Vítor Constâncio: Challenges faced by the European banking sector
Speech by Mr Vítor Constâncio, Vice-President of the European Central Bank, at the Risk & Supervision 2017 Conference, organised by the Italian Banking Association (Associazione Bancaria Italiana), Rome, 14 June 2017.
Let me start by thanking the Italian Banking Association for inviting me to participate in this important event.
Euro area banks have faced a number of headwinds since the financial crisis. As a result, euro area banks have raised significant amounts of capital which have led to a substantial strengthening of solvency ratios: average common equity tier 1 capital (CET1) increased from 7% in 2007 to 14% at present. The solvency position of European banks is thus robust. The lingering concerns about their condition relate primarily to their low profitability, an issue that has already lasted several years.
The subdued profitability is reflected in depressed bank valuations. Despite a broad-based improvement in bank valuations in recent quarters, a wide dispersion persists between euro area and U.S. banks' valuations. While euro area banks' price-to-book ratios have recovered from the lows of mid-2016, the gap between euro area and U.S. banks' valuations remains significant, with an average price-to-book ratio of 0.77 against 1.26 for U.S. listed banks.
The large cross-sectional variation of price-to-book ratios within the euro area partly reflects cyclical factors, as the pace of economic recovery varies across countries, but possibly also differences in the progress made by institutions in tackling structural challenges. To give you an idea, average price-to-book-value is 0.94 for Spain, 0.75 for Italy, 0.76 for France and 0.42 for Germany.
In the euro area, listed banks aggregate return on equity (ROE) stood below 3% in 2016, a slight decline from the previous year. In international comparison, euro area banks' financial performance continues to lag behind that of most of their global peers: in 2016, U.S. average ROE stood at 8% with Nordic banks above 9%, led by Swedish banks with 12% ROE, a notable performance in a country where key official interest rates are negative and much lower than in the euro area. Regarding cross-country differences, listed Italian and German banks recorded negative average ROE in 2016 which, in the case of Italy was due to sharp increases in loan impairment charges at some banks, mainly linked with increased efforts to clean up their balance sheets. In other large euro area countries, French and Spanish listed banks recorded a ROE of 5% and 7%, respectively.
The continued weakness in bank profitability reflects a number of cyclical and structural factors. A key cyclical challenge is linked to the difficulties in increasing revenues in a low nominal growth and low interest rate environment and a relatively flat yield curve. As a matter of fact, net interest income declined somewhat in 2016, as the compression of margins was only partly offset by increased credit growth albeit still at moderate levels.
While profitability headwinds stemming from cyclical factors should abate as the economic recovery progresses, structural challenges remain and need to be tackled. These include the large stock of non-performing loans (NPLs), a legacy of the 2008 financial crisis, cost inefficiency and excess capacity. Let me address each of these factors in turn.
A large stock of legacy NPL in some euro area countries continues to dampen profitability prospects. The average NPL ratio for euro area banks at just above 2% in 2007, reached 8% in 2013 and stands now at 6%. For the six "high-NPL" national banking sectors, the averages went however from 5% in 2007 to 23% last year.1 Elevated loan impairment costs remain an important driver of low profitability in "high-NPL" countries. Indeed, following the capital and provision increases since the financial crisis, the NPL issue does not primarily concern solvency but rather banks' profitability. The coverage of NPLs by provisions and collateral is, on average, 82% in the euro area and 80% in Italy. Net NPLs are therefore reassuring, from a solvency perspective. Profitability however, is affected by the lower returns provided by the NPLs, given the weight of gross exposures in total assets. The share of total assets that generates lower revenues than full performing loans is a considerable European problem: gross NPLs represent 4% of total assets of euro area banks against only 0.8 % for U.S. banks. The figure for Italy however stands at 11%, thereby with an unavoidable significant impact on profitability, despite the visible reduction in the NPL ratio for the system since the peak in 2013. Naturally, high NPLs also tie up capital, erode funding, as well as operational capacity, thereby constraining banks' ability to support the economic recovery. Despite a visible reduction in system-wide NPL ratio in Italy, progress in reducing stocks of high NPLs to manageable levels remains insufficient.
Amid continued difficulties in boosting revenues, remaining cost-inefficiencies also weigh on banks' profitability. On aggregate, euro area banks' cost-efficiency has deteriorated somewhat since 2010, based on both a cost-to-income and a cost-to-assets basis and compares unfavourably with some international peers, most notably the Nordic countries.
Against this background, for many euro area banks, a return to sustainable profitability is increasingly dependent on improvements in operational efficiency. A cross-country comparison suggests that the relative importance of physical versus digital distribution channels may be one of the differentiating factors across countries in terms of cost efficiency, as illustrated by the positive correlation between branch network reduction since the late 90s and the usage of internet banking in EU countries.
Structural challenges to profitability in some banking sectors are also linked to industry structure and excess capacity. In addition to banks' efforts to improve operational efficiency via cost-cutting, consolidation could bring some profitability benefits at the sector level. These could be particularly relevant in countries where banking systems remain fragmented and are characterised by low market concentration and high cost-to-income ratios. Ideally, consolidation should go hand-in-hand with greater geographical diversification. This would allow banks to achieve economies of scope and scale from cross-border mergers and acquisitions, thereby also contributing to greater macroeconomic risk-sharing by diversifying country risks.
The way forward
Looking forward, there are several ways of estimating what could be "realistic", longer-term ROE targets for banks. In terms of historical comparison, we need to bear in mind that high leverage observed in pre-crisis years can flatter measured profitability. Therefore, an upper bound estimate for ROE targets could be derived by multiplying pre-crisis return on assets (ROA) ratios by the current, structurally lower leverage multiples. From another perspective, the long-term average of banks' estimated cost of equity (i.e. the rate of return investors expect from an investment in bank equity) is a useful benchmark. In addition, survey-based information on banks' own estimates of long-term sustainable ROE can complement approximations of what could be sustainable levels of profitability. Combining estimates from these different approaches, we arrive at a target range of 8-10% ROE for euro area banks.
Should we expect banks' profitability to reach their "sustainable" targets over the next 3-5 years? Even in a benign scenario of a robust economic recovery, this may not necessarily be the case. For instance, in its latest Global Financial Stability Report, the IMF argues that even in a cyclical upturn, many euro area banks may not be able to raise their profitability to a sufficient degree.2
Against this backdrop of cyclical and structural challenges to banks' profitability, possible solutions include: (i) revenue growth; (ii) cost efficiency improvements; (iii) consolidation through mergers and acquisitions; and (iv) NPL resolution.
The first avenue that banks may want to explore is to revisit their revenue growth strategies, particularly in the field of fees and commissions that still show relatively low levels in international comparisons.
Adjustments to the business model could also include actively targeting a change in the product mix, seeking adjustments to the funding structure or increasing the geographic diversification of activities.
These discussed strategies have limitations, as they require capital, entail risks or on account of competitive pressures. From a financial stability perspective, one should also be mindful of the possible impact of revenue diversification on bank performance and riskiness.3
As highlighted earlier, many euro area banks continue to operate with relatively high cost levels, which constitute a drag on efficiency and hamper the sector's ability to generate sufficient net income. Improving cost efficiency is therefore another strategy some banks should pursue in order to return to more sustainable levels of profitability.
Cost reductions and efficiency improvements can be achieved in a variety of ways, ranging from more traditional measures such as downsizing, branch closures to the adoption of new, cost-saving technologies aiming at digitalising financial intermediation services such as increasing the reliance on internet-based banking. There is significant room to improve cost efficiency among euro area banks considering that the ratio of costs to total assets is, on average, 1.4% for euro area banks, and just half of that, 0.7%, for Swedish banks.
Apart from potentially reducing costs, a higher reliance on digitalised forms of financial services may also result in more contestable bank retail markets, as it becomes easier for bank customers to "shop around" and compare bank products and prices.4 This, in turn, may have a positive impact on the sector's overall efficiency.
It should be acknowledged though that cutting costs is easier said than done. It may entail substantial upfront investments and costs (e.g. new IT systems, compensation for staff reductions), may be hampered by country-specific legal and institutional settings and may be surrounded by non-negligible execution risks. The more capital-constrained banks especially may therefore face difficulties undertaking radical cost reduction schemes.
Another avenue towards improving efficiencies and eliminating redundancies is through banking sector consolidation via mergers and acquisitions (M&A). Well-managed M&As can deliver important cost synergies (e.g. lower administrative expenses, branch rationalisation) as well as revenue synergies (e.g. lower funding costs of the merged unit) that might contribute to improving the banking sector's profitability prospects.
M&A activity has, however, remained subdued in recent years, in particular cross-border M&As. In the ECB's latest Financial Integration Report, we document that M&A activity in the euro area banking sector has been on a declining trend since 2000, and in particular in the aftermath of the crisis. This is especially true as regards cross-border mergers.5
In general, there is an ambiguous relationship between consolidation and financial stability.6 Banking sector consolidation could strengthen the market power of the merged entities which may help them restore profitability.
At the same time, by lowering competition, consolidation may increase the costs of retail financial services for households and firms7 and may also provide disincentives to keep costs under control. These negative effects are likely to be more pronounced in the case of domestic M&As, whereas cross-border mergers should be broadly neutral in terms of immediate impact on market concentration.8
In this context, let me reflect on the Italian perspective. The Italian banking sector is one example among euro area countries where there seems to be scope for further consolidation. Indeed, in terms of standard concentration measures the Italian banking sector is among the least concentrated in Europe.9
This has been duly recognised by the Italian authorities, which in recent years have introduced a number of measures aimed at consolidating and modernising the Italian banking sector, such as the reform of the mutual banks10, the reform of the cooperative banks11, and the self-reform of the banking foundations.12
These are important steps to facilitate further consolidation of the sector which are welcomed in Italy and elsewhere in the euro area.
Excess capacity and fragmentation along national lines are, to some extent, hampering the profitability and performance of some euro area banking sectors.13The banking union, including single supervision and resolution mechanisms, in principle provides ideal conditions for banks to reap the benefits of new cross-border merger and acquisition opportunities.
However, progress in both domestic and, in particular, cross-border bank consolidation remains limited to date. Ultimately, the euro area economy needs banks that are large and efficient enough to operate and diversify risks on a cross-border basis within a European single market, but small enough to be resolved with the resources of the Single Resolution Fund. This would make the most of the banking union while improving the trade-off between financial stability and economic efficiency.
One potential deterrent of further consolidation at the domestic and euro area level is the high NPL stock that I just mentioned, which may reduce the attractiveness of potential targets in some jurisdictions. This is another reason why resolving the NPL problem is of utmost importance.
Resolving the NPL problem
Resolution of the NPL problem is first and foremost a task for the banks. Should NPL markets be fully efficient, it would be straightforward to sell the NPLs to other investors. We know however, that the NPL markets in the EU are shallow. Notwithstanding the recent pick-up in transactions, only a small fraction - less than 10% - of the stock of EU NPLs changed hands over the last twelve months.14
Notwithstanding the existing obstacles, a wide range of solutions to the NPL problem is available to banks and policy-makers.
Internal work-out would always feature among the available tools. In this vein, ECB Banking Supervision has recently issued guidance to improve bank capabilities in working out NPLs, as these vary significantly. Direct sales could be complemented by securitisations, asset management companies, and NPL trading platforms that may be used to transfer the risk of NPL work-out outside of the banking sector to those who may be better suited to recover value.
Clearly, more needs to be done to fully unlock the potential of market-based solutions to the NPL issues. Due to asymmetric information between the investor and the seller of NPLs, and to very long duration and high cost of debt enforcement procedures, there is a wide bid-ask spread in the NPL markets. The exact data on that gap are difficult to come by. Estimates suggest that it might reach up to 40% for a fully collateralised loan.15
Reduction of that gap, which has many causes, requires a comprehensive strategy. Determined structural reforms should play a prominent role in the action plan for national and European authorities, regardless of the actual solutions adopted by individual countries.
For instance, debt enforcement procedures should be streamlined, and their cost lowered. This may require in particular that judicial and out-of-court capacity is increased. To mitigate the information asymmetry, access to financial information about distressed debtors and public data registers should be made easier. Investment in improving quality of that information is often needed, and may be rewarded by investors with higher transaction prices.
Within the comprehensive strategy, there may be a role for national governments to support the correction of the NPL market failure. Necessary structural reforms may require a lot of time to yield fruit and to earn credit with investors. This is a very tangible issue also in Italy, where recent reforms of the debt enforcement claims have not yet produced the expected results, and therefore have not influenced NPL prices to the desired extent.
Asset management companies (AMCs) may aid in correcting the market failure. They can swiftly clean up NPLs from bank balance sheets, and resolve them over a longer period of time. Acquisition of assets at their long-term economic value, instead of market value which is depressed by low liquidity and high uncertainty, minimises fire sale losses. Sweden, Germany, Ireland, Spain, Slovenia and Korea, for example, used these tools to manage their banking crises, often with a focus on loans backed by real estate. There is one common feature in this type of AMC: state support. By putting capital and funding guarantees at stake, governments can signal their commitment to the structural reforms and bring forward the related benefits. A similar role may be played by securitisation schemes.16
The EU legal framework does not close the door to setting up an AMC with government sponsorship. The rules also allow for precautionary recapitalisation of banks by the public sector in cases of significant financial stability concerns, subject to strict conditions laid down in the BRRD (Art. 32, 4), covering hypothetical losses estimated under an adverse scenario of a stress test. Precautionary recapitalisations also have to respect State Aid rules that generally require bail-in of subordinated debt as stated in the European Commission's communication of July 2013, where possible exemptions are nevertheless foreseen (paragraph 45).
Let me remark here that, in the steady state, this would not be a major issue as "bail-in-able" debt would in principle only be sold to knowledgeable, qualifying investors, capable of understanding the related risks. That unfortunately, is now not the case: our resolution rules entered into force before the requirements for MREL i.e. the specification of "bail-in-able" securities, entered into force. However, the rules give scope to address valid concerns about retail investors through compensation. The European Commission recently announced what could become an important initiative: a proposal regarding a comprehensive strategy to deal with NPLs in Europe, including a blueprint for national AMCs accompanied by a clarification of the regulatory aspects involved.
The new European Regulation on recovery and resolution deserves full support as it represents a change from easy public bailouts to a new culture of private bail-in, minimizing moral hazard. Notwithstanding this, the regulation does not ignore financial stability considerations. "We have to bear in mind that it is not only direct public support for banks that has a cost for taxpayers, but also financial instability - indeed, the costs of the latter may, in some circumstances, be higher. Compare the worldwide costs for taxpayers stemming from the absence of public intervention to rescue Lehman Brothers, with the zero cost for taxpayers following the USD 700 billion injection into U.S. banks in 2008 (which have been totally repaid by the banks). In other words, financial instability can have a meaningful cost to taxpayers even if it is not visible in the very short-term - a notion that all policy-makers should keep in mind.17
I have highlighted a number of structural, cyclical and legacy-related challenges confronting the euro area banking sector. We need to recognise the progresses made in increasing robustness of euro area banks, the establishment of the banking union, the firming of the economic recovery due to the ECB monetary policy, with the consequent improvement in banks' future profitability as recognised by the stock market in the recent quarters. Much still needs to be done to fully overcome the consequences of the financial crisis and fully restore the condition of all national banking sectors.
The first line of response obviously lies with the banks that have to adjust their business model to the new environment, entailing new regulation and increased competition from non-bank financial institutions and market-based finance as well as from technological innovations in financial services. The adjustment is already underway and I have no doubt that the banking industry, even after some restructuring and downsizing, will continue to play a dominant role in the financing of the European economy.
Public authorities, regulators as well as supervisors, also have to do their part by providing the proper incentives to steer and support a strong and efficient banking sector that the European economy crucially needs.
Thank you for your attention.
1 See Constâncio, V. (2017), "Resolving Europe's NPL burden: challenges and benefits" at Bruegel Conference, February. NPL ratio is defined as NPLs as a share of total loans.
2 See IMF, GFSR April 2017.
3 The literature provides mixed evidence on the benefits of revenue diversification; see e.g. Altunbas, Y., S. Manganelli, and D. Marques-Ibáñez (2011), "Bank Risk During the Financial Crisis - Do Business Models Matter?", ECB Working Paper Series No. 1394 ; Demirgüç-Kunt, A. and H. Huizinga (2010) , "Bank Activity and Funding Strategies: The Impact on Risk and Returns", Journal of Financial Economics, Vol. 98, pp. 626-650; Köhler, M. (2015 ), "Which Banks are More Risky? The Impact of Business Models on Bank Stability", Journal of Financial Stability, Vol. 16, pp. 195-212; DeYoung, R. and K. P. Roland, (2001), "Product Mix and Earnings Volatility at Commercial Banks: Evidence from a Degree of Total Leverage Model", Journal of Financial Intermediation, Vol. 10, pp. 54-84; and Stiroh, K. J, (2004), "Diversification in Banking: Is Noninterest Income the Answer?", Journal of Money, Credit and Banking, Vol. 36(5), pp. 853-882.
4 See Gropp, R. and C. Kok (2017), "Competition and contestability in bank retail markets", forthcoming in Handbook on Competition in Banking and Finance (eds. J. Bikker and L. Spierdijk), Edward Elgar Publishing.
5 See ECB Financial Integration in Europe, May 2017. The report also provides an in-depth review of the literature studying the costs and benefits of bank consolidation.
6 For references to the literature, see ECB Financial Integration in Europe, May 2017.
7 See e.g. Corvoisier, S. and R. Gropp, (2002), "Bank concentration and retail interest rates", Journal of Banking and Finance, Vol. 26, pp. 2155-2189; Focarelli, D. and F. Panetta, (2003), "Are mergers beneficial to consumers? Evidence from the market for bank deposits", American Economic Review, Vol. 93, pp. 1152-1171.
8 See ECB Financial Integration in Europe, May 2017.
9 For instance, the share of the 5 largest credit institutions in total banking sector assets was just 43% in Italy compared to an average of 63% across the euro area countries (based on 2016 data). Similarly, the Herfindahl index (another market concentration measure) for Italy was 0.045 in 2016 compared to an average across the euro area countries of 0.12. Only Germany, Luxembourg and Austria have a less concentrated banking sector than Italy.
10 Through various modifications of the Consolidated Act of Banking and Credit Laws (Legislative Decree No. 385 of 1 September 1993).
11 As part of the package of measures included in the Decree-law No. 18 of 14 February 2016.
12 As reflected in the protocol of intent signed on 22 April 2015 by the Minister of the Economy and Finance and by the Chairman of the Association of Foundations and Savings Banks ("ACRI").
13 In fact, there is some empirical evidence that euro area banks operating in less-concentrated markets tended to be less profitable in the period between 1991 and 2013; see special feature by Kok, C., C. Móré, and C. Pancaro, "Bank profitability challenges in euro area banks: the role of cyclical and structural factors", in ECB Financial Stability Review, May 2015.
14 The total volume of transactions in the loan portfolio market - including also non-distressed portfolios - in Europe amounted to about €103 billion in 2016, which is less than 10% of the EU NPL stock. See Deloitte (2017), "Uncovering opportunities in 2017. Deleveraging Europe 2016-2017".
15 Fell, J., M. Grodzicki, R. Martin and E. O'Brien (2016), "Addressing market failures in the resolution of non-performing loans in the euro area", Special Feature B in ECB Financial Stability Review, November 2016.
16 See Fell, J., C. Moldovan and E. O'Brien (2017), "Resolving non-performing loans: a role for securitisation and other financial structures?", Special Feature C in ECB Financial Stability Review, May 2017.
17 Constâncio, V. (2016), "Challenges for the European banking industry", speech at the Conference on "European Banking Industry: what's next?" , organised by the University of Navarra, Madrid, 7 July 2016.