in spreads have disappeared as investors have recognised that legislation elsewhere in Europe is being brought into line with the UK’s.”

Another complication arising from heterogeneity of legal regimes across Europe is the theoretical conversion of debt to equity, which could again lead to differences in pricing levels of bail-inable debt. 

“Issuing subordinated debt that converts into equity is relatively straightforward for a Spanish bank like BBVA, where corporate law is accommodative to companies asking for shareholder approval to issue convertibles on a non pre-emptive basis,” says Krim. “In the UK it has been more difficult for banks to set a conversion price and issue on a pre-emptive basis. Industrial corporates have blanket approval to issue convertible bonds, but banks don’t yet have the same freedom.” 

Barclays, Lloyds and RBS have all secured approval from their shareholders at recent annual general meetings to allow for non pre-emptive share issuance, which is valid for a year and needs to be renewed on an annual basis. 

While UK issuers appear to have addressed the issue of shareholder approval for conversion, in some other European jurisdictions the process may not be so straightforward. Sidley Austin’s Vivian Root points out that French corporate law on pre-emptive rights, for example, may make conversions unwieldy or even impractical. That, she says, may have significant market implications if it leads French issuers to adopt write-down rather than equity conversion structures, when investors’ preference is generally for conversion. 

Cross-border issuance of Cocos also creates at least one notable legislative complication for issuers and investors, given that US market practice and underwriter risk management policies have meant that, traditionally, virtually all SEC-registered issuance has been governed by New York law. Article 50 of the RRD, however, rules that an additional contractual term is required stipulating that the bonds are potentially subject to bail-in under the powers in the RRD.

“Governing law on US deals will be a big question,” says Sidley’s Walsh. “Historically, it has been the practice on US SEC-registered and 144A deals for local law — let’s say, Dutch law — to govern subordination and New York law to govern everything else. The recent Barclays deal followed this approach by using New York law to govern everything apart from subordination.”

“But due to article 50,” Walsh adds, “some European issuers may now push to have their US deals governed exclusively by local law.” That, he says, may give rise to serious concerns among US investors about the choice of a civil law such as those that govern subordination in many European countries.

To date, this does not seem to have had an impact on distribution or pricing in the US. “We would be dealing with a liquidation scenario, and historically only a court in the issuer’s country of incorporation has the competence to deal with matters related to insolvency,” says Krim. “So from a marketing and pricing standpoint, we see no objection in having the issuer’s governing law covering loss absorption at the point of non-viability.”

The question of subordination law, Krim adds, should be regarded chiefly as a disclosure issue. “From a marketing perspective, the key point is to ensure that investors are warned in advance that this would be governed by the issuer’s law.” To date, he adds, investors have not demanded a premium for deals from issuers such as Barclays, UBS and Deutsche Bank in which any litigation arising from losses borne by bondholders recognises the authority of a UK, Swiss or German court.

Nor, thinks Krim, is it any longer likely that investors will require different premiums for different European laws governing subordination. “18 to 24 months ago this may have been a valid concern,” he says. “Today, if you believe that Europe is moving towards banking union, you should not be differentiating between Italian or Spanish and UK or German law.”

Perhaps. But at Sidley Austin, Walsh says that one solution might be for English law, a common law system — as distinct from European civil law — to be applied to the instrument other than the subordination provisions. “There will continue to be a strong preference for New York governing law, but English law could be a sensible compromise to the extent that concerns arise about Article 50,” he says.

The temptation may be to assume that most if not all of these legal and regulatory complications have to date been largely theoretical considerations, given that issuers in the nascent Coco market have generally been national champions which nobody seriously believes will be in danger of breaching their conversion or write-down triggers.

But as Ineke cautions, that may also be a dangerous assumption that makes the mistake of looking just at the capital that defines the trigger. “Investors are looking at the gap to the trigger,” she says. “But we think they will need to pay increased attention to the denominator of the ratio, which is risk-weighted assets (RWAs).”

Ineke cautions that the size of the buffer could itself become vulnerable to regulatory discretion. “We’ve just seen the introduction of a floor on the risk weightings of residential mortgages in Sweden, which is the sort of thing that could ultimately cut into core tier one, if introduced as a Pillar 1 measure,” she says. “So while as an investor you might be feeling very comfortable with a 9.5% CET1, regulator discretion on RWAs could easily drop this by 100bp, or more, in short order.”