The story of developments around the European banking sector has turned into a convoluted and complex tale.
A strong regulatory sense of purpose and direction of travel around capital, liquidity and, more specifically, bank resolution has run headlong at a macro level into a potentially toxic triumph of politics over policy and at a micro level into puzzlement around how point of non-viability is determined and what processes ensue.
The result has been what look to all the world like finagled smoke-filled backroom solutions – the post hoc restitution of retail investors for subordinated debt exposure to failed banks an act of pure political populism – that asked more questions than they have answered.
Markets dislike uncertainty, but uncertainty looks set to be all-pervasive while banks and investors lack specifics even with regard to known elements (eg detailed MREL requirements, impacts of the impending IFRS 9 NPL accounting convention and MIFiD II implementation). Let alone outcomes of in-process streams: Fundamental Review of the Trading Book, the latest iteration of the Capital Requirements Directive (CRD V) and Basel guidelines (Basel IV) with its RWA modelling, US re-regulation, and Brexit effects.
Amid the uncertainty, however, banks have resolutely set about updating market participants to the best of their ability, not shy about commenting on potential impacts and expected outcomes. And they have been tapping the markets to meet actual or estimated capital and bail-in debt targets, invariably with aplomb.
In this, they have been helped immeasurably by a yield environment heavily distorted by ECB asset purchase activities and monetary policy that have crushed credit risk-return metrics and created an environment of rampant yield tourism. But they have also been assisted by some inspired product innovation.
The creation of senior non-preferred (SNP) debt in 2016 has been a huge success. “The emergence of senior non-preferred debt is a welcome development for continental European banks not organised with holding companies as it provides an equivalent instrument to senior holdco used by UK and Swiss banks for TLAC/MREL,” said Khalid Krim, head of EMEA FIG capital markets advisory at Morgan Stanley.
“This second class of senior debt offers a cost-efficient alternative to Tier 2 to comply with MREL requirements. It is also a clear and distinct tool to traditional senior funding enabling EU banks to draw a line between capital securities issued for regulatory/resolution purposes and debt securities issued for funding and liquidity.”
The emergence of SNP debt fuelled a huge rally in senior preferreds in 2017 driven by a supply-demand imbalance such that senior preferreds entered spread territory occupied by covered bonds and secured debt.
Around a dozen predominantly French and Spanish but Danish and Belgian banks too had hammered SNP to the tune of around US$32.5bn in 2017 by mid-September via multiple benchmark and smaller trades in euros, US dollars and yen, plus trades in half a dozen other currencies, and in fixed and floating-rate formats.
Issuers encountered an audience more than willing to book the pick-up – all while the market is still uncertain as to whether SNP pricing should buttress against traditional senior or Tier 2 lines.
“The market has been very focused on issuers establishing their TLAC or MREL curves, especially the pricing differential between senior non-preferred and senior preferred,” said Christoph Hittmair, global head of FIG DCM at HSBC.
“A further interesting dynamic going forward will be issuers seeking to maintain both an “old-style” senior preferred curve and a “new-style” senior non-preferred curve. This presents bank issuers with different questions on how they approach the issuance of these two instrument types. Which instrument is more suitable for benchmarks, which instrument for private placements, and should there be a differentiated issuance strategy for the two instruments in terms of domestic or international investors or in terms of currency/market?”
Having a separate senior non-preferred bucket certainly offers clarity around where each creditor sits in the capital stack, which investors and other stakeholders such as ratings agencies and resolution authorities have welcomed. Even though MREL calibration and composition – ie the mix across buckets – remains in flux, market participants have been supportive, helping the emergence of this new asset class.
Nik Dhanani, global head of the financing solutions group at HSBC, expects that for many banks, senior non-preferred debt will replace a significant portion of existing unsecured wholesale funding.
“It will be interesting to see if total capital levels begin to have more of an impact on TLAC/MREL spreads in the future, as this might influence how and to what extent banks issue Tier 2 in excess of the amount that is efficient for regulatory capital,” he said.
Even with the resolution events in Spain and Italy and earlier wobbles around the solvency of individual banks, the Additional Tier 1 market has similarly seen solid demand. The national champion names that have dominated the market have seen heavy levels of oversubscription, and that has paved the way for the next leg of AT1 market formation: second-tier names from core jurisdictions, as well as non-core names and EM issuers. On the basis that the BofA Merrill CoCo index tightened from 6.14% at the start of the year to 4.775% at one point in August, the message is clear.
The question on everyone’s lips is what happens to the market for capital and bail-in debt once the European Central Bank ends its unorthodox monetary policy experiment, the asset purchase programme comes to an end and the market starts to revert to normal yield-curve patterns; the economic cycle turns and we enter an era of monetary tightening; public funding sources start to be withdrawn; and banks that have deferred MREL issuance start to attack the market to build their buffers.
Expectations for a tapering cliff event that had been spooking market participants amid fears of market volatility may have lessened, but the uncertainties caused by the process and the bursting of what many consider to be an asset bubble in credit could alter buyer behaviour and force a re-evaluation of levels. The EBA is concerned about fragmentation between large banks and small/medium-sized banks vis-a-vis capital markets access.
Mitigating against a complete re-appraisal of pricing expectations is the fact that senior debtholders were spared in the resolutions of Banco Popular, Veneto Banca and Banca Popolare di Vicenza. That provided a highly expedient example of how regulators think about red lines.
The bigger question on everyone’s lips is how European bank resolution is expected to work going forward. While individual bank distress may be idiosyncratic, EU policymakers were supposed to have created a transparent and level playing field. The way in which failing banks in Spain and Italy were dealt with has left unanswered questions.
Monte dei Paschi
The €5.4bn state bailout of Banca Monte dei Paschi di Siena under an EU precautionary recapitalisation, approved on July 4, was in some ways a nail in the coffin of the EU’s resolution framework because it flew in the face of the most fundamental tenet of post-GFC policy-making: the end of taxpayer bailouts of failing banks.
Ten years after the onset of the GFC, bailout is still very much of the European resolution toolkit. MPS’s shareholders and junior creditors may have been bailed in to the tune of €4.3bn but eligible retail junior bondholders can claim up to €1.5bn in a compensatory equity-to-senior-bonds conversion on the basis of mis-selling contraventions.
Mis-selling has become the go-to policymaker euphemism for getting around the thorny issue of compensating investors against the stated rules. In the absence of civil or criminal investigations into serial management-approved mis-selling schemes, resorting to such double-talk is scarcely believable.
Veneto and BPVi
The €1 state-assisted sweetheart deal that allowed Intesa Sanpaolo to buy the good assets only of insolvent Veneto Banca and Banca Popolare di Vicenza while Italian taxpayers took the pain was another eloquent demonstration that Europe is still a long way from the end of taxpayer bailouts.
The two banks failed the SRB’s public interest test that underpins BRRD resolution, yet the EU commissioner in charge of competition policy, Margrethe Vestager, bought the Italian government’s narrative around threats to the stability of the regional economy had Veneto and BPVi been liquidated under normal insolvency proceedings. And duly granted state aid.
The banks, creaking under the weight of a staggering 37% of non-performing loans, had already burned through a €3.5bn 2016 cash injection from bail-out fund Atlante to no avail. Between June 2015 and March 2017, they suffered massive deposit withdrawals – equivalent to some 44% – which rendered them unfit for EU precautionary recaps.
In receiving about €4.78bn of Italian government cash as well as state guarantees of up to €12bn to cover against losses, Intesa arguably did the deal of the century. Shareholders and subordinated creditors were wiped out but Intesa set aside €60m to make restitution payments to retail subordinated bondholders.
Meanwhile, Banco Popular Espanol’s heavy concentration of corporate deposits walking out of the door almost literally overnight culminated in the bank (which passed its most recent stress test and sported reasonable Q1 17 solvency marks) sinking into insolvency. Straight into the welcoming arms of Grupo Santander, also for €1.
Some market observers considered the BPE resolution provided demonstrable evidence that capital instruments – AT1 and Tier 2 – did their job as holders were wiped out. But others believe the fact that BPE’s hybrid capital instruments failed to deploy as intended, including the conversion/write-down triggers, arguably rendered the complex embedded features of AT1 instruments irrelevant in an accelerated slide into the abyss.
Under the BRRD and Single Resolution Mechanism Regulation, the SRB agreed with the ECB that BPE was failing or likely to fail and concluded there was no reasonable prospect that an alternative private-sector solution or supervisory action would have prevented its failure within a reasonable timeframe and considered a winding-up under normal insolvency proceedings would not have met its resolution objectives to the same extent. An independent valuation estimated BPE’s economic value at a negative €2bn in the baseline scenario and a negative €8.2bn in the most adverse scenario.
As with the Italian banks, the Spanish regulator CNMV on September 12 approved the issuance of up to €980m in restitution payments to junior retail creditors via non-call seven perpetual Fidelity Bonds. Creditors with exposures of up to €100,000 will be compensated in full; even bondholders with €1m will get half their money back. The bonds will not qualify as regulatory capital but they will count towards Santander’s MREL/TLAC requirements. To get their hands on the bonds, though, creditors will have to waive any legal claims against the bank.
Cries that Single Resolution Board officials made inappropriate public comments about BPE at a critical time which accelerated the bank’s demise will put the resolution process under intense scrutiny. The dozens of lawsuits now filed at the European Court of Justice and elsewhere targeting the SRB, ECB and Spanish resolution agency FROB by the likes of Pimco, Anchorage Capital, Algebris Investments, Ronit Capital, pension funds, Spanish retail investors and an investor group led by Mexican billionaire Antonio del Valle – all of whom suffered enormous losses – will see to that.
“The resolution events were country and institution specific and a reminder that, in Europe, we are still in the early days and transition phase from the previous regime to a fully fledged and operational Banking Union and resolution framework,” said Krim.
“Banco Popular was evidence that BRRD can be used to deal with bank resolution. It is also a reminder of the role of AT1 and Tier 2 as gone-concern capital in resolution. It is also worth noting that while some market participants were putting “bail-in” as the base case scenario for resolution, this is a case study showing that there are other ways to resolve EU banks.” Namely, in the case of BPE, the sale of business tool.
There are some key takeaways from the resolution actions of 2017. “One important lesson learned was that it is difficult to impose losses on retail customers, either for political or business reasons. Hence, going forward, it is important that subordinated or loss-absorbing instruments are appropriately designed and only placed with sophisticated investors who understand the associated risks to enable ease of recapitalisation or resolution,” said Sandeep Agarwal, head of EMEA DCM and co-head of the EMEA capital markets solutions group at Credit Suisse.
“Secondly, it highlighted that unless each individual capital layer is sufficiently sized, these instruments by themselves may not be enough to recover a distressed bank, particularly leaving very little recovery differential between Tier 1 and Tier 2 when both are triggered by PONV at the same time.”
There is arguably also a need for greater transparency around regulatory decision-making vis-a-vis banks in distress. Discussions are taking place now at the level of EU policy-making as to whether banks that meet certain conditions should have a moratorium imposed that among other things restricts or prevents deposit flight.
Bearing in mind that in all the 2017-vintage European resolution cases, supervisors were unable to act as depositors yanked out their cash and it was this that ultimately forced the banks’ demise, some question whether policymakers have spent too much time on capital issues – with serious knock-on effects on lending to the real economy – when failures are invariably matters of liquidity.
The notion of having independent advisors appointed in the event of a moratorium to sell assets while deposits stay – a form of bankruptcy protection – is not without merit on paper, although figuring out how it would work in practice is a challenge.
After BPE’s collapse, for example, Santander quickly did a deal with Blackstone for the latter to acquire 51% of BPE’s portfolio of repossessed properties, real estate NPLs and related assets with an aggregate gross book value of around €30bn – adding around 12bp to Santander’s fully loaded CET1 ratio. Could that have been conducted with BPE as a going concern under moratorium conditions?
Bearing in mind BMPS has been granted state support while it disposes of its €26.1bn NPL portfolio on market terms – via a privately funded SPV partially financed by Atlante II where private investors will be offered senior notes under the Italian state guarantee GACS scheme – there are parallels.
Let them go?
Another key question is what would have happened if the insolvent banks had been allowed to go bust. That may be anathema to national and regional regulators but were any of the banks under the spotlight truly systemic? There were no transmission impacts in the capital markets at the time of their passing: the markets took it in their stride.
“The resolution events in Spain were an important milestone for regulators in terms of applying some of the new tools developed to deal with instability at a banking institution. The market reacted calmly to events at Popular and negative sentiment did not spread to other peer institutions. Certainly a much better outcome for the market than events during the crisis,” said HSBC’s Hittmair.
Does that mean that claims that allowing the banks in question to fail would have caused systemic disruption were baseless? We will never know. The thing is: the European banking sector looks relatively robust. The Basel Committee’s semi-annual Basel III Monitoring Report published on September 12 shows sector-wide CET1 capital ratios under the fully phased-in Basel III framework for 105 large internationally active (Group 1) banks, including all G-SIBs, increased from 11.9% to 12.3% between June and December 2016 – an increase of 510bp since June 2011.
Applying 2022 minimum requirements, 12 of the 25 G-SIBs reporting TLAC data had a combined shortfall of just €116.4bn. That is not far from a two-thirds reduction from the €318.2bn they showed at the end of June 2016. Whatever metric you choose, there has been an improvement: Group 1 LCR up five percentage points to 131.4%; NSFR up from 114.0% to 115.8%; average fully phased-in Basel III Tier 1 leverage ratios 5.8% for Group 1 and 5.5% for Group 2 banks.
Perhaps EU policymakers need to let a bank go – maybe under laboratory conditions – to test disruption hypotheses. If ever the time was right, it is now.