European banking supervision three years on
Welcome remarks by Mario Draghi, President of the ECB, at the second ECB Forum on Banking Supervision, Frankfurt am Main, 7 November 2017
It is my pleasure to welcome you to this second ECB Forum on Banking Supervision.
When I spoke at the Forum in 2015, European banking supervision had only been in operation for one year. Much had been achieved in that time – not least the comprehensive assessment of bank balance sheets – but in many ways the single supervisor was still untested.
We saw European banking supervision as having two main objectives: to reduce bank-specific risks through tough and forward-looking supervision; and to reduce country-specific risks in the banking sector by applying those same high standards across the whole of the euro area.
Now, three years on, we can begin to take stock of what has been accomplished.
What is clear is that European supervision has been instrumental in building a stronger and more resilient banking sector. The country in which a bank is located has also become a less important factor in how its credit risk is perceived.
These two achievements have been a crucial complement for our monetary policy, too, since banks are the main channel of financial intermediation in the euro area.
A well-integrated financial sector with sound banks has helped transmit our policy impulses more evenly across the euro area. And it has allowed us to pursue an accommodative policy for as long as necessary, without building up significant financial stability risks.
Progress with European banking supervision
There is no doubt that building European supervision has been a remarkable undertaking. Today we have 900 supervisory staff working at the ECB who, together with 4,700 national supervisors, directly oversee around €22 trillion in assets, representing around 200% of euro area GDP.
But more important than its scale have been the changes the single supervisor has prompted in the conduct of supervision. It has broken with the past in a single, but fundamental way.
That is: it has brought about a more uniform approach in how banks are supervised, leading to a more resilient banking sector overall. The key catalyst for this change – alongside the new EU regulations – has been the harmonisation of the Supervisory Review and Evaluation Process (SREP).
This harmonisation has allowed supervisors to converge towards common benchmarks in how they assess risks; and it has helped them to be consistent in how their risk assessments are then linked to supervisory capital add-ons and other measures.
To illustrate the difference this has made, in 2014 the correlation between SREP scores and Pillar 2 capital requirements was 26% in the euro area. In 2016, it was 76%.
European supervision has therefore resulted in a substantial strengthening of shock-absorbing capacity within the sector. The total capital ratio of banks supervised by the ECB has increased by more than 170 basis points since early 2015. The quality of capital has gone up as well: the high-loss absorbing component – CET1 – now makes up the largest share of total capital of euro area banks.
Specific weaknesses are also now being addressed in their entirety across the euro area. Currently the most important issue here is tackling non-performing loans (NPLs).
We all know the damage that persistently high levels of NPLs can do to banks’ health and credit growth. Internal ECB analysis shows that, over recent years, banks with high stocks of NPLs have consistently lent less than banks with better credit quality, therefore providing less support to firms and households.
And though NPL levels have been coming down for significant institutions – from around 7.5% in early 2015 to 5.5% now – the problem is not yet solved. Many banks still lack the ability to absorb large losses, as their ratio of bad loans to capital and provisions remains high.
We therefore need a joint effort by banks, supervisors, regulators and national authorities to address this issue in an orderly manner, first and foremost by creating an environment where NPLs can be effectively managed and efficiently disposed of.
Importantly, the development of European supervision has not only reduced the risk of individual banks failing. It has also – as we hoped – had some success in reducing the importance of location in perceptions of bank risk, because single, rigorous supervision is an essential precondition for the other pillars of the banking union that more decisively sever the bank-sovereign nexus.
Indeed, looking at the largest banks for which we have data available, the correlation between bank credit default swaps and those of sovereigns is now considerably weaker than at the height of the euro area crisis.
Still, there is no room for complacency, since these improvements are likely to have been driven, in part, by the improved economic situation. It is therefore crucial that further reforms to de-link banks from sovereigns do not lose steam, notably completing the other pillars of banking union.
The benefits of strong supervision for monetary policy
All these supervisory efforts have not only produced a more robust banking sector; they have also provided crucial support for our monetary policy since we entered a new easing phase in mid-2014. This support has come from two main sources.
First, stronger supervision has improved the transmission of our policy impulses through banks.
There is now plenty of evidence to suggest that the level of bank capital is a key variable in how banks lend during downturns, and hence also for monetary policy transmission. It is therefore no surprise that the improved health of the banking sector, coupled with our credit easing measures, has coincided with a marked improvement in the transmission process.
Previous asymmetries in bank lending rates across the euro area have now largely been reversed, the cost of bank borrowing has stabilised at record lows everywhere in the euro area, and so has its dispersion across countries. We can now say that the pass-through from our past policy measures to lending rates is nearing completion.
The decline in lending rates has been particularly noticeable for small loans in vulnerable countries, which are a proxy for the financing conditions faced by SMEs. SMEs are of course highly dependent on well-functioning banks as their options are limited when it comes to accessing market finance. For these small loans, the spread between vulnerable and more resilient countries has now narrowed to a record low of 5 basis points.
This more even credit pricing across countries and firms has, in turn, been crucial to the broadening of the economic and employment recovery, not least because SMEs represent 60% of euro area value added and employ 70% of the labour force.
The second way in which stronger supervision has supported our monetary policy is by helping contain any financial stability risks that may emerge during a long period of low rates.
One channel through which such risks can appear is search-for-yield effects: low rates can, in principle, induce banks into making lower-quality loans, leading to higher loan losses. But with a strong supervisor ensuring well-capitalised banks, the quality of lending tends to be higher.
This is confirmed by a major study looking at loan-level data in Spain, which finds that, when overnight rates fall, highly capitalised banks grant fewer loan applications to risky firms than lowly capitalised banks, and have fewer loan defaults.
With a more resilient banking sector in the euro area, we have seen this more positive picture develop. Credit risk exposures in banks’ loan books have declined as monetary policy has eased. Default rates have fallen, and forward-looking measures also suggest a decline in credit risk.
This has of course been driven by improvements in credit quality as the macroeconomic situation has improved. But it may also reflect the role of higher capital in resolving agency problems: the more a bank is capitalised, the more its owners stand to lose if borrowers default and cause losses. So the more equity a bank holds, the greater its incentive to make higher-quality loans.
Other financial stability issues associated with low rates have also not materialised, thanks in part to the stronger supervisory framework.
At euro area level, we currently see no signs of credit-fuelled housing bubbles, which are at the root of most serious financial crises. Since 2016, bank lending for house purchases has risen, on average, by 2.9% per year – well below the growth rates of up to 12% recorded in the run-up to the crisis. Some local pockets of risk have emerged, but both supervisors and macroprudential authorities are actively taking steps to counter them.
We have also seen little evidence that negative interest rates are undermining bank profitability, an issue which has caused a lot of concern. This would pose a financial stability risk to the extent that it hinders banks from building up capital through retained earnings and makes raising market equity too expensive. It would also affect monetary transmission for the same reasons.
In fact, net interest income has remained quite stable over the past two years, even as overnight rates have drifted lower. And thanks to gains in other income components, banks’ return on equity has been rising and is converging towards their cost of equity. For the banks under ECB supervision, return on equity has risen from 4.4% at the end of 2015 to 7.1% at the start of this year.
This neutral impact of negative rates is largely due to the general equilibrium effects of monetary policy that we have explained many times: when policy is accommodative, the main components of profitability largely offset each other, since the positive impact of a stronger economy on loan-loss provisions largely cancels out any negative effect on net interest income.
For some banks, however, these negative effects may be larger than for others. This is where strong supervision is again crucial. As part of its SREP, ECB Banking Supervision carries out detailed, comparative assessments of banks’ business models, which feed into the ongoing supervisory dialogue between the supervisory teams and banks.
This process is not prescriptive, but it helps bring to light important issues such as the sustainability of banks’ business models in a low rate environment, and their operating costs in comparison with their peers, which in some countries are a contributing factor to low profitability.
Let me conclude.
We are now three years into the life of European banking supervision, and the track record so far is encouraging. Though the single supervisor is still a young and developing institution, it has in many ways lived up to the high expectations that accompanied its founding.
Rigorous and uniform supervision has led to higher levels of capital and a more resilient sector overall. The credit risk of banks is now less determined by the credit risk of their country of establishment.
Healthier banks have, in turn, helped transmit the ECB’s accommodative monetary policy more evenly across the euro area, leading to a stronger and broader recovery. And the new supervisory framework has helped mitigate any financial stability risks that might have arisen as a result.
In short, European supervision and European monetary policy have proven to complement each other well. It is an approach which confirms the synergies that can be reaped when the right policies are combined at euro area level.
 See, for example: Buch, C. M. and Prieto, E. (2014), “Do Better Capitalized Banks Lend Less? Long-Run Panel Evidence from Germany”, International Finance, Vol. 17 (1), pp. 1-23; Kapan, T. and Minoiu, C. (2013), “Balance sheet strength and bank lending during the global financial crisis”, Discussion Papers, No 33, Deutsche Bundesbank; and Osborne, M., Fuertes, A.M. and Milne, A. (2017), “In good times and in bad: Bank capital ratios and lending rates”, International Review of Financial Analysis, Vol. 51, May, pp. 102-112.
 See Jiménez, G., Ongena, S., Peydró, J.-L. and Saurina, J. (2014), “Hazardous Times for Monetary Policy: What Do Twenty-Three Million Bank Loans Say About the Effects of Monetary Policy on Credit Risk-Taking?”, Econometrica, Vol. 82(2), March, pp. 463-505.
 See Altavilla C., Boucinha, M. and Peydró, J.-L. (2017), “Monetary policy and bank profitability in a low interest rate environment”, Working Paper Series, No. 2105, ECB, October.