Since October 2015 the European banking sector has been under severe pressure. This is reflected in the -29.6% return for European bank stocks since then, compared to a +0.65% return for the broader EuroStoxx 50 Index. This massive divergence can be attributed to an environment of unprecedented low interest rates, tighter regulation from the European Banking Authority (EBA) and, finally, the recent Brexit decision, which further increases uncertainties for the sector. In a nutshell: European banks are experiencing falling revenues within a slow-growth environment and are facing regulatory pressure to increase capital.
On top of these known factors, the EBA will release the results of the second European bank stress test on July 29, 2016. This is expected to expose additional risks to the system, especially for Italian banks.
Within the European banking sector, Italy’s banks are viewed as a particular weak spot. Concerns stem from historically fast rising levels of non-performing loans (NPLs) combined with the low capitalization of many institutions. The second EBA stress test since 2014 is designed to evaluate and compare banking risk across the European Union (EU). We think two issues, especially with respect to Italian Banks, are worth highlighting.
First, in the EBA’s effort to harmonize European regulation, the definition of NPLs was revised, resulting in a far broader scope of loans to be included for Italian banks. The key terminology is “nonperforming exposure (NPE)”: This raised the level of NPLs in the Italian banking system from EUR 210 billion previously to around EUR 360 billion now. The newly added EUR 150 billion have a far lower coverage ratio of 25% (the coverage ratio represents the provisions banks have already made for bad loans as a percentage of those loans) as compared to the 59% coverage ratio for bad loans accounted for under the previous definition. The higher level of 59% is actually around the European standard. In absolute numbers, Italian banks would need to increase their loan provisions by around EUR 48 billion to be on par with their European peers with respect to overall bad loan provisioning.
Second, the EBA stress test puts the book value and the actual market value of NPL in the spotlight. Using Deutsche Bank Research assumptions, the actual market value in case NPLs are sold might be below current book value. Using their case, it could be 20% below already reduced book values. That would amount to additional provisions of around EUR 43 billion for the Italian banking system.
To make things worse, a combination of two events occurred in the past 4 weeks. First, the result of the Brexit referendum put ad-hoc pressure on European bank stocks starting June 24, 2016. Second, a letter from the ECB notifying an Italian bank (Banca Monte Dei Paschi Di Siena Spa) in early July that they need to reduce their bad-loan exposure by around EUR 10 billion by the end of 2018 (source: Bloomberg, July 3, 2016). That again highlighted the pressing need to take a look at the Italian banking sector overall.
It is important to note that the EBA stress test will result in “capital guidance [that] does not constitute any form of binding capital requirements” (July 1, 2016 EBA)1. This means that there is no outright call for additional capital in case of a shortfall – there is no fail or pass. However, it suggests Italian bank risks are high compared to some European peers even though the Italian government has taken some steps to alleviate the situation. Among other things, the government orchestrated the creation of a EUR 4.25 billion bank-support fund called “Atlante”, sponsored by non-government investors, to for example purchase NPLs as well as bank equity in April 2016.
In total, there are good reasons, like rising NPLs, higher potential provisions and possible regulatory action, to take a very close look at the Italian banking sector – overall, but even more on a bottom-up basis per institute.
We expect some form of state support to be needed for select Italian banks. The key question is how this can be done under the rules of the European Bank Recovery and Resolution Directive (BRRD), which was enacted as of January 2016. These rules basically do not allow state aid to banks until at least 8% of total liabilities including capital have been “bailed in”. Such action would therefore hit not only banks’ equity investors but also holders of bonds, with the usual cascade of capital rankings applied (first losses on subordinated bonds). This creates a significant degree of political risk as Italian retail clients have heavily invested in these subordinated instruments. An adverse outcome for private investors might in fact have negative consequences on the Italian Senate referendum scheduled to take place in October/November this year. The referendum is on reforming the Italian Senate, but has even bigger consequences as Italy’s prime minister Matteo Renzi has staked his political future to its success, meaning that a failure could hurt the pro-European Union political forces in Italy.
The following three scenarios represent our basic thoughts on the outcome on this matter. This is not and cannot be an exhaustive overview, but could serve as a guide in interpreting possible results and reviewing this in an overall portfolio context. Further, please note that we do not comment on single names, but will make a general assessment of the situation.
In our base case, there is an orderly process under the BRRD. Banks which are deemed to be undercapitalized based on the EBA stress test could be required to sufficiently recapitalize within a reasonable period of time. For banks that cannot fulfill this requirement on their own, we expect government support triggering burden sharing by equity and subordinated-debt holders; senior debt would be left untouched. This decision should result in viable banks working on a going-concern basis, reducing the risk of European contagion. Risky assets could be provisioned according to market prices and could either be sold or transferred to a second privately funded rescue fund (Atlante II/ Giasone; actual set-up to de defined) or to a bad bank to be orderly run down.
Still, this base case would require the political will to expose (private) Italian investors in subordinated debt to losses, making this “bail in” a politically hazardous undertaking, particularly as it might affect the result of the Senate referendum. Currently there are discussions of differentiating between private and institutional investors, effectively compensating private investors for some/all losses to reduce the negative political impact. While this may be possible, it seems unlikely to happen. Compensation for potential mis-selling of subordinated debt to private investors as an inappropriate investment would be more likely and could go over more easily, but it still bears some political uncertainty.
Beyond this, a unilateral decision by the Italian government to change the rules and offer state aid without burden sharing is not likely as it would clearly undermine the European banking union and, more broadly, confidence in the European Union.
In a worst-case scenario, banks following the EBA stress test would be required to increase provisions and/or reduce their bad loans in (too) short a period of time, resulting in massive capital shortfalls. In such a situation, we would expect the ability of those banks to tap the capital markets as very difficult. If in such a situation the EU and Italy fail to seal an agreement on government-funded recapitalization or state aid on the basis of “exceptional circumstances” (Article 107 of the BRRD), this could make things significantly worse. The banks would end up in a “gone-concern” situation; full resolution laws would be enforced. In such a scenario, all stakeholders would be affected and the bail in would not only be shared by equity and sub-debt holders but also by senior-note holders. This might undermine confidence in banks overall, creating potential liquidity issues in addition to solvency issues. Such a broad-based bail in could induce a sell-off in European bank equity and debt. Further, we would expect severe knock-on effects in other markets, with financial stability put at risk, raising uncertainty about the EU’s future. We see this as an unlikely scenario.
The best-case scenario would be if European regulation finally allows the injection of public money into under-capitalized banks (potentially based on Article 32, BRRD). This might be argued on the basis of it being a “precautionary measure” that does not trigger a bail in, i.e. without the required 8% burden sharing. However, it is unclear how this could be achieved without bending the rules set by the regulatory bodies. Currently we see this scenario as unlikely, but from an investor’s perspective it would be the best outcome. Still, such a solution would likely trigger a reduction in confidence in European regulation and potentially the EU itself.
As outlined, the planned publication of the EBA stress test on July 29 will be an important date. Further to that, the European Court will opine on July 19 about state aid previously extended to Slovenian banks in 2013. At the time, the Slovenian government put more than EUR 3 billion into the banking system, wiping out about EUR 600 million in subordinated bonds. On a cautious note, the ECB has authorized Italy to provide a precautionary liquidity line of EUR 150 billion for banks until the end of this year. Also outside the EU, there is some general support for the idea that state aid is possible. The IMF has tried to help the process along by arguing in the past that there is “adequate” flexibility within the BRRD and existing state-aid provisions to allow this to happen.
As mentioned at the beginning, a main issue banks are currently struggling with is the low-interest-rate environment and the flat yield curve. Taking this into account, the ECB might find (or already does find) itself in a stretch between required low interest rates to accommodate the economy (low rates to stipulate demand and investment) and the low rates itself hurting the main players in the monetary transmission mechanism: banks. Further, a government-funded bail out in Italy could increase Italy’s debt-to-GDP (gross domestic product) ratio – already Europe’s highest – further. That, in turn, then could lead to widening Italian yield spreads versus Germany, perhaps sparking talk of more ECB quantitative easing (QE) and asset buying. The resulting even lower bond yields in other European countries might further hurt profitability of the financial sector in other countries if the result is again a flatter yield curve. There is for sure more to come and to be discussed.
Even after an effective intervention, we believe that there will be reasons for investors to keep a close eye on Italian banks (and European banks overall). The Italian banking sector, even if recapitalized, faces longer-term structural problems (e.g. a high cost base resulting from excess staff and branches, implying a need for consolidation). European bank earnings are expected to go even lower this year and this might exert further downward pressure on bank stocks. Falling equity prices might make it more difficult for banks to raise capital externally at a time when low and flat yield curves make it difficult to raise capital internally (via retained profits). This in turn could impact banks’ ability to lend – and bank lending is a much more important source of corporate financing in Europe than in the U.S. The ongoing worry is that while the immediate shock of a banking collapse is averted, the European economy could still suffer from lower availability of credit.
These are all very big “question marks”, even if there is general agreement that prevention is the best strategy. So we will probably end up with a “muddling-through” scenario.
We believe that there will be some form of bail out for Italian banks. Although Italian banks’ problems are currently crystallizing against an uncertain economic and political background, one big positive is that the EU now has a formal bank rescue system in place, based around the BRRD and the European Stability Mechanism (ESM). However, while such an intervention is likely to help to reduce tail risks, it is unlikely on its own to resolve the underlying issues. Further volatility is to be expected, and the potential for additional problems to emerge will remain.
A clear understanding of potential spillover mechanisms, a thorough bottom-up stock and credit selection and decisive actions in implementing the views will be key going forward.