New Bank Liquidity Rules: d a e h A s r e g n a D Dangers Ahead s: e ul R y t di ui q Li k A Position Paper by EBA’s Banking Stakeholder Group n a B w e N “As the new liquidity rules in enter the implementation stage, a number of black spots must be clearly brought to attention. The aim of the Liquidity Coverage Ratio is not to enable banks to withstand liquidity pressures on their own and to survive through stressed scenarios without supervisory support. Rather, it is to buy time and make sure that supervisors can rescue ailing institutions – and possibly wind them down – without the impending threat of a disordered meltdown and its potentially unmanageable systemic costs. Accordingly, prudential rules should not be brought too far, as the wish to provide all banks in Europe with a bulletproof jacket in times of distress may, in fact, lead to imposing a straightjacket on the everyday business of financial institutions and their customers.” 0 d a he A s er g Main acronyms and abbreviations: n a D BCBS: Basel Committee on Banking Supervision s: e ul BSG: Banking Stakeholder Group R y t CET1: Common Equity Tier 1 di ui q CRD4: 4th Capital Requirements Directive Li k n CRR: Capital Requirements Regulation a B w DGS: Deposit Guarantee Scheme e N EBA: European Banking Authority ECAI: External Credit Assessment Institutions HQLA: High-Quality Liquid Assets IIF: Institute of International Finance IRB: Internal Ratings-Based LCR: Liquidity Coverage Ratio NSFR: Net Stable Funding Ratio SME: Small-Medium Enterprise 1 d a e h A s r e g n a D s: e ul R y t di ui q Li k n a B w e N 2 New Bank Liquidity Rules: Working Group on Dangers Liquidity This document was d a e Ahead drafted by a Working h A s Group on Liquidity set r e g n up by EBA’s Banking a A Position Paper by EBA’s Banking D s: Stakeholder Group and e Stakeholder Group ul R is based on y t di contributions by the ui 1 Overview and Key Issues q Li following Group k n a 1.1 Liquidity rules and the role of EBA Members: Erik B w e Credit institutions across Europe face an unprecedented Berggren, Sylvie N amount of regulatory reforms, originating from the 2009 Bourguignon, Monica De-Larosière Report and the third release of the Basel Cueva, Javier De Accord (“Basel 3”) in 2010. The new Capital Requirements Directive (CRD4) and Regulation (CRR), Andrés, Arnold Kuijpers, that the European Union is currently debating, will Christian Lajoie, Louise create a common sine qua non for institutions Lindgren, Hiltrud throughout the European Union. The CRR will establish a consistent and integrated regulatory framework for Thelen-Pischke, Andrea many aspects of bank management – including liquidity – Resti, Pamela Walkden, providing a homogeneous standard under a unified set Giles Williams. of prudential rules. In relation to liquidity, two new requirements have been proposed by Basel 3 to ensure that financial institutions are more stable and will require them to hold more 3 liquid assets and issue more long-term debt. The Liquidity Coverage Ratio (LCR) is aimed at ensuring short-term resilience of financial institutions. They will be required to hold at all times liquid assets, the total value of which equals, or is greater than, the net liquidity outflows which might be experienced under stressed conditions over a short period of time (30 days). Net cash outflows are to be computed on the basis of a number of assumptions concerning run-off and draw-down rates. The LCR will be monitored in the EU after January 2013 and the European Banking Authority (EBA) will test various eligibility criteria for liquid assets. Calibration will also be undertaken regarding net cash outflows. This fine-tuning will provide input for the level-two regulations to be introduced by the European Commission before January 2015, when the LCR will become binding for all credit institutions in the EU. The Net Stable Funding Requirement (NSFR) requires that available stable funding (equity and liability financing expected to remain stable over a one-year time horizon) at least d a equals the matching assets, i.e. illiquid assets which cannot be easily turned into cash over e h A the following 12 months. In the European Union the components of the NSFR will be s r monitored from 2013 with a view to introducing a binding requirement in 2018. e g n Da The CRR – while setting a clear and comprehensive framework for the measurement and s: control of bank liquidity – leaves many details open for calibration, impact assessment and e Rul review. As mentioned above, the EBA has been assigned a key role in the implementation of ty the new regulatory framework; it is required to provide supervisors and European di ui institutions with criteria, standards and technical advice on a wide-ranging set of issues. q Li k 1.2 The LCR: expected impact and scope for calibration n a B w As it will be phased in first, the case for calibration and careful implementation of the LCR is e N stronger. Similar attention will be required for the NSFR once a full consensus on the its structure has been achieved. Accordingly, although this report covers all liquidity rules introduced by the CRR, the LCR has been the main the focus of the contributions. Based on the latest impact studies1, the LCR shortfall of EU banks (that is, the absolute amount of extra liquid assets needed for all banks to comply with the ratio) currently exceeds 1 trillion euros. Between 2009 and 2011, this shortfall has not improved. Actually, it has deteriorated from €1tn to €1.15tn, with a 15% increase in the (almost overwhelming) amount that EU banks would need to invest in liquid assets in order to be compliant. This clear risk or threat is that European banks may channel new funding towards LCR- eligible assets rather than to loans and other “illiquid” assets. E.g., European banks could increase their liquidity buffer through additional deposits with central banks (which play no role in financing the real economy and which, in 2011, already amounted to about €850 million for large EU banks). Essentially the LCR would have the effect of crowding out productive investments and sterilize €1 trillion of liquidity out of the real European economy. In other words, unless the funding base available to European banks can quickly be increased 4 (which appears quite unlikely in the current macroeconomic scenario), the LCR might lead to €1trillion loan deleveraging process by December 2014. Such a risk would become especially acute if a narrow definition of LCR-eligible assets were enacted, which would deny adequate recognition to some financial instruments supporting the financing of companies and individuals, like corporate bonds, covered bonds or asset backed securities. While, in principle, capital markets may provide a substitute for reduced bank funding, this looks improbable given the limited development of corporate debt markets in many European countries and the high degree of risk aversion currently shown by investors. All the above provides a strong incentive for a rigorous calibration of the LCR. There are indeed, a number of steps in the computation of the ratio which could be reconsidered, in order to make it closer to market practices and to reduce the foreseeable burden for banks and the European economy. Any ratio is made up of two components. One can look at the LCR numerator, and consider d ways to enhance the set of assets eligible as liquidity buffer. By allowing banks to use, for a e h example corporate bonds and asset-backed securities as liquid assets, regulators would A s greatly support the development of those asset classes throughout Europe, and thus help er g the European capital market absorb the loans that banks will no longer be able to provide. n a D Alternatively (or, rather, jointly) one can look at the denominator and carefully revise the s: e assumptions on runoff/drawdown/rollover factors underlying the computation of the net ul R cash flows. y t di The definition of liquid assets in the LCR will affect the behaviour of market participants, ui q hence the liquidity of different asset classes. Banks will prioritise “liquid assets” as defined in Li k n LCR and “down-prioritise” other assets, which will alter the demand for different securities. a B Additionally, during a crisis banks while trying to comply with the LCR will generate liquidity w e in the first place by selling assets which are not eligible for the ratio. The definition of liquid N assets in the CRR will not just depend on the current market conditions, but rather will drive behaviours affecting the future liquidity of different security types. If such definitions were to prove inadequate, unintended consequences could build up through a snowball effect. Assumptions on cash flows (including run-off, draw-down and roll-over factors) will also have a dramatic impact on the underlying bank products, and may shift funds across business lines and different categories of bank stakeholders. E.g., limited recognition for the benefits of self-liquidating facilities (including trade finance) may increase their cost and ultimately undermine the economic viability of some lending activities. Credit provided to SMEs might become unduly expensive. Interbank lines of credit – a key tool to improve bank resilience to liquidity shocks – may prove less and less attractive due to over-conservative rules. 1.3 This report This position paper was produced by the Banking Stakeholder Group to provide the EBA and European policy makers with a technical discussion of several areas where the new rules risk 5 to have unintended effects unless properly calibrated and carefully implemented. Its structure is the following. Part 1 provides a general framework to introduce the calibrations in liquid assets and net cash flows that will be discussed in the following sections. We highlight the main implications of the new liquidity ratios for banks and for the European real economy; we then go back to the rationale of the liquidity requirements and discuss whether their anticipated costs are consistent with expected benefits. The next contribution surveys national regulations on liquidity – prior to and after the 2008-2009 financial crisis – and finds that the provisions in the CRR appear comparatively stricter than most pre-existing requirements. Finally, we discuss how the new liquidity-related ratios could modify the accounting choices of banks. Part 2 focuses on caveats and possible adjustments concerning the numerator of the LCR, that is, liquid assets that banks are allowed to use to meet their liquidity buffer. We review the eligibility criteria set out by the CRR for high quality liquid assets, highlighting why they may prove inadequate in capturing systematic liquidity risk. We then discuss the d a appropriateness of such criteria for Europe, to find that, unless appropriately calibrated, they e h A may prove a major source of disadvantage compared to the US. Subsequently, we address s r the link between liquidity ratios and possible changes in credit risk weights for government e g n debt, to conclude that a more risk-sensitive approach to sovereign risk weights could a D introduce a pronounced “cliff edge” effect into the LCR and reduce the demand by banks for s: e government debt. Finally, we look at a specific asset class, notably covered bonds, whose full ul R eligibility as a liquid asset may help incentivise portfolio diversification and keep credit y t di flowing to European consumers. ui q Li Part 3 discusses a number of potential calibrations which may be introduced in the k n computation of the LCR’s denominator, i.e., the net cash outflows experienced by a bank a B w under a 30-day distressed scenario. This includes customer deposits (where the new liquidity e N rules may unduly penalise retail and commercial banks), credit and liquidity facilities (where banks would be discouraged from holding liquidity lines with other institutions, a key tool that can be used to ease liquidity pressures) and trade finance (where the parameters of the LCR could prove detrimental for a low-risk industry that underpins global economic growth). 1.4 Time to sound the alarm As the CRR, and therefore the liquidity rules, are about to enter implementation stage, a number of hot spots must be clearly identified to stimulate further debate and highlight the risk of unintended consequences. The first issue is the definition of liquid assets in the LCR and whether this risks being too prescriptive and rigid. The Eurozone sovereign crisis has shown that liquid assets can become illiquid quickly, so flexibility is needed to accommodate different market conditions and the changing economic environment. The changing risk profile of government bonds has shown how important it is to create incentives for portfolio diversification. Another area for further consideration is the link between LCR-eligibility and central bank 6 eligibility. The CRR requires that liquid assets be central-bank eligible, but states that not all central bank collateral will be acceptable for the LCR. During a crisis, central bank eligibility is crucial in facilitating the provision of liquidity to cash-strapped institutions and markets. The definition of liquid assets and the rules on central bank collateral need to be looked at the same time – they cannot be regarded as two separate issues. Furthermore, liquidity is an elusive concept, which fluctuates over time and cannot be predicted in infinitum. Hence, supervisors should resist the temptation to draw up lists and to create parameters that will stay unchanged over time. Any “black and white” approach to liquid assets’ definition will prove increasingly unhelpful. Developing criteria that are granular enough to accommodate many different scenarios would be better to provide for a smooth transition of asset classes between different liquidity grades. Conversely, a liquidity scale which has only one or two levels is most likely to prompt cliff effects when changes occur in the perceived characteristics of specific eligible assets. Overlooking the different degrees of liquidity provided by a wide range of investable assets, could result in regulators setting the threshold too high and consequently focusing on many fewer asset classes. The wish to “err on the safe side” would ultimately lead to d a investment concentration and higher risk. he A s As concerns cash flows, no set of rules, no matter how conservative, will ever isolate a bank er g from systemic risk. E.g., assuming that all liquidity and credit lines that an institution has n a D secured on the wholesale market will suddenly become unavailable in a crisis could give a s: e false sense of security, while increasing banks’ costs and creating wrong incentives. The ul R potential for dirigisme in the new rules should not be underestimated, as minor changes in y t the factors imposed to banks (including drawdown, rollover and runoff coefficients) may di ui cause huge shifts of funds across business lines in a way which interferes with the free q Li interplay of demand and supply. k n a B It is important to remember that the aim of the LCR is not to enable banks to withstand w e liquidity pressures on their own and to survive through stressed scenarios without N supervisory support. Rather, it is to buy time and make sure that supervisors can rescue ailing institutions – and possibly wind them down – without the impending threat of a disordered meltdown and its potentially unmanageable systemic costs. There is a fundamental difference between liquidity as a micro and macro phenomenon. Many assets might well be liquid if one single bank needs to sell them, but can quickly become illiquid if all banks want to get cash out of them. The quest for assets which stay liquid “at all times” might prove frustrating, since under severe systemic scenarios liquidity can only be ensured by monetary authorities. Accordingly, prudential rules should not be carried too far, as providing banks with a bulletproof jacket in times of distress may, in fact, lead to imposing a straightjacket on the everyday business of financial institutions and their customers. Increasing compliance costs may not only make credit more expensive and undermine growth; it may also move intermediation towards shadow banking, channelling money through weakly-regulated 7 schemes which rely significantly on wholesale funding and may prove strongly pro-cyclical. The calibrations mentioned above should be carried out by the regulators and policy makers with representatives of the financial industry, users of financial services and banking scholars. The availability of reliable data sources is a major bottleneck for any effort to investigate funding and market liquidity risk; databases should be shared loyally and transparently. Any rule developed without a thorough involvement of banks and other stakeholders is bound to prove both short-lived and short-sighted. Andrea Resti d ea Head of the Working Group on Liquidity h A s Banking Stakeholder Group r e g of the European Banking Authority n a D s: e ul R y t di ui London, October 3, 2012. q Li k n a 1 For further details, see Box 1 on page 7 of this report. B w e N 8 Think again: why do the liquidity rules need further calibration Part 1 of this report provides a general framework to introduce the possible calibrations in d a e liquid assets and net cash flows that will be covered in Parts 2 and 3. To this aim: h A s r §2 (“Does the cat catch the mice? Rationale and implications of liquidity ratios”, page 10) e g n highlights the main implications of the new liquidity ratios for banks and the real economy a D in Europe. It then revisits the rationale of the new liquidity requirements and discusses s: e whether their anticipated costs are consistent with expected benefits. While a Liquidity ul R Coverage Ratio appears intrinsically correct and sound, the present version has ample room y t for recalibration, as too much emphasis on the benefits in terms of financial stability may di ui lead to overlook the costs for the economy, including SMEs. As for the NSFR, by requiring Liq banks to reduce their role in maturity transformation, it is likely to lead to higher interest nk a rates and to weaker supply of long-term finance for non-financial companies and B w households; e N §3 (“The State-of-the-Art of Bank Liquidity Rules in Europe”, page 18) provides a quick overview of national regulatory requirements on liquidity – prior to and after the 2008-2009 financial crisis – including countries from different areas of the European Union. This allows for a general preview of the reforms’ implications. with an emphasis on credit granting to private-sector corporations. The provisions in the CRR, generally speaking, appear comparatively stricter than pre-existing requirements. While bolstering liquidity buffers across the financial sector and therefore mitigating risk in distressed markets, they may have unintended implications, especially on the financing of small to medium-sized corporations; §4 (“Accounting and the liquidity ratios”, page 26) dwells on the accounting implications of the new liquidity-related ratios. The latter are likely to modify the accounting choices of banks: such rebalancing of the incentives can modify the relations of the firm with owners and other users of accounting information. 9