- Since we last commented on the changing regulatory environment for banks in the aftermath of the global financial crisis, much progress has been made in all the key areas and we believe that the regulatory jigsaw puzzle is essentially complete. The inauguration of a European Banking Union and progress towards a credible resolution/”bail-in” regime are perhaps the most eye-catching of developments.
- Less headline-worthy, but equally important, is the massively increased supervision of virtually everything that is going on in a bank. In Europe this takes the form of the SREP, making banks subject to the most intrusive supervision, on an almost real-time basis, of any industry in the world. As such it is a significant safeguard for us as credit investors.
- Banks have now largely regained their traditional pricing advantage relative to non-bank corporates. Even without the historic “bail-out” subsidy, we consider that this reversion is justified. With insight into both the macro and micro dynamics of banking systems and individual banks, we believe there is value to be found in bank credit investing.
Just under three years ago, in October 2012, we published a white paper entitled “Northern Rock Five Years On.” This paper examined the changing face of banking since the onset of the global financial crisis (GFC). We focussed on what we referred to as the “four pillars of banking wisdom”—micro-prudential regulation; macro-prudential regulation; recovery and resolution; and structural reform. How has this agenda moved on in the intervening three years? Are we closer to having a final view of the shape of the banking industry ahead and, in particular, of our role in it as fixed-income investors?
One of the problems we noted in the aforementioned paper was that the GFC, to which these measures are a response, had not itself fully run its course. Since then, we have had another banking crisis in Cyprus and, most recently, have witnessed the next chapter of the Greek crisis. Both cases have demonstrated the age-old truth that the banks stand at the nexus of the financial/monetary economy and the political economy. Essentially all the measures being undertaken by the global regulatory community, with full backing from the political community, are aimed at reducing the potential friction at that nexus. To the extent that that friction cannot be removed entirely, the complementary aim of regulators and politicians is to pass a substantial proportion of the costs of that friction on to the private sector via both the equity and fixed-income markets. The question is whether the market can understand, accept and price these costs appropriately.
The realm of micro-prudential regulation—the detailed everyday rules by which banks operate—has seen the most progress. The Basel III agenda has been substantially translated into legislation and we are progressing through the various transitional periods towards full implementation of significantly tighter standards for both capital and liquidity. In practice, most major banks have been front running the new regime and are fully compliant not only on a transitional basis, but on a so-called fully loaded basis. Capital positions have improved significantly relative to the pre-crisis period. These measures substantially improve banks’ capacity to absorb shocks, but it is important to remember that banks remain highly leveraged (both by comparison with non-bank corporates and by historic banking standards, see Exhibit 1) and reliant on confidence-sensitive funding.
On top of that, the micro-regulatory regime, while laudable in its aims, has a staggering degree of complexity. As well as entailing a substantial cost burden for both the banks and the regulators, this also raises substantial issues concerning disclosure and transparency.
The macro regulation pillar remains largely untested, as it is designed to diminish some of the pro-cyclicality that we have witnessed in previous financial crises. Nevertheless, some regulators are using their discretion to raise system-wide capital requirements in the face of potential asset price bubbles. For example, Sweden has announced successive increases in the countercyclical buffer, rising to 1.5% from June 2015. This comes on top of raising the minimum risk weight for mortgage loans from 15% to 25%. With Swedish house prices continuing to rise, it would appear questionable how effective these tools have been in actually dampening the credit cycle, although it has to be said that the Swedish banks are now consistently among the best capitalised in Europe.
In the area of structural reform, the focus has been on the establishment of a “banking union” across the 19 member states of the eurozone common currency area. As the European Commission explains: “As the financial crisis evolved and turned into the eurozone debt crisis in 2010/11, it became clear that, for those countries which shared a currency and were even more interdependent, more had to be done, in particular to break the vicious circle between banks and national finances… That is why, in June 2012, the Heads of State and Government agreed to create a banking union.” The key components of a banking union are:
- A single rulebook under the auspices of the European Banking Authority (EBA),
- Single Supervision under the auspices of a newly created regulatory division within the European Central Bank (ECB) and
- Single Resolution under a Single Resolution Board made up of permanent executive members as well as representatives of the EU Commission (the Brussels bureaucrats/technocrats), the Council (i.e., the heads of government of the participating states) and the national resolution authorities.
The institutional framework has largely been put in place with the ECB taking over supervisory responsibility for the eurozone’s 5,500 banks in November 2014 after the completion of an extensive comprehensive assessment and stress test exercise. Given that the 123 largest banking groups directly supervised by the ECB represent close to 85% of total banking assets within the eurozone, the massive scale of the undertaking needs to be acknowledged. Nevertheless, major questions remain as to how this arrangement will work in practice, and recent events in Greece highlight the scale of the difficulties with the ECB struggling to reconcile its roles as regulator, lender of last resort and monetary policy setter. It is too early to judge whether the ECB has successfully kept its balance on this highly politicised tightrope. It can be argued, perhaps fairly, that the concept of a banking union was dreamt up to help avoid and depoliticise future banking crises and not to resolve the already politically toxic residue of the last crisis, currently exemplified by Greece.
Recovery and resolution is, not surprisingly, the pillar that has generated the most immediate concern among fixed-income investors, as it contains the concept of “bail-in” (the process whereby a bank’s bondholders and/or depositors, moving up through the capital structure, are progressively forced to recapitalise a failing institution). In the eyes of some people, this seems to have become equated with a notion of unfair (and possibly illegal) interference with property rights. We think this reflects a fundamentally flawed understanding of both debt generally and the nature and intent of bank resolution specifically. The risk of bail-in is default risk and as such is no different for a bank bond investor than for any other corporate bond. The point of intervention in both cases should be such that recoveries for creditors are maximised, the key argument being that that point occurs at a far earlier juncture for a bank than for a non-bank corporate. In both cases, in standard theoretical terms, investing in a bond should generate a return that is the sum of a risk-free payoff and a short position in a put option written on the firm’s assets, with the put option representing the loss given default. Resolution is simply the process by which the loss given default is established and attributed to different classes of creditors in accordance with their seniority in the liability structure. In other words, resolution is simply bankruptcy for banks!
The theory is indeed simple. The practice, however, is massively complex—and not only because of the political dimension referred to previously. The real complexity is establishing with any reliability the value of a bank’s assets. On the eve of its collapse, Lehman Brothers had a Tier 1 ratio (essentially the regulatory measure of the value of the bank’s assets) of 11.3%. On the basis of latest figures, the Greek banks have Core Equity Tier 1 ratios of between 8.7% and 12.6%, yet their senior debt is trading at approximately 40 cents on the euro. The “real” value of the Greek banks’ assets is anybody’s guess and is largely dependent on the ECB’s valuation of the bank’s holdings of Greek government bonds for collateral purposes (as the responsible monetary authority) and on the ECB’s valuation of the banks’ Greek corporate loans (in its capacity as the banks’ supervisor). This in turn relies on the willingness of the ECB to provide liquidity to the Greek economy to continue functioning with any degree of normality. It is unclear exactly how it will now play out for the Greek banks, but that is not really the point of this paper. We do not hold any Greek bank bonds, and, as noted earlier, Greece is an unresolved legacy problem rather than a useful example of the new norm. The more important question is, why, against this kind of background, do we think it is appropriate to own any bank bonds at all?
This is clearly a non-trivial question. Bank senior debt comprises 27% of the euro investment-grade corporate index by market value. When a further 6% of subordinated debt is added in, a third of the euro investment-grade universe is attributable to banks. We believe the answer to this question lies in the supervisory application of the enhanced micro-prudential regulation. As noted previously, significantly enhanced levels of capital and liquidity are being put in place, essentially creating a larger equity (and liquidity) buffer to absorb volatility in the value of the banks’ assets; it is enhanced supervision that is the key to dampening that level of volatility to an acceptable level. The stress tests that banks will now be subjected to on a regular basis are the most obvious manifestation of the regulators’ new zeal. Stress testing did not get off to the most auspicious of starts with both Dexia and SNS Bank collapsing well within the stress horizon of the 2011 EBA stress test that they had comfortably passed. Nevertheless, we do believe that methodologies are becoming more rigorous, and represent a substantial discipline upon the banks. This is especially true given that failure (or indeed near failure) will inevitably result in further regulatory intervention, and in particular restrictions on equity distributions or the requirement to raise fresh equity, which are measures that are clearly supportive of debt investors. Stress tests cannot be foolproof and are only as good as the scenarios that are modelled, as the Greek banks again exemplify.
Away from the headlines of the stress tests, however, an arguably more significant development has gained momentum. In December 2014, the EBA published its “Guidelines on common procedures and methodologies for the supervisory review and evaluation process (SREP).” This document runs over 200 pages, and the best way to describe it is as a rating methodology complete with scoring system and output on a rating scale of 1 to 4 plus “F” for an institution that is “considered to be failing or likely to fail.” In effect, it is very similar to the rating agencies’ stand-alone rating scales. The EBA’s SREP is exhaustively comprehensive, covering everything from business model analysis, strategic and financial planning and governance issues (including remuneration policies) to credit risk, market risk, liquidity risk and capital. Whereas the private sector rating agencies may spend at most a handful of days on site at even the largest banks (typically engaging with top management), the resources that the regulators are increasingly deploying are many orders of magnitude greater. Having been the beneficiaries of the so-called “light-touch” regulation in the pre-crisis years, banks are now subject to the most intrusive supervision, on an almost real-time basis, of any industry in the world.
Another big difference between the regulator-as-rating-agency and the credit rating agencies Moody’s, S&P and Fitch, is that the feedback mechanism from rating output to bank behaviour is immediate and concrete. Whereas the agency ratings are supposed to impose discipline on bank behaviour by being a partial determinant of the cost of funding, the feedback from SREP will bite directly on the banks’ capital requirement. As Nordea, the largest Nordic bank, says in its recent half-year results publication, “We also expect, as part of the SREP, that the SFSA [the Swedish regulator] will add Pillar 2 requirements that will increase the capital requirement from the current level.” And that is for a bank that already posts a Core Equity Tier 1 ratio of a very high 16%! Of course, there is one other big difference and that concerns disclosure and transparency. The SREP ratings will not be in the public domain. This is particularly significant as this element could play a decisive role in the behaviour of deeply subordinated instruments such as additional Tier 1 contingent convertible capital instruments. The SREP guidelines are quite explicit on this, with the SREP output potentially “restricting or prohibiting distributions or interest payments by the institution to shareholders, members or holders of Additional Tier 1 instruments.” To a certain extent, therefore, we are being asked to put our trust in the regulatory community, a community that arguably failed miserably in the years preceding 2008. However, given the strength of the political imperative to stabilise the financial system, and the sheer scale of the resources being committed to this end, we believe that this trust is largely merited, at least as far as the vast bulk of bank issuance that is not Tier 1 is concerned.
One of the more interesting commentators on the banking crisis has been Andrew Haldane. He has done this with the advantage of a ringside seat at the Bank of England, first as Head of Financial Stability and now as Chief Economist. Haldane is fond of using analogies from outside the financial sphere, and in particular from the realm of biological sciences. In conclusion, I therefore take a leaf out of his book to describe our approach to investing in bank debt. Our starting point is to understand the broad ecosystem in which any given bank is operating. A crucially important part of this, as set out in this paper, is clearly the regulatory framework in all its manifestations. On balance we think that, specific exceptions such as Greece aside, the banking ecosystem has become a safer place for credit investors to inhabit. However, as the Greek example clearly reminds us, banks are inseparable from their macro-economic environment—banking is the ultimate cyclical industry. So although there are measures being put in place that are intended to dampen banks’ capacity to turbocharge the economic cycle, we remain cautious about their efficacy. The second strand of our thinking is to understand the DNA of any given bank; essentially this “DNA” is its business model, but also it is the broader corporate culture that dictates both the kind of business it does and the way it does it. Just as in the natural world DNA has evolved into many different life forms, so it is the case in banking. It is our job to make sure that we are being adequately rewarded for investing in the differing DNA types, and those required rewards are likely to vary both in absolute terms and relative to each other through the economic cycle. Of course DNA is constantly evolving, and occasionally will mutate in an adverse way. This happened on a significant scale in the years preceding the crisis, and there was very little appetite to check the process—indeed there is abundant evidence that it was cheered on in the pursuit of deregulation, national champions and global reach which arguably reached its apotheosis in Royal Bank of Scotland (RBS). Briefly in October 2007, after the takeover of ABN Amro, RBS became the largest bank in the world by total assets. Twelve months later, on receipt of £46 billion of taxpayer money, it became the largest state bail-out in UK corporate history, posting a staggering £24 billion annual loss for 2008. We believe that this scale of adverse mutation is virtually inconceivable for the foreseeable future.
And finally, we need to monitor a bank’s current behavior or—in this analogy—what it had for its last meal. Given the scale and complexity of a large bank’s activities, this is actually an incredibly difficult task even with the reams of disclosure made available to the market. To a large extent, the quality of the “last meal” should be determined by the safety (or otherwise) of the ecosystem and the DNA-driven appetite, but nobody is claiming that banks can or should operate in a zero-risk environment or manner. It is in this regard that the SREP process, described earlier, should be seen as a significant constraint on bank’s capacity to load up with disproportionate, and ultimately life-threatening, amounts of risk.
Having traditionally priced tight of non-bank corporates prior to 2007/08, it is not surprising that during one of the two most significant financial crises of the last 100 years, banks have traded at a significant discount to non-banks. That discount took a second leg wider as the initial crisis exposed structural weaknesses within the eurozone in 2011/12, and again, unsurprisingly, the impact of that was greater for the euro indices than for US dollars. That spread relationship has now substantially reverted in both the major currencies, although Europe is clearly lagging the US, with the most recent blip in euro spreads attributable to the recent Greek episode. It is notable, however, that despite the political severity of recent events in Greece, the impact on bank spreads has been extremely muted in comparison to 2011/12. This is in part due to the structural firewalls that have been built into the eurozone itself, but we believe it is also attributable to the firewalls that have been built into the banking system, as outlined in this paper. With resolution/bail-in replacing tax-payer-funded bail-out, the severity of a tail-risk event has undoubtedly become substantially greater. However, more importantly, the probability of such a tail-risk event occurring has been substantially and sustainably reduced. Diligent credit analysis and stock selection can reduce exposure to that risk even further, which allows us to see real value in bank credit.