Liquidity Risk, Paul Sharma (speech)
Liquidity Risk
Friday, 8 October 2004
Speech by Paul Sharma
The subject of my speech is liquidity risk. The Department which I head at the Financial Services Authority covers all prudential risks – market risk, credit risk, operational risk, insurance risk etc as well as of course liquidity risk – across the three sectors of banking, insurance and securities. My Department has of course therefore been heavily engaged in negotiating and implementing the reforms to the regulation of credit, market and operational risk in Basel 2 and the reforms to the regulation of insurance risk in our new realistic solvency regime. These reforms have been based on years of detailed work, extensive consultation and intensive negotiation and are now the subject of major implementation projects at the both the FSA and regulated firms. The reforms have succeeded in focusing attention on credit, market, insurance and operational risk. A vast literature, both regulatory and academic, exists on these risks. In sharp contrast almost no attention has been given to liquidity risk.
However this too is now changing and in my speech I will focus on two recent initiatives. Firstly there are the reforms set out in the FSA’s Consultation Paper 128 and in the feedback statement to that consultation paper. These reforms set out new rules and guidance on the non-quantitative aspects of liquidity risk management. The rules and guidance are due to come into force at the end of this year1. Secondly there is the work of the Joint Forum of the Basel Committee and its international regulatory counterparts, IOSCO and the IAIS, in the securities and insurance fields. The Joint Forum is the senior international body, at present under the chairmanship of the FSA’s Gay Huey Evans, that looks at key regulatory issues that are of common interest across the three sectors. It has set up a sub-group under the joint chairmanship of Richard Mead (New York Federal Reserve) and myself to report on liquidity risk. Joint Forum does not itself legislate regulatory standards but in the past its finding and recommendations have lead to others carrying out significant regulatory reforms. For example the initiative which lead to the EU directive on conglomerates was based initially on a Joint Forum report.
Prior to these recent initiatives we have grown so accustomed to extensive regulatory and academic discussion of credit risk, market and even operational risk that the comparative absence until recently of liquidity risk from the regulatory debate no longer seems surprising. Actually it should surprise us a great deal. Historically in the 1930s and again in the 1970s liquidity runs have been the causes of multiple banking failures within the UK leading to systemic instability and there are of course much more recent examples if one looks outside the UK. Also analytically if one looks at the basic business model it is obvious that liquidity is one of the fundamental risks that arises in banking.
Within their banking books, banks borrow short and lend long. This gives rise to three obvious risks – credit risk on the lending and long-term interest rate risk and liquidity risk from the mismatch between the term structure of the assets and liabilities. Basel 2 sets out regulatory standards for the first two risks but is almost silent on the liquidity. Within their trading books, banks and investment banks hold trading assets backed by regulatory capital calculated on a VaR basis that assumes that the market risk from those assets could be mitigated within a short period of time. This in turn assumes that those assets are liquid although there is increasing evidence that in recent years banks and investment banks have started including significant volumes of illiquid assets in their trading books.
I shall speak about the latest trends in the regulatory responses to these risks in a moment but before doing so I shall first set out what I mean by liquidity risk. A conventional analysis of liquidity risk distinguishes between funding liquidity risk and market liquidity risk.
- Funding liquidity risk is the risk that the counterparties who provide the bank with short-term funding will withdraw or not roll over that funding, e.g. there will be a ‘run on the banks’ as depositors withdraw their funds.
- Market liquidity risk is the risk of a generalised disruption in asset markets that make normally-liquid assets illiquid.
The first is more important in the context of the maturity transformation that occurs in the banking book. The second is more important in the context of tradable assets in the trading book.
However rather than define liquidity risk in this rather conventional way I would prefer to describe it using some of the concepts being developed in the Joint Forum work. This sees liquidity risk in terms of adverse liquidity outcomes that arise from a combination of an external or non-liquidity trigger event and an internal vulnerability.
- An obvious adverse liquidity outcome is (1) the inability to pay liabilities as they fall due. However many firms take liquidity risk further than this. They include as adverse liquidity outcomes (2) realising a market loss as a result of the premature or force sale of assets to raise liquidity and (3) loss of business opportunity or franchise due to a lack of liquidity. This broader vision of liquidity is also relevant to regulators. It illustrates the link between liquidity risk and market risk. There is a correlation between the times at which market volatility – and therefore market risk – is highest and times at which the market also loses its liquidity. It also illustrates the focus on liquidity risk which regulators need to have where a bank is of systemic importance such that the loss of its business franchise would disrupt the financial system as a whole.
- Internal vulnerabilities to liquidity risk arise principally either because (1) assets are in relative terms less liquid than liabilities or (2) a bank has granted its counterparties significant optionality. The first point here draws attention to the role of asset-liability matching in creating and managing liquidity risk. Neither illiquid assets nor liquid liabilities (that is liabilities whose timing is uncertain) in themselves create liquidity risk. It is the mismatch which creates the risk. This may sound as if I am stating the obvious but conventional regulatory approaches often focus in a one-sided way on the virtue of liquid assets, as arguably our sterling stock liquidity regime for banks does, or on the harm from illiquid assets, as arguably our illiquid assets deduction rules for investment firms do. The second point here draws attention to another fundamental aspect of liquidity risk. It is part of the business model of banks and investment banks to do business with their customers and counterparties in ways which give those customers or counterparties a significant degree of choice, that is optionality, as to the timing of cash outflows to them. For example customers who hold sight deposits may withdraw those deposits at any time. Counterparties who are providing short term secured or unsecured funding may choose not to roll over that funding. This leaves banks and, to a lesser extent, investment banks open to the risk that customer or counterparty behaviour will alter suddenly and radically – the so-called risk of a ‘run on the bank’. This in turn leads us to focus on reputational risk and contagion risk of which more in a few moments.
- External or non-liquidity triggers need to combine with these internal vulnerabilities to create an adverse liquidity outcome. The Joint Forum work identifies three main triggers. First a non-liquidity risk may crystallise either directly leading to the creation of a short term liability or to the loss of use of a short term or liquid asset or indirectly leading to that result by first causing damage to reputation that in turn leads to a change in customer or counterparty behaviour. Second a similar result may follow due to reputational damage that occurs due to misinformation, misunderstanding or overreaction in the market. A particularly relevant example of this is the contagion damage that occurs to a bank’s reputation when a similar but unrelated bank suffers financial or other difficulties. Third a more general market disruption of liquidity may frustrate a bank’s ability to raise unsecured funds and/or to convert assets into liquidity either through repos or secured lending or through sale.
In risk management terms preventing trigger events, reducing vulnerabilities and, when they occur, mitigating adverse outcomes are all important strategies. In comparison conventional regulatory responses appear somewhat one-dimensional focusing nearly exclusively on a few narrow methods of reducing one particular aspect of vulnerabilities. The next stages of the FSA’s reform of its regulatory approach to liquidity risk have, and will have, a broader, but also more flexible, regulatory focus. I shall now turn to this.
In discussing the regulatory approach to liquidity risk it is important to distinguish between quantitative (pillar 1), qualitative (pillar 2) and disclosure (pillar 3) requirements. Turning first to pillar 1 we have at the UK level at present a patchwork of different regimes as set out in the various interim prudential sourcebooks. At the EU and international levels there is as yet no harmonisation of quantitative liquidity rules. Following Basel 2 and the draft EU Capital Requirements Directive this remains by the far the most important aspect of banking regulation on which there are no internationally or EU agreed standards.
As part of the FSA’s programme to introduce an Integrated Prudential Sourcebook that treats like risks in a like manner regardless of the sector in which they occur, the FSA set out in Discussion Paper 24 in October 2003 some ideas of how a single quantitative regime across the banking, building society and investment banking sectors might be created.
In principle there are three broad approaches that a firm might use in the quantitative control of its liquidity risk.
- The stock approach. The firm maintains stocks of liquid instruments that can be drawn upon when needed. The amount, quality and nature of the stock of assets are most typically calibrated to deal with crisis events but the stock of assets is of course also potentially available to deal with normal variations in cash flows.
- The cash flow matching approach. The firm attempts to match cash outflows and inflows. The cash flow matching approach may be applied using solely contractual cash flows or using adjusted cash flows. Adjusted cash flows adjust the contractual cash flows to take account of the likely behaviour of counterparties and/or non-contractual cash outflows needed to preserve the business franchise. These adjustments may be made either on the basis of expected normal conditions or assuming stress conditions.
- The mixed approach. This approach combines elements of the cash flow matching approach and the stock approach. Adjusted cash flows are projected as under the cash flow matching approach except that it is assumed that the stock of liquid assets are not merely held to maturity but are used to generate cash inflows either through their sale or through repo or other secured lending transactions. This adaptation of the cash flow model is targeted to stress environments.
The existing IPRU regimes for banks and building societies follow a stock approach for large banks and cash flow matching and mixed approaches for smaller banks and for building societies. However the detail of these various different existing approaches as expressed in the Interim Prudential Sourcebook is open to criticism as needing to be update for modern market developments and also to create a level playing field across the sectors on the basis, as already explained, of “like risk like treatment”. For example the stock liquidity approach that is at present used for large banks suffers from the defects that it only applies to sterling, it only accepts a narrow range of assets as stock liquidity assets and it is not the main way large banks actually manage their liquidity risk for internal risk management purposes. The existing regime for investment firms uses an approach based on illiquid asset deductions which in effect fuses and conflates capital rules and liquidity rules.
Discussion Paper 24 set out ideas for a new integrated approach to Pillar 1 requirements for liquidity risk. It basically followed the ‘mixed approach’ described above. The responses to the discussion paper received from the industry and others supported the idea that there was a need for reform but also, as we expected, pointed out that significant further work was needed. The ideas in the discussion paper were not yet fit to be taken forward to a consultation paper stage. In particular four important points were made by industry commentators with which we agree. First, the outline new approach set out in the discussion paper was far too prescriptive and detailed. Second, it would create new divergences between the way in which liquidity risk was regulated and ways in which it was managed by firms for their own internal purposes, especially for the investment banking sector which typically uses an approach closer to the stock approach. Third it placed too much weight on a Pillar 1 approach of the regulator setting hard quantitative limits as opposed to relying upon and developing further the FSA’s own Pillar 2 ideas that rest on the concept of the senior management of a firm forming its own view – subject to supervisory review – as to their firm’s liquidity needs. Fourth it risked creating a Pillar 1 regime for liquidity in the UK that was significantly out of step with the emerging international regulatory trends and was not sufficiently cross-sectoral in its approach especially for mixed banking-insurance conglomerates. The FSA will not therefore be taking forward the Discussion Paper 24 approach to Pillar 1 liquidity standards to the consultation stage. Instead the next steps in our work on the reform of liquidity risk will be to implement our existing Pillar 2 proposals and to participate further in the international work on Liquidity. That is in particular the Joint Forum sub-group which I co-chair. Further domestic reform will await the outcome of this international work and will be reassessed in the light of that outcome.
The FSA set out its Pillar 2 requirements in its feedback statement to Consultation Paper 128. This sets out rules and guidance that are due to come into force at the end of this year that require a firm:
- to carry out stress testing and scenario analysis of liquidity needs;
- to put in place contingency funding plans for dealing with a liquidity crises were it to occur; and
- to document its liquidity risk management policy.
These rules and guidance anticipate and amplify the requirement set out in the draft EU Capital Requirement Directive that firms have in place “policies and processes for the measurement and management of their net funding on an on-going and forward-looking basis”. The directive adds that “Alternative scenarios shall be considered and the assumptions underpinning the net funding position shall be regularly reviewed” and that “Contingency plans to deal with liquidity crises shall be in place”.
Turning now to Pillar 3 (disclosure requirements), this is the aspect of the regulation of liquidity risk that has received the least attention of all at either the UK or international level. However the Joint Forum sub-group is now looking at this and the FSA will await its findings.
Drawing my remarks to a conclusion I would repeat the thought with which I started. Liquidity risk has perhaps for too long been overlooked as a focus of attention for regulatory reform. That is now changing both at the UK and International levels.
by Paul Sharma - Head of Prudential & Accounting Standards, FSA.
1See the FSA’s letter of 15 September to trade associations on the PSB Implementation page (under the Integrated Prudential Sourcebook heading) on the FSA’s website for a precise statement of which aspect of the CP128 rules and guidance are to be brought into force.
http://www.fsa.gov.uk/Pages/Library/Communication/Speeches/2004/SP201.shtml