Basel III & beyond: revised Liquidity Coverage Ratio (LCR)

Published



Background to the changes


The key changes


High Quality Liquid Assets


Cash outflows and run-off rates


Cash inflows


Application of the metric


Monitoring tools


PwC preliminary assessment

 

Introduction

On 7 January 2013 the Basel Committee on Banking Supervision (BCBS) issued a finalised standard on the Liquidity Coverage Ratio (LCR). The LCR forms one of the key planks of the Basel III reform package. The package consists of (i) a revised standard on minimum levels of capital (largely defined, but with further changes expected such as the proposed reform of the trading book regime and finalisation of the leverage ratio); (ii) the LCR; and (iii) the Net Stable Funding Ratio (NSFR – a measure designed to restrain the amount of short-term wholesale borrowing and stimulate stable funding over a one year time horizon).

The package announced on 7 January concerns only the LCR, the short-term liquidity measure. The complementary longer term ratio, the NSFR, is not covered by the recent announcement but the original proposed standard on both the LCR and NSFR envisaged that the latter would be subject to review and, if necessary further calibration, by mid-2017. The Committee notes that the NSFR will be a key area of its focus over the next two years.

This note summarises the key changes to the LCR, and provides an initial assessment of the impact of these changes.

go to the top

Background to the changes

The original proposed standard was issued in December 2010. It was always envisaged that the BCBS would review and, if necessary, recalibrate the LCR, and the timetable for revisions to be announced by mid-2013 has been kept with the latest announcement.

The BCBS says that it has ‘been mindful of the implications of the standard for financial markets, credit extension and economic growth, and of introducing the LCR at a time of ongoing strains in some banking systems’ – a reference no doubt to the ongoing problems in the Eurozone, in particular.

The BCBS has therefore decided to delay the full implementation of the measure, and although the initial start date of 1 January 2015 has been maintained, there is a transitional period until 1 January 2019, during which the minimum level of liquidity coverage will gradually rise from 60% of the minimum standard to the full 100%, at increments of 10 percentage-points a year.

The standard also states that countries that are receiving financial support for macroeconomic and structural reform purposes may choose a different implementation schedule, recognising that these countries still have a long way to go before their banking systems may have recovered sufficiently to be able to comply with the Basel III package.

It should also be remembered that, as always with BCBS standards, these are regarded as minimum standards and national supervisors may choose to implement stricter requirements.

go to the top

The key changes

In its press release, the BCBS highlighted the following changes:

  • An increase in the range of eligible assets that can be held as part of the required liquidity buffer (but note that this is subject to the discretion of each national supervisor);
  • Changes to the assumptions as to cash outflows and inflows ‘to reflect better experience in times of stress’;
  • The revised timetable (as highlighted above); and
  • A clarification that it is expected that supervisors will permit the use of the liquidity buffer in times of stress i.e. banks will be allowed to use the buffer, and thus fall below the minimum ratio, in a liquidity stress scenario.

There is significant additional detail, with the new standard running to some 75 pages (the December 2010 standard, which covered both LCR and NSFR, ran to 47 pages).

Overall, the approach is as set out in the December 2010 standard – banks must assess the potential outflow in a liquidity crisis based on supervisory assumptions as to the level of outflow (not banks’ internal models) and must maintain a stock of unencumbered High Quality Liquid Assets (HQLAs) to be able to withstand a 30-day stress scenario.

The standard is divided into two parts. Part 1 sets out the rules for the LCR, and Part 2 sets out other standard metrics that should be used as monitoring tools, without any specific limits or ratios that need to be met as part of these.

go to the top

High Quality Liquid Assets

It was probably inevitable that the definition of eligible assets, HQLAs, would be broadened. Without this change, the new rules would have pushed the banks to lend more to governments at a time when the Eurozone crisis has thrown into question the traditional regulatory assumption that lending to governments is risk free. This assumption has been replaced with an understanding that there is a systemic feedback loop, whereby government support for weak banks can cause a sovereign debt crisis, and bank support for fiscally-challenged governments (through holding sovereign bonds) can cause a banking crisis.

The characteristics of HQLAs remain broadly unchanged, although the first characteristic has been amended from ‘low market and credit risk’ to simply ‘low risk’, which implies a tightening of the definition (to include for example ‘low liquidity risk’) which is somewhat at variance to the loosening of the eligibility criteria.

While there is also no longer a specific requirement under the ‘fundamental characteristics’ for such assets to be supported by ‘presence of committed market makers’, which would be consistent with the broader eligibility criteria, it should be noted that under the ‘market related characteristics’ there is a requirement that ‘There should be robust market infrastructure in place. The presence of multiple committed market makers increases liquidity as quotes will most likely be available for buying or selling HQLA’, which is essentially the same thing.

A new requirement is that the stock of HQLAs should be ‘well diversified’ (para 44), in recognition of the fact that the potential liquidity of individual asset classes in a future stress scenario cannot be predicted with any certainty.

The operational requirements related to HQLAs have been clarified, including making it clear that banks should be permitted by their supervisors to use the liquidity pool at times of stress. Also, additional details regarding the control of the HQLAs have been provided in order to better reflect industry practice.

BCBS has also confirmed that supervisors have national discretion to include or exclude central banks reserves.

The previous Level 1 and Level 2 classifications have been further refined with a new, additional classification of Level 2B. The 40% cap on Level 2 assets remains, and there is a new cap of 15% on Level 2B (the cap is expressed as a % of the total HQLA requirement). More detail on how to calculate the cap is given in Annex 1, which provides more detail than the previous standard.

The definition of Level 1 is unchanged, although the language has been modified slightly. Level 2A assets are the same as the previous Level 2, and are subject to the same 15% haircut.

Level 2B assets, which are new, may only be included at the discretion of the national supervisor. The eligible assets, and the haircuts to be applied, are:

  • Residential mortgage-backed securities (RMBS) that are rated AA or higher, not issued by the bank or its affiliates, which are traded in deep and liquid markets and have not fallen by more than 20% over 30 days in previous stress situations – subject to a 25% haircut. It should be noted however that RMBS can only include ‘full recourse’ loans, which would exclude most US RMBS. There is also a loan-to-value cap of 80% at the time of issuance.
  • Corporate debt securities rated between A+ and BBB-, with other conditions similar to RMBS – subject to a 50% haircut.
  • Common equities, which are exchange traded, a constituent of a major market index, denominated in the same currency as the deposits giving rise to liquidity risk, and with a market history of no more than a 40% fall over 30 days – subject to a 50% haircut.

Note that this is a summary of the criteria – the detailed requirements should also be looked at carefully.

The new standard maintains the original options for those jurisdictions which do not have a sufficient supply of Level 1 assets, but the use of these options is now constrained by a ‘maximum use’ cap (paras 63-65). Greater specificity as to when alternative approaches would be eligible is new, set out in Annex 2 in the form of three principles, with eligibility criteria set out under each principle. A new Annex 3 further sets out standards for banks’ use of these alternative approaches.

There is also a new provision to allow for the inclusion of Shariah-compliant products for Islamic banks, but these products do not count as HQLAs for non-Islamic banks.

go to the top

Cash outflows and run-off rates

The assumed run-off rate for a sub-set of ‘stable’ retail deposits has been reduced from 5% to 3%, but that for ‘less stable’ deposits remains at 10%. To qualify as ‘stable’, the deposits must be covered by an effective deposit insurance scheme or equivalent (as before), and the new rules make it clear that only the portion of retail deposits covered by such schemes qualify for the ‘stable’ category (e.g. some jurisdictions, such as the US, cap the insurance per depositor at a fixed amount). However, retail deposits meeting the original criteria as ‘stable’ remain at 5% unless further conditions are met as to what qualifies as an ‘effective’ deposit insurance scheme for the purposes of the 3% run-off assumption, notably that it must be pre-funded.

The previous category of ‘retail fixed term deposits’ has been removed, and subsumed within the ‘stable’ and ‘non-stable’ categories of retail deposits, but the basic rule remains – deposits which cannot be withdrawn within the 30-day period covered by the LCR (or which would be subject to significant penalties if they were to be withdrawn) remain excluded from the LCR.

Run-off rates are as follows:

  • The run-off rate for unsecured wholesale funding provided by SMEs remains, as before, the same as for retail (i.e. 5% or 10%). However a lower run-off rate of 3% is now allowed on retail deposits covered by qualifying deposit insurance
  • The category of ‘operational deposits’ (previously ‘unsecured wholesale funding with operational relationships’) has been slightly re-defined and set out in more detail, but the 25% outflow assumption remains unchanged
  • Deposits in networks of co-operative banks remain unchanged (25% or 100%)
  • The assumed run-off rate for wholesale unsecured funding provided by non-financial corporate and sovereigns (including central banks and PSEs) has been significantly reduced, from 75% to 40%, or even to 20% if the deposits are covered by an effective deposit guarantee scheme or public guarantee
  • The run-off levels for maturing secured funding transactions now include those backed by Level 2B assetsSpecifically, run-off rates for transactions backed by RMBS eligible for inclusion in Level 2B and by other Level 2B assets will be 25% and 50% respectively. Secured funding transactions with central banks will benefit from a run-off rate of zero
  • For the ‘other’ category, which includes things like derivative cash outflows, there are a number of changes as follows:
    • Drawdown on committed liquidity facilities to non-financial entities – reduced from 100% to 30%
    • Drawdown on committed credit and liquidity facilities to banks – reduced from 100% to 40% (as long as the entity is subject to prudential supervision)
    • The definition of liquidity facilities has been clarified; in particular, facilities provided to hedge funds, money market funds and special purpose vehicles or vehicles used to finance the banks’ own assets will be classified as liquidity facilities
    • Additional derivatives risks included in the LCR with 100% outflow (relates to collateral substitution, and excess collateral that the bank is contractually obligated to return/provide if required by a counterparty)
    • Introduction of a standardised approach for liquidity risk related to market value changes in derivatives positions, based on the largest absolute net 30-day collateral flow realised during the preceding 24 months – supervisors may adjust the treatment if deemed necessary
    • 0% outflow for derivatives and commitments secured by HQLAs
    • Clarification of treatment of activities related to client servicing brokerage (generally leading to an increase in net outflows)
    • Guidance to indicate a low outflow rate for trade finance related activities: 0-5%.

The assumptions as to drawdown rates for committed credit and liquidity facilities have been expanded, with new 30% and 40% drawdown bands in addition to the existing 5%, 10% and 100% bands.

National discretion remains the approach for other contingent funding obligations (such as guarantees, letters of credit etc). However, the new rules allow national supervisors to apply a ‘relatively low run-off rate’ of 5% or less for trade finance commitments.

go to the top

Cash inflows

The overall constraints on offsetting inflows against outflows remain unchanged – inflows have to be contractual, with no expected defaults, and there is an overall cap on inflows vs. outflows of 75%.

The assumptions on the extent to which reverse-repo and similar transactions will roll over (and not result in cash inflows) are broadly unchanged, with the following exceptions:

  • The addition of Level 2B HQLAs has led to additional assumptions for this type of asset (25% inflow for eligible RMBS and 50% for other Level 2 assets); and

The inflow rate for margin lending backed by other collateral has been reduced from 100% to 50%. The rules for cash inflows remain materially the same in all other respects.

go to the top

Application of the metric

There are no material changes to the frequency of reporting (monthly, or more frequently in a stress situation), or the scope of application (consolidated and also at the subsidiary/overseas branch level, where relevant).

go to the top

Monitoring tools

The second part of the standard sets out the standard liquidity monitoring metrics that supervisors should expect to see and use. These metrics are for information purposes, and there are no minimum ratios, caps etc which need to be met.

The list of metrics (unchanged over the previous standard) covers:

  • Contractual maturity mismatch;
  • Concentration of funding;
  • Available unencumbered assets;
  • LCR by significant currency; and
  • Market-related monitoring tools.
go to the top

PwC preliminary assessment

Banks should not take too much comfort from the much-heralded relaxation of the timetable. The LCR still becomes binding (effectively a Pillar 1-style requirement) as of 1 January 2015, although at a reduced level.

The most recent BCBS Quantitative Impact Study (QIS), published in September 2012, indicated that there was an aggregate shortfall of Euro 1.8tn for the banks that would have failed to meet the 100% LCR target as at December 2011 (the shortfall had not shrunk since the previous QIS). The majority of Group 1 (larger banks) failed to hit the target, whereas the majority of Group 2 (smaller banks) exceeded it. Therefore Group 1 banks are likely to benefit most from the phasing in period and the broader criteria for HQLA.

From an operational and systems perspective, banks still need to be in a position to report the LCR to their supervisors as of 1 January 2015 (and probably earlier). In order to avoid any unpleasant surprises, banks need to have the infrastructure in place long before then. In addition they will need to make any adjustments to their balance sheets in good time (recognising that the phase-in arrangements will give more time to make adjustments to meet the 100% minimum requirements) and resolve the inevitable data issues that the new reporting requirements will reveal. The level of reporting detail, and in particular the need to be able to report on individual legal entities and overseas branches where required, and with tailoring to meet home and host supervisory requirements, should not be underestimated. For EU banks, CRDIV is likely to require greater granularity than Basel III.

From a business perspective, the impact of the inclusion of Level 2B assets and the watering-down of some outflow assumptions will depend on a bank’s funding model and the availability of assets in the markets in which the bank operates. It will be interesting to see if the broader definition of eligibility stimulates demand for the paper in question, particularly as banks seek to optimise the yield on their buffer assets.

Clearly the reduction of outflow rates for some ‘stable’ retail deposits from 5% to 3% could be a significant benefit for banks with a big retail deposit franchise, and the reduction in outflow assumptions for unsecured wholesale credit from the non-financial sector will be a significant benefit for those with a large corporate depositor base. However, it should be noted that to qualify for the lower 3% category, there needs to be an effective deposit insurance scheme or equivalent public guarantee in place, which must be pre-funded. This would exclude (currently) retail deposits, for example in the UK and some Eurozone countries, so the ‘benefit’ provided by this relaxation may, in fact, be quite limited.

Run-off rates have been reduced for transactions backed by Level 2B assets and secured funding transactions with central banks, outflows related to derivatives and client servicing (prime) brokerage could increase.

The reduction of the outflow assumption for trade finance removes what was feared to be a major blow to this line of business. Similarly, the reduction in the runoff rate for non-operational, non-financial corporate deposits from 75% to 40% will be welcomed by corporate business lines who would otherwise have faced greatly reduced customer profitability. Finally, the reduced drawdown assumption, from 100% to 30%, for committed corporate liquidity facilities, helps address a major pricing and credit availability concern shared by corporate bankers and corporate treasurers.

How the standards will be adopted in various jurisdictions remains to be seen. There are still some concerns around the consistency of application with some clear national discretion areas such as the inclusion of central banks reserves and several areas left to local regulators (such as triggers and actions associated with LCR depletion). In the EU, the European Banking Authority said in December that it expected firms to submit monthly liquidity and leverage data in Q1 2014 but the exact requirements are still to be defined in the regulations as part of CRDIV. It will be interesting to see how much detail of the HQLA definition is specified in the European regulation.

In conclusion, the transitional arrangements, with the LCR starting at a minimum 60% and then increasing by 10 percentage-points a year until it reaches 100% on 1 January 2019, must be seen as what they are – provision of breathing space for banks that would find it hard to meet the full metric by 1 January 2015. However, it is simply a deferral of the eventual, revised full impact, and, as we have seen with the Basel III capital ratios, the market will be quick to expect banks to meet the full ratio in advance, and to punish those who are perceived to be lagging.

To discuss any of the issues raised in the briefing note please speak to your usual PwC contact or one of the contacts listed below.

go to the top

Author