Monetary policy and the euro area banking system
Speech by Peter Praet, Member of the Executive Board of the ECB,
at ECMI Annual Conference, Brussels, 9 November 2016
The euro area banking system today is stronger than when it entered the financial crisis. At aggregate level, risk-weighted capital ratios have steadily improved – the median Tier 1 ratio of euro area banks almost doubled since 2008 – while banks have reduced assets and raised equity, leading to a marked decline in leverage. Resilience has been further buttressed by a reduced relative reliance on wholesale funding – which proved particularly run-prone in the early stages of the financial crisis – in favour of a greater share of deposits in their liabilities.
But building resilience has been a long and hard-earned process, bolstered by public support – in the form of capital support and guarantees – at its peak amounting to as much as one and a half trillion euro in gross terms. And though the sector is now stronger and more stable, it is also facing major challenges in terms of earnings capacity and profitability.
The return on equity of significant euro area banking groups still remains well below their cost of capital. Market valuations point to an expectation of subdued earnings prospects going forward. And further signs of weakness have emerged in response to the different episodes of financial market turbulence since the start of 2016, which saw euro area banks’ price-to-book ratios particularly hard-hit.
This troubling picture can be explained by four interlinked challenges.
The first is the legacy challenge bequeathed by the euro area’s twin crisis, in which the financial crisis morphed into a sovereign debt crisis and a double-dip recession. This has left a legacy of persistently high levels of non-performing loans (NPLs) in parts of the euro area, which drive down bank profitability through loan-loss provisioning charges. NPLs also hamper the contribution of the banking system to the recovery in parts of the euro area economy, which in turn leads to weaker macroeconomic outcomes and still slower balance sheet repair. Question marks over how and when these NPLs will be resolved do little to improve the situation.
Second is the cyclical challenge posed by the low growth and low inflation environment, which goes along with lower levels of policy interest rates. So far, low rates have not had a marked effect on bank profitability, with net income actually increasing between 2014 and 2015. But with interest rates expected to remain “low for long” in the euro area, market analysts expect a significant reduction in (already low) returns on equity over the next five years, especially for banks whose business models are strongly reliant on maturity transformation. The relatively subdued growth outlook is also expected to weigh on credit demand and hence income growth from loan volumes.
Structural challenges are a third factor in weak profitability. Though the efficiency of the euro area banking sector has improved since the crisis, with median cost-to-income ratios declining by almost 10 percentage points, operating costs remain high in some national banking systems, especially those which are characterised by overcapacity. Banks with high fixed costs are also facing increasing competitive pressures from non-banks and new FinTech entrants. While there has been some industry consolidation in response, it has largely taken place within countries, meaning higher efficiency has come at the cost of increased national concentration.
Further rationalisation of branches and consolidation of entities appears necessary. But in a truly integrated euro area banking sector, what I would ideally like to see – as a medium-term goal – is such consolidation to go hand-in-hand with greater geographical diversification. Which is to say: banks achieving economies of scope and scale from cross-border mergers and acquisitions, while also deepening macroeconomic risk-sharing by diversifying country risks. This is, in my view, the litmus test of whether we have completed a genuine banking union.
Fourth are regulatory challenges. The increased resilience of the euro area banking sector is in large part down to the regulatory agenda since the crisis, but it is hard to deny that this agenda has also created some uncertainties, for instance over steady state capital levels, which are reflected in bank share prices. Regulatory constraints have had a direct impact on earnings capacity too, especially in the investment banking space. Revenue generation from fee, commission and trading income has fallen significantly as a percentage of assets since the crisis. This is a welcome and intended consequence of restrictions on excessive risk-taking, but comes at a time when interest income is also being compressed.
The slump in bank profitability is not, per se, a concern for policymakers. The banking sector needs to adjust to a changing market environment just like any other economic sector. But – as I will explain in today’s speech – the struggles of banks are not innocuous from a monetary policy perspective given their integral role for monetary policy transmission and macroeconomic performance more generally. And the newfound financial resilience of the euro area banking system can only be sustained if it is accompanied by improved financial performance in the future.
So the situation needs close attention and the risks must be carefully monitored. But in my view, the path towards a more healthy and sustainable sector involves a clear division of responsibilities.
Monetary policy will continue its accommodative stance until inflation returns to our aim, which will remove some of the cyclical drivers of weak profitability linked to disappointing nominal growth. This will require interest rates to remain low for an extended period of time. As such, it is urgent that, in parallel, steps be taken to address the other challenges depressing profitability: tackling legacy NPLs, responding to structural change, providing regulatory clarity. Such a comprehensive approach means both private actors and other policymakers must play their part.
Let me address these points in greater detail.
Monetary policy effectiveness and the state of the domestic banking system
In a predominantly bank-intermediated financial system like the euro area’s, monetary policy-makers have an inherent interest in the position and prospects of the domestic banking system. This interest has three dimensions: the efficiency of monetary policy transmission; economic shock-absorption; and longer-term macroeconomic performance.
Efficiency of transmission relates to the capacity of banks to pass on monetary impulses to the real economy, which is an essential pre-requisite for central banks to influence cyclical conditions and ensure price stability over a medium-term horizon. Bank-enabled economic shock-absorption relates to the capacity of banks to accommodate temporary financing needs of firms, households, and the public sector, thereby helping them follow smoother spending and investment profiles and, in turn, mitigating the macroeconomic fluctuations that follow from a given shock. Finally, the contribution of banks to longer-term macroeconomic performance stems from their capacity to allocate capital to the economic sectors deploying it most productively and to facilitate savings and investment decisions by bridging the gap between current and future spending decisions.
In other words: a healthy banking system makes it easier for central banks to respond to a given shock; it reduces the overall impact that this shock would exert on the economy, even absent a monetary response; and it improves the steady state to which the economy gravitates when the shock has dissipated.
These critical interactions between monetary policy and the banking system were already well-understood prior to the crisis. But, during the crisis, their relevance became palpable as the banking system failed as a conduit of monetary policy impulses, transformed from a shock-absorber into a shock-propagator, and – bogged down by necessary but painful balance sheet repair efforts – has not made a decisive contribution to the recovery.
This crucial role of the banking system explains why many of our monetary policy interventions during the crisis were aimed at repairing the bank lending channel. Those interventions can be divided into three phases.
Bank-based monetary policy transmission during the crisis
The first phase, which started in the summer of 2007 and escalated after the collapse of Lehman Brothers a year later, consisted in a systemic liquidity run at global level triggered by lack of transparency on the amount and distribution of risks related to certain types of privately issued securities. The resultant seizing-up of the interbank market spilt-over to the entire financial system and forced banks and other institutions into a fire-sale of assets so as to stay current on their short-term liabilities.
The ECB confronted this situation by elastically accommodating the liquidity needs of the banking sector and by lengthening the duration of its refinancing operations. Thus, we avoided that the liquidity shock morphed into a self-fulfilling solvency crisis, which would have induced a severe monetary contraction when recessionary forces instead called for forceful monetary expansion.
The second phase of the crisis, which occurred in 2011-2012, had a specific euro area connotation, but followed a similar script as the first – only that the shift in market perceptions, this time, did not pertain to privately issued paper but to the debt instruments of certain euro area governments. The risks originating from this situation and the corresponding challenges to monetary policy transmission were no less severe than those from the first phase of the crisis. In fact, the consequence for the broader economy was, again, an acute failure of monetary policy transmission, with countries most in need of further monetary policy accommodation instead facing deteriorating financing conditions.
And, again, the ECB responded: we adopted OMT, a commitment to stamp out unwarranted break-up risk premia from the price of the securities issued by stressed jurisdictions conditional on economic adjustment.
But the confidence crisis had already left a harmful heritage in transmission, which heralded a third phase of crisis: banks in a vast portion of the euro area had lost their willingness and capacity to keep credit flowing to the real economy. Borrowing conditions were being tightened precisely at the time when the economy was entering a new slump. This required further flexibility in our policy framework, articulated in three main instruments: a series of targeted long-term refinancing operations (TLTROs), a negative interest rate policy, and an asset purchase programme including private and public securities.
The TLTROs offer banks long-term refinancing conditional on their using the funds to expand lending. The asset purchase program, together with a negative rate applied on excess reserves, quickens the pace at which banks and other lenders are returning to economic risk-taking. Lower term spreads on public securities encourage a shift in the composition of banks' portfolios toward other types of exposures with a higher risk-adjusted return. Finally, the tax on electronic reserves discourages liquidity hoarding and speeds up the process of asset reallocation.
As a result of this series of measures, we are now seeing the bank-based transmission process working more smoothly. Banks have responded to the combination of measures by slashing their lending rates and shifting out their loan supply schedule. Lending volumes are picking up gradually. And lending rates are easing considerably: between June 2014 and July 2016, the composite cost-of-borrowing indicator for new corporate loans fell by 95 basis points in the euro area and up to 150 basis points in stressed countries.
The role played by monetary policy in supporting bank-based transmission has been confirmed by recent academic research. In particular, studies suggest that the elastic provision of liquidity by the ECB earlier in the crisis – e.g. via the longer-term refinancing operations that started in December 2011 – was instrumental in counteracting the freezing of bank funding markets and in curtailing the risk of a credit crunch.
The evidence also suggests that the shift in our policy strategy towards a more active support of bank lending has been successful. The combination of targeted longer-term refinancing operations and large-scale asset purchases has been found to have induced an increase in bank lending throughout the euro area, as well as a simultaneous reduction in the exposure of banks in stressed countries to their domestic sovereigns. This latter effect was not observable for the earlier, non-targeted longer-term refinancing operations that mainly served to provide banks with funding reassurance in distressed market conditions.
At the same time, research indicates that the beneficial effects of the APP on lending volumes have been accompanied by downward pressure on lending rates, which has further reinforced the pass-through to the real economy.
Accordingly, there are increasing signs that our monetary policy has contributed decisively to the ongoing cyclical recovery. And there is no doubt that a robust cyclical recovery benefits all sectors of the economy.
But a question that has arisen in some quarters is whether, at some point, the extraordinary steps we have taken to sustain bank-based transmission may have negative side effects. Certainly, in the initial phases of the crisis, the side effects were positive from the perspective of banks, as they were partly shielded from the consequences of often misguided and imprudent business models they had adopted prior to the crisis.
In recent times, however, several observers have instead emphasised another type of side effect of our measures, namely its negative impact on bank profitability. In fact, several market participants have laid much of the blame for the prevailing bank profitability concerns on monetary policy, in particular stressing the negative effects of low interest rates on net interest margins. But this perspective is partial.
The interaction of monetary policy and bank profitability
Net interest margins have indeed declined in recent years in the wake of low policy interest rates. But monetary policy also affects bank profitability through various other channels. Lower interest income has been partly offset by our measures leading to reductions in bank funding costs, higher asset valuations, and a more robust recovery– which in turn creates further demand for bank credit and improves the capacity of borrowers to service their debts. Accounting for these factors, we do not see strong evidence that bank profitability, on aggregate, has been suffering from our measures at this stage.
However, the longer the current low-interest rate environment persists, the greater the challenges for bank profitability will be. This is because many of the benefits of monetary accommodation for bank balance sheets, such as the concomitant capital gains, are unlikely to occur on a continuous basis, whereas the drawbacks are likely to remain in place – for instance due to the downward stickiness of interest rates on retail deposits, which implies a squeeze in net interest margins if lending rates and yields were to remain at their current low levels for long. And the underperformance of bank equities may actually be testament to such concerns.
Of course, weak bank equity markets are not, in the first instance, a public policy concern. But they may become such concern if they persist and morph into a systemic phenomenon. The reason is that the newfound financial resilience of the euro area banking system can only be sustained if it is accompanied by improved financial performance in the future. Otherwise, banks will struggle to generate capital organically or raise equity in capital markets, which is the cushion that allows banks to weather adverse shocks, retain or even increase their exposure to the real economy, respond to potential regulatory changes, and attract future investors.
And, besides the dampening effects on their own financial resilience, this scenario would put a dent into the contribution that banks can make to the economic recovery: deteriorating earnings prospects weigh on bank equities, which in turn increases the cost banks face when raising equity in the market. While this increase in the cost of equity may initially be compensated by the substitution of equity with debt funding, the scope for such substitution is limited by regulatory requirements and banks’ own motives to establish adequate capital buffers. Hence, a persistent decline in equity valuations will ultimately render credit more costly and less profitable for banks which, in response, may curtail lending to the real economy.
This adverse link from low interest rates to weak bank profitability to lower loan supply is particularly problematic in view of evidence suggesting that the low interest rate environment is likely to persist for an extended period of time. According to this evidence, declining potential growth together with a relative scarcity of safe assets to absorb global savings have led to a declining natural interest rate – that is: the rate consistent with full resource utilisation and steady inflation rates around the central bank objective over a longer-term perspective.
In the context of such decline of the natural rate and the continued cyclical weakness, the necessary degree of monetary policy accommodation implies a constellation of very low interest rates for an extended period of time. Therefore, the ECB will preserve its accommodative stance until inflation returns to our aim, which will remove some of the cyclical drivers of weak profitability linked to disappointing nominal growth. It will be crucial to continue monitoring the implications of our monetary policy measures on the position and prospects of the banking system and approach any future re-calibration of our monetary policy toolkit, should this be necessary at all, in a way that optimises the stimulus for the broader economy without undermining the incentives for financial intermediaries to remain actively engaged in transmission.
At the same time, additional adjustment needs should be addressed decisively to ensure a healthy euro area banking sector: tackling legacy NPLs, responding to structural change, providing regulatory clarity. These relate both to individual banks and to the institutional environment.
Adjustment needs for individual banks
A number of banks are now tackling their legacy and structural challenges rather than waiting for better times. But many still need to adjust faster to the new conditions they face. The process of disposing of NPLs is moving too slowly in some jurisdictions. And cost efficiency has to adapt to the new macroeconomic and regulatory environment. As noted, euro area banks have made important progress in reducing costs since the start of the crisis. But there is clearly scope for further improvements: cost-to-income ratios, for example, have remained high for euro area banks compared to their global peers, and progress in efficiency improvements has levelled off.
Opportunities for efficiency gains are there. The technological challenge posed by FinTech, for example, is also an opening for established market participants to push out the efficient frontier, thereby protecting market shares and raising new sources of revenue in the customer-to-bank relationship, and/or increasing efficiency and cutting costs on infrastructure. There is also scope for both further rationalisation within banks, especially those burdened with expensive branch structures, and further consolidation among credit institutions, in particular in countries where fragmentation leads to unsound competitive behaviour and low operational efficiency.
Indeed, even though the number of branches in the euro area declined by 16% between 2008 and 2014, almost half of the decrease in the number of local bank units was accounted for by Spain. Countries undergoing structural adjustment programmes have also seen the largest relative decrease in the number of credit institutions. Banking systems in some larger countries have also adjusted but remain at relatively low concentration levels, implying scope for rationalisation and consolidation without reinforcing the “too-big-to-fail” problem.
To navigate the new environment, banks’ business models will need to adjust as well. As a consequence of both market pressures and regulation, many banks have shifted from revenue generation based on investment banking, high leverage and/or reliance on relatively cheap wholesale funding, towards business models built around retail banking, lower leverage and increased reliance on deposit funding. These are welcome developments from the perspective of financial stability – but they may also increase vulnerability to a very low interest rate environment.
This calls for stronger efforts to increase income diversification, which is currently widespread only across larger banks. Funding costs, especially in terms of deposits, also remain high in comparison with global peers, squeezing net interest income. But admittedly, neither more stringent pricing policies nor greater focus on fee-generating activities will likely compensate for the downward pressure on profitability from other factors. And after many years of encouraging banks towards more “plain vanilla” business models, it may be unfair to now expect them to reverse course.
That is why, if the banking sector is to adjust effectively, public authorities also need to play their part – indeed, it is their responsibility to create a sound and supportive environment for such adjustment to take place. To my mind there are three priority areas for action.
Reforms at the institutional level
The first is to stabilise the regulatory environment. As underlined in the communiqués of the recent G20 meetings, to remove one source of uncertainty, the post-crisis regulatory framework needs to be finalised in a timely, full and consistent manner – most pressingly through the completion of the Basel III framework by end-2016, without further significantly increasing overall capital requirements across the banking sector. Those forums have also pointed to the need for enhanced monitoring of the implementation of reforms and their effects, which is important to ensure consistency with macroeconomic policy objectives and to address any material unintended consequences.
The second is to improve further the institutional environment for NPL resolution. This hinges on both supervisory pressure towards individual banks, using qualitative and quantitative measures, and on system-wide reforms to reduce the cost for banks and investors of getting NPLs off bank balance sheets. Most important in this context are measures to speed up asset recovery, such as improving the efficiency of judicial and insolvency procedures, and appropriate tax incentives.
The third priority is to complete banking union. If we are to reap the economies of scale of a genuine single banking market, we need a euro area institutional framework that allows banks to thrive as euro area banks. So the main responsibility for policymakers is not to stop halfway.
Indeed, in crucial ways, banking union remains incomplete. A case in point is the fact that the euro area is not treated as a single jurisdiction for the purposes of bank regulation. There is still non-negligible national discretion in implementing rules. Regulatory uncertainty remains on how fungible excess liquidity and capital is across borders. And the lack of single-jurisdiction status may impose – albeit limited – additional capital charges on euro area banks. Symptomatic is the treatment of additional capital charges for systemically important banks related to cross-border euro area exposures, which are still considered as international exposures from a regulatory perspective.
This is just one example for a bigger point: if we are to achieve a fully integrated euro area banking sector and to reverse the retrenchment behind national borders that took place during the crisis, we need to remove any remnants of the regulatory framework that implicitly support home bias, raise the impediments to cross-border business and disadvantage euro area banks internationally. Though some banks have had cautionary experiences with cross-border expansions in the past, the development of pan-European banks, alongside strong local banks serving local markets, must be part of the financial landscape of monetary union.
In a monetary union the banking landscape cannot be made up of a collection of standalone national banking systems. And in an environment where bank profitability is weak, and where macroeconomic stabilisation policies are already at full throttle, the benefits of such cross-border integration – efficiency and risk-sharing – are even more in demand. Thus a true banking union needs to be completed in a reasonable period of time. And that includes, as the ECB has noted on many occasions, establishing a European Deposit Insurance Scheme that can ensure the fungibility of insured bank money across all parts of the monetary union.
Let me conclude.
A resilient banking system is crucial to provide for a smooth transmission of monetary policy in response to cyclical fluctuations and to support economic performance over a longer-term horizon. And, indeed, since the onset of crisis, euro area banks have made important progress in strengthening their financial resilience.
At the same time, this newfound resilience will come under pressure if the current slump in bank profitability were to persist going forward.
The prevailing low interest rate environment is one factor that dampens profitability prospects. But, since the ongoing recovery remains strongly reliant on continued monetary accommodation in the context of a declining natural rate, the constellation of very low interest rates will probably prevail for an extended period of time.
Against this background, the ECB will continue monitoring the implications of our monetary policy measures on the position and prospects of the banking system. The analytical underpinnings of the ECB’s strategy, which have granted a prominent role to the bank-based transmission mechanism in the context of our monetary pillar, are well-suited for this task.
At the same time, many other factors, relating to legacy and structural problems, are holding down bank profitability. Hence a durable improvement in the prospects of the euro area banking system requires further efforts outside the realm of monetary policy. These include, notably, the swift and thorough completion of banking union. Additional adjustment in the banking sector towards greater cost efficiency and business models that remain viable in the new regulatory and macroeconomic environment is also needed.
For some parts of the banking system, these adjustment needs have arisen because of regulation discouraging unsustainable pre-crisis business models that were highly reliant on aggressive leverage and maturity transformation, regulatory arbitrage (in particular via off-balance sheet activities), and a lack of transparency and due diligence in the origination of financial products. In others, the adjustment needs reflect structural challenges and over-capacity. In both cases revenue sources will have to be augmented by more sustainable business models that are consistent with the new policy environment.
Beyond the banking sector, policy needs to pave the way for a more dynamic recovery and – even more importantly – a more favourable end-point of that recovery. For the crisis has not just led to an underutilisation of economic capacity, but has also reduced the pace at which economic capacity itself expands – its potential growth. Pre-crisis, the potential growth rate in the euro area, on average, amounted to almost 2% per annum; after the crisis, this growth rate roughly halved.
The implications for economic prosperity are striking: an economy growing at 2% more than doubles over a 40-year period; an economy growing at 1% expands by less than half over the same horizon. Meanwhile a strong economic basis is essential to cope with other challenges facing the euro area over the longer-term. These include the burden of high legacy debt, large implicit liabilities related to unfunded pension claims, and intensifying distributional concerns in the context of a widening income gap.
We are therefore facing a mixed outlook. The cyclical recovery is proceeding at a moderate pace and, despite continued downside risks, has proven resilient to a range of adverse external shocks. The ECB’s accommodative monetary policy has been instrumental in sustaining the recovery and the gradual absorption of economic slack. To secure a return of inflation rates to levels below, but close to, 2% over the medium term, it will remain necessary to preserve a very substantial amount of monetary support.
Yet, at the same time, there are many open questions as to how to strengthen the structural recovery – that is, how to raise the growth path on which the economy will settle once it returns to full capacity. Here, monetary policy has limited influence and actions have to come from other policy areas. Thus far the actions taken have not been commensurate with the challenges the euro area is facing. For instance, economic policy has done too little to support the recovery in investment. In fact, the ratio of public investment to GDP has followed a consistent downward trend in recent years; so, instead of counteracting the crisis-related downturn in private investment, it has added to it. Hence a more growth-friendly composition of fiscal policy remains crucial, coupled with structural reforms to improve the business climate and re-energise the private investment recovery.
I would like to thank Fédéric Holm-Hadulla for his contributions to this speech and Nadya Jahn for valuable inputs. This text, in large parts, draws on the speech held in Madrid on 4 October 2016. It contains additional detail on bank-based monetary policy transmission, inter alia reflecting new research presented at the ECB conference on "Monetary policy pass-through and credit markets" on 27-28 October 2016 (see section on bank-based monetary policy transmission during the crisis) and on the role of other policy domains (see Conclusion).