Banking performance, competition and financial stability: a supervisory view
Speech by Kerstin af Jochnick, Member of the Supervisory Board, of the ECB at the 7th Frankfurt Conference on Financial Market Policy entitled “European Banking – Too Much Competition?” organised by the Research Centre SAFE
Frankfurt am Main, 15 November 2019
It is a pleasure for me to be here, and I would like to thank the organisers for inviting me to this prestigious conference.
The conference proceedings today are structured around the question of whether there is too much competition in the European banking market. From a supervisory point of view, the focus after the financial crisis of 2008 has been on strengthening the banking system in Europe. A lot has been done in terms of both regulation and supervision in order to make banks more robust and better able to withstand future problems. However, although the capital and liquidity situation is better and governance in banks has improved, there are still structural problems in the European banking market. Supervisors are not in the business of promoting structural changes but we do focus on making sure that banks have viable business models. Amongst other things, this means that banks need to be profitable on a sustainable basis.
However, although profitability has improved in the aftermath of the financial crisis, the return on equity of many banks in Europe is still below the cost of capital. And while there is considerable variance across individual banks, the fact that price-to-book ratios also remain below one, on average, suggests that markets have yet to be convinced that banks will be able to earn their cost of equity anytime soon. Moreover, lifting banks’ profitability will likely become more challenging in a scenario where interest rates are “low for longer” and economic growth prospects are more sluggish than previously expected.
Having established that banks are likely to face sizeable profitability challenges for some time to come, we need to tackle at least four questions in order to inform public policy.
First, to what extent will a more protracted period of low banking profitability than previously envisaged affect financial stability? Second, to what extent is low banking profitability a by-product of competitive market structures? Third, is there a potential trade-off between profitability and competition in terms of their impact on financial stability? Fourth, what could various stakeholders in the banking system do to address these challenges?
Unfortunately, there are no straightforward answers to these questions, for a variety of reasons. The very low interest rate environment is a relatively recent development, so it would be premature to pass firm judgement on its likely effect on banking structures. The number of factors potentially affecting banking profitability makes it hard to gauge the precise role played by competitive forces. And while the economic literature provides many insights concerning the nature of the relationship between bank profitability and financial stability, on the one hand, and bank competition and financial stability, on the other, the evidence remains mixed on both counts. I will return to these points later on, but I take the fact that we have a full day of discussions ahead of us as proof in itself that the debate on these issues remains very much alive in both academic and policy circles.
Notwithstanding these caveats, I would like to share with you some thoughts on these questions from a supervisory perspective.
In essence, I will be arguing that although the overall resilience of the banking system has greatly improved in recent years, weak profitability will remain a cause for concern for financial stability because it hampers banks’ ability to build buffers against negative shocks. And it may also negatively affect banks’ ability to support the real economy. Mergers have been put forward as one of the ways for banks to shore up profitability through cost efficiencies, and there is broad consensus among both policymakers and industry officials that the European banking market is too large at present.
However, if banking consolidation is to gather momentum, a number of legal and regulatory issues need to be tackled to encourage sustainable Europe-wide players to develop and ensure that non-viable banks can exit the market without disruption. In addition, the banking union needs to be completed. Unless these things happen, the promise of a truly integrated and competitive banking market which benefits millions of Europeans will remain elusive.
Let me briefly examine these points in turn, starting with the issue of banking sector resilience.
Improving the resilience of the banking system
The shift in global banking standards in the aftermath of the crisis
In the aftermath of the global financial crisis, the G20 and other multilateral bodies embarked on a wide-ranging set of co-ordinated reforms of supervisory and regulatory frameworks to bridge the gaps which had been exposed by the crisis. The underlying rationale of the reforms for the banking sector was to move from a system based on one indicator (the capital ratio) to a broader approach where one indicator would guard against one risk (e.g. the capital ratio to address solvency risk, or the leverage ratio to address excessive leverage risk). The bulk of measures were focused on two areas.
The first set of measures was devoted to building more resilient financial institutions, in light of the fact that banks’ capital and liquidity buffers had proven to be insufficient during the crisis. The Basel Committee on Banking Supervision (BCBS) thus did a major overhaul of global banking standards focusing on significantly increasing the quality and quantity of capital to be held by banks, as well as increasing their available liquidity buffers at both short- and medium-term horizons.
A constraint on banks’ overall leverage was also added in order to complement risk-based capital requirements. Under the revised standards, banks using internal models to calculate capital requirements must hold at least 72.5% of the capital which would be held under the (typically more conservative) standardised approach. This measure, which is designed to reduce unwarranted variability in banks’ risk-weights, is known as the ‘output floor’.
This set of measures, most of which were announced in 2010, became known as the ‘Basel III’ reforms. As part of the work on building more resilient financial institutions, the Financial Stability Board (FSB) issued principles for remuneration practices with a view to ensuring that banks’ compensation schemes reflect prudent risk-taking and capital planning, rather than the perverse incentives for ‘excessive’ risk taking over short-term horizons which were seen to have contributed to the crisis.
In Europe, most of these measures have been progressively implemented since 2014 through a number of amendments to the Capital Requirements Directive as well as through a new Capital Requirements Regulation. These are two of the fundamental pieces of legislation regulating banking activity in the EU. However, the Basel III output floor, which was only agreed in 2017, has yet to be implemented in EU law.
Overall, Basel III is helping to make banks safer and sounder. However, some bankers claim that the pendulum has swung too far the other way, arguing that tighter rules are weighing on their profits and making it harder for them to finance the economy. I do not agree with this view. Those banks that have been faster in adjusting to the Basel standards are generally in a better position to support their customers.
The second set of post-crisis reforms has focused on addressing the “too big to fail” problem, which meant that taxpayers often had to rescue failing banks on account of systemic concerns or significant interconnectedness that would have put both the financial system and the real economy at risk. Governments provided the banking sector with substantial financial assistance during the crisis, with the European Commission estimating that a state aid totalling €1.5 trillion was granted to recapitalise EU banks between 2008 and 2017 (or around 10% of EU GDP in 2017, of which just under half was ultimately used). The reforms implemented at both the global and the European level, have sought to redress this by shifting the burden of bank rescues from taxpayers to banks’ shareholders and creditors.
Among the efforts that were made to make banks “resolvable”, I would highlight two.
First, the FSB adopted the “Key Attributes of Effective Resolution Regimes for Financial Institutions” in 2011, setting international standards for resolution regimes while seeking to avoid systemic disruptions and taxpayer losses. In Europe, the FSB standards have been implemented through the Bank Recovery and Resolution Directive (BRRD, adopted in 2014). The BRRD requires resolution authorities to be established in all EU countries, providing them with tools to deal with failing banks, while accounting for cross-border issues through group resolution plans and cooperation arrangements with third parties.
Second, the FSB standards established the principle of “bail-in” to ensure that losses and recapitalisation needs resulting from resolution are borne by banks’ shareholders and creditors rather than taxpayers. This led to the establishment at the global level of the total loss-absorbing capacity standard for global systemically important banks, which requires such banks to hold a given share of liabilities that could be “bailed in” to cover the costs of resolution. In Europe, the BRRD implements the bail-in provisos foreseen by the FSB, while the minimum requirement for own funds and eligible liabilities aims to ensure that sufficient funds are readily available to absorb potential losses and recapitalise the bank in the event of resolution.
At the global level, the overhaul of the bank regulatory framework which I have just described has played a major role in improving the resilience of the banking system in the aftermath of the crisis. But in Europe, another development has also contributed decisively to this trend, and that is the banking union, which I will turn to now.
Banking union under the first pillar
The financial crisis of 2008 was a global phenomenon, but it was particularly vicious and virulent in Europe, showing that financial sector risks can spread to the broader economy with devastating repercussions. As mentioned earlier, European governments intervened to support troubled banks during the sovereign debt phase of the crisis, thereby increasing the pressure on public finances. In some cases, the causality worked in reverse, with a country’s weak fiscal standing undermining confidence in the banking sector.
The mutual dependency between banks and sovereign finances gave rise to a vicious circle (known as “the doom loop”) which ultimately threatened to undermine the entire construct of European Monetary Union itself. An integral part of the problem was that responsibility for supervisory and crisis management policies rested at the national level, because “national” responses to the crisis resulted in increasing financial fragmentation.
In order to overcome these problems, political leaders decided in 2012 to create a banking union by transferring the pursuit of bank stability to the European level. The three pillars of the banking union – a Single Supervisory Mechanism (SSM) with the ECB at its core, underpinned by a single rulebook, a Single Resolution Mechanism and a common deposit insurance scheme – were thus designed to work together so as to restore public confidence in the banking system.
On the supervisory front, ECB Banking Supervision sought to do this in a number of ways.
The first order of business was to raise the capital bar for banks in order to dispel the notion that their balance sheets were not credible on account of hidden losses. Five years on from the operational start of the single supervisor, we can see notable progress on this front: from the fourth quarter of 2014 to the second quarter of 2019, the weighted average of banks’ CET1 ratio improved by 3.1 percentage points to 14.1%. Beyond capitalisation, banks’ balance sheets have also become stronger in other areas such as liquidity and funding, with the average Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR) remaining above the 100% minimum threshold since the inception of the SSM, and banks’ weighted average NSFR increasing by 10.7 percentage points to 112.6% over the same period.
A second line of action was to reduce credit risks in the system, especially those posed by the high level of non-performing loans (NPLs) on banks’ balance sheets. We took a phased approach, initially issuing a guidance to banks on how to tackle NPLs (in 2017); subsequently publishing an addendum to this guidance setting out supervisory expectations for the provisioning of new NPLs (in 2018); and finally announcing further steps in the supervisory approach for addressing the stock of NPLs (also in 2018), seeking to address this as part of the supervisory dialogue through bank-specific supervisory expectations aimed at achieving adequate provisioning of legacy NPLs. These measures helped to put NPLs on a steady downward path, with the average NPL ratio dropping by 4.3 percentage points to 3.6% from the fourth quarter of 2014 to the second quarter of 2019. The stock of NPLs almost halved over the same period, from around €1 trillion to €562 billion.
A third priority has been to improve the overall quality of supervision to foster convergence, harmonisation, transparency and even-handedness across the banking system. We undertook many initiatives in this domain, for example as regards the harmonisation of options and discretions granted to the national supervisors by the European legislator. All national supervisors participating in the SSM agreed to apply these options and discretions in line with the ECB guide released in 2016.
We also published a guide on our fit and proper assessment criteria, applied when assessing the suitability of banks’ managers and administrators, so candidates for these positions know what to expect as part of the normal vetting process. We attach great importance to these assessments because our experience suggests that bank weaknesses can often be traced back to malfunctioning governing bodies. And we are also about to finalise our Targeted Review of Internal Models project, which seeks to reduce unwarranted variability in the calculation of own funds requirements and thus complements the ‘output floor’ under Basel III that I mentioned earlier. Apart from raising the bar for banks using internal models, this project reflects our commitment to ensuring consistency and fairness across banks.
Shoring-up bank profitability
I hope to have convinced you by now that the banking system is in better shape than before, and that this is partly thanks to the initiatives which the ECB has undertaken as the single supervisor. The system’s greater ability to deal with shocks is good news in itself – especially because there is now ample evidence that banking crises can lead to large and persistent output losses, especially in advanced economies.
However, I also noted in my introduction that the profitability of the euro area banking sector has remained subdued overall in recent years, and that the future macroeconomic environment is more likely to reinforce rather than to revert this trend. How much of a concern is this for financial stability?
One common way of tackling the relationship between bank profitability and financial stability in economic literature is through the implications for bank risk-taking. Many of the post-crisis discussions in policy circles emphasised that banks had increased their risk-taking at a time when profits were high. So it is perhaps ironic that policymakers now face questions about exactly the same kind of bank behaviour materialising for the opposite reason of low profitability. This apparent contradiction is also captured in the literature, with some studies suggesting that bank risk-taking is reduced in a context of higher profitability (as it also indicates higher long-expected profitability or higher “charter value”) while other works argue that risk-taking could increase when profits are high (for example because may be more lax). In this context, there is also evidence to suggest that high profits in good times could be indicative of systemic tail risk in bad times.
In order to shed light on banks’ risk-taking behaviour, ECB Banking Supervision – as one of its supervisory priorities for 2020 – is currently conducting a systematic analysis of banks’ underwriting criteria, with a focus on new lending. It is examining the quality of banks’ lending practices and lending standards with a view to mitigating potential risks.
Nonetheless, it is clear that, all else being equal, a protracted period of low profitability will make it harder for banks to build up buffers against negative shocks through organic growth. Banks therefore need to continue to adapt both their business models and their operations to the low interest rate environment in order to buttress their financial bottom line.
As regards cost efficiency, euro area banks as a group still seem to have some catching up to do in comparison to banking peers in other jurisdictions which are facing, or have recently faced, a similar macroeconomic and interest rate environment. Although the situation of individual banks varies considerably across the euro area, the average cost-to-income for euro area banks is around 66%, as compared with 55% to 57% for Nordic and US banks. All in all, improvements in cost efficiency should benefit bank profitability, as some research also suggests. In addition, banks could also augment their revenues by making more use of sources of income that are not dependent on net interest margins, such as fees and commissions. Here too, some evidence suggests that diversification is positively associated with profitability.
Investing in digitalisation could also be an integral part of banks’ successful strategies to shore-up profitability. This is likely to entail a sizeable sunk cost in the near term, but the payoff should accrue in the medium-term (through declining cost ratios). However, this should not be interpreted as a universal recipe for success, as banks can also expect competition from non-financial firms in the provisions of some banking services which will be “unbundled” from those traditionally offered by banks. Experience with banks’ successful IT strategies suggests that local conditions also matter, including the general digitalisation of society at large to begin with.
In addition to these factors, mergers and acquisitions are usually mentioned as a way for banks to improve profitability through economies of scale. I will next discuss some of the aspects surrounding mergers and acquisitions, including competition and financial stability.
Fostering consolidation and orderly market exit
Profitability, overbanking and financial stability
There is a broad consensus among policymakers that the European banking sector is oversized. For example, the Chair of the ECB’s Supervisory Board, Andrea Enria, recently noted that while “the optimal size of the banking sector is hard to gauge, it seems clear […] that the European banking sector is still too large”  meaning that consolidation needs to take place. A number of structural indicators for the banking system that are regularly compiled by the ECB, such as market concentration shares, population per credit institution, population per branch, and assets per bank employee, support the view that there is room for consolidation in some markets and jurisdictions. This view is also supported by the general observation that, on an unconsolidated basis, the number of credit institutions in the euro area has been generally declining since the crisis.
The perception that the system is overbanked is also shared by bankers themselves. In a recent dialogue between the ECB and the banking industry, overcapacity was recognised as a challenge and mergers and acquisitions were seen as the best way to overcome this. In addition, a number of public officials have pointed to overbanking as a contributing factor to banks’ low profitability, a conclusion which is supported by an ESRB report on the topic.
If excess capacity is seen as a factor behind banks’ lacklustre profitability and mergers are proposed as a (partial) remedy, the next question we should ask is whether a reduced number of players in the system would have any implications for financial stability. To answer this question, we would need to assume that overcapacity is closely associated with strong competition in the market.
Under this premise, economic theory under the “competition-fragility” view suggests that more competition would lead to an erosion of market power and decreased proﬁt margins, and thereby encourages banks to take risks. However, a “competition-stability” view would instead imply that more intense competition (and hence, lower market power) leads to lower interest rates for borrowers, thus reducing borrower default and asset portfolio risk. This would suggest that banks become riskier as markets become more concentrated . Some studies note that the two theories may be actually reconciled because, even if market power in the loan market leads to riskier loan portfolios, banks’ overall risks can be kept in check if they protect their franchise values through risk-mitigating measures (such as increased capital).
My overall impression, which other policymakers also seem to share, is that the relationship between competition and financial stability remains inconclusive in the literature.
Why cross-border mergers have not happened
While the findings of the literature remain ambiguous, our practical policy experience with banking mergers and acquisitions in recent years is clear: they have taken place mostly within domestic boundaries, and rather infrequently across borders. This runs counter to the widely-held expectation that the establishment of the SSM would result in an increase in cross-border bank and financial integration.
So why has this not been the case? We often hear the banking industry say that concerns on banks’ asset quality and regulatory uncertainty have been discouraging factors. However, the ECB’s efforts to tackle NPLs have shed light on banks’ balance sheets and put banks' NPL ratios on a firm downward path. And given that the regulatory reform process is now coming to an end, I would also expect this concern to dissipate going forward.
However, I would like to mention another discouraging element which will unfortunately remain with us for the foreseeable future, legal fragmentation. The current legal framework for single supervision in Europe is a three-tiered system that includes rules directly applicable to banks (such as the CRR); provisions set by European directives (such as the BRRD) that are not directly applicable and need to be transposed into national law; and provisions that are purely national in nature. This set-up leads to national discrepancies which create an uneven playing field across countries in various areas of prudential relevance, such as the granting and withdrawal of bank licenses, and the fit and proper assessment of bank managers.
For cross-border mergers and acquisitions, Member States still have access to national options and discretions that can prevent liquidity from being allocated freely within banking groups across different countries. This is a problem. And this occurs against a backdrop where cross-border capital waivers are not possible under the currently applicable regulation, and where national insolvency regimes are also different. With this in mind, it is perhaps unsurprising that banks have generally been reluctant to venture across national boundaries.
So legislators, regulators and supervisors need to continue to work together to eliminate obstacles to cross-border bank integration. This should help banks diversify their portfolios across borders, which should make them more resilient. It should also weaken the sovereign-bank nexus and improve private risk-sharing, which should improve the stability of the system as a whole.
At the ECB, we are now looking at what else we can do to facilitate cross-border consolidation for banks that want to pursue this option. This includes making further use of intragroup financial support agreements as part of banks’ recovery plans and reflecting on which revisions to options and national discretions could be supportive of such a process. However, I would like to highlight that although we would welcome cross-border mergers as a sign of an integrating banking market, we do not favour one form of consolidation (e.g. cross-border) over another (e.g. domestic). We assess all mergers and acquisitions proposals on their own merits, purely on technical grounds, and recognise that the consolidation process needs to be driven by the market.
Another aspect related to the bank consolidation process where harmonisation is also needed is in the legal frameworks governing the process by which non-viable, weaker banks exit the market. In the past, some banks whose futures were questionable were simply rescued by their governments for lack of a better alternative, and thus perhaps out of fear of contagion. From this point of view, the establishment of the Single Resolution Board (SRB) and Single Resolution Fund (SRF) is very welcome, as is the fact that we now have a dedicated framework to deal with failing banks (i.e. the BRRD and the Single Resolution Mechanism Regulation).
However, of the handful of banks which we have declared ‘failing or likely to fail’ since the inception of the SSM, only one has been resolved by the SRB – the rest have been liquidated under national insolvency regimes. As with other domains where national law applies, we have seen that these insolvency regimes vary from country to country, so harmonisation is needed here as well in order to have coherent outcomes in both resolution and liquidation across the union. This harmonisation will at best be gradual, however, so adding a bank liquidation tool to the European resolution toolkit would also help to ensure that the degree of consolidation needed in the banking sector as a whole takes place in an orderly fashion.
Cementing the European financial architecture
Before concluding, let me briefly touch on something which is important for every aspect of the relationship between bank performance, competition and financial stability which I have discussed so far. This is the need to complete the banking union and make further progress towards a genuine capital markets union.
I am sure that I don’t need to say much about this, because, after all, I am speaking at the Center for Sustainable Architecture for Finance in Europe (SAFE). It should be obvious to everyone that a banking union which is not complete can hardly be deemed sustainable in the medium-to-long term. Work on the second pillar – single resolution – is advancing, with the SRB making efforts to improve banks’ resolvability through its resolution plans and the SRF being progressively built. Agreement on a ‘backstop’ to the SRF has also been reached, but this still needs to be operationalised to enhance the credibility of European resolution.
However, progress is most needed on the third pillar of the banking union – a common deposit insurance scheme – which is entirely missing.
This pillar is critical on many fronts. It is key to foster cross-border consolidation among banks, because it will create mutual trust among participating Member States, thereby reducing their perceived need to ring-fence capital and liquidity. With a common deposit insurance scheme in place, it should be easier to reach agreement on eliminating the remaining provisions in the current regulatory framework that are trapping capital and liquidity within national boundaries. Therefore, progress on this pillar is also critical for banking competition and performance in an integrated system where banks can provide services on an equal footing across the entire market. And it will also be important for financial stability, because a common deposit insurance scheme would remove the major element still remaining in the bank-sovereign nexus in line with the founding aims of the banking union. We would therefore welcome any progress on this front.
My conclusions can be brief. If there is one overriding message which I would like to convey to you today is that we have to continue working together to deepen the integration of European markets to ensure that the single rule book and the Banking Union work effectively and as envisaged. Bank consolidation would be facilitated as a result, including across borders, and allow weak banks to exit the market without disruptions. If this doesn’t happen, the necessary process of consolidating the banking system is likely to be more costly and more protracted in time than would otherwise be the case.
However, there is one big unknown in this regard, which is how digital innovation and new information technology will shape this process. My impression is that the entry of new providers of financial services will probably be positive for competition in the banking market but at the same time give rise to new banking and financial stability challenges which supervisors and central banks should carefully monitor.