1. Introduction In order to understand the usefulness of VaR and other risk metrics for the setting of capital adequacy requirements it is useful to compare various measures used by financial institutions and legislative statutes of the Basel Frameworks. Traditional approaches to banking regulation emphasises the understanding that the existence of capital adequacy plays a central role in the long term financing and solvency positions of banks, especially in helping the banks to avoid bankruptcies and their negative externalities on the financial system (Dewaitpont and Tirole 1994) The notion of liquidity must be well defined 'unfortunately the word, liquidity has so many facets that it is often counter-productive to use it without further and closer definition' (Goodhart 2008). However; in the context of liquidity risk management, a bank's liquidity can be defined as the ability to fund increases in assets and to finance obligations as they fall due. Therefore liquidity refers to the risk resulting (Nier 2005) from a financial institutions failure to pay its debts and obligations when due because of its inability to convert assets into cash readily. Moreover, liquidity risk also refers to the inability to procure sufficient funds due to high costs of liquidity transformation that may affect the financial institutions revenues and capital funding either now or in the future. The main objective of liquidity management is to ensure adequate liquidity in all circumstances so that banks have the ability to meet its cash flow obligations. Since maturity transformation of short-term deposits into long term loans is one of the banks fundamental roles banks are therefore inherently vulnerable to liquidity risk stemming from both an institutional-specific nature and a contagion effect which has the ability to cause a 'ripple' effect throughout global markets.
2. Liquidity Management Several areas are of concern in the context of liquidity risk management, (Nier 2005) firstly data may be scarce and lacking in quality and historical data is not necessarily an accurate predictive agent; thus data may not be a reliable proxy for stress testing. Sound liquidity management for both short term and long run purposes is an integral component of a banks contingency funding plan that would aid banks in the event of a financial crisis. Fundamentally, liquidity risk measurement comprises four measurement systems (i) use of ratio analysis (Dowd 2002) where the applications of ratios are developed to measure various components of a bank's balance sheet. Such ratios include the minimum liquid asset (MLA), the capital asset ratio (CAR) and the minimum cash balance (MCB). In addition a banks liquidity position needs to be monitored with the application of these ratios both on-balance-sheet and off-balance-sheet terms (ii) Cash flow measures; where a projection of cash flows based on both supply and demand for liquidity exists under normal market conditions. The recent global financial crisis has highlighted the importance of adequate liquidity of banks coupled with five key features relating to financial regulation and (Cross 2010) supervision; systematic risk, pro-cyclicality, regulatory arbitrage and transparency. The inadequate regulation and supervision of banks globally has prompted regulators to review current liquidity requirements and statute in order to mitigate liquidity risk and prevent future crises' from recurring. The existing approach to capital regulation in the US and E.U is based on Basel I and Basel II and has been identified by regulators and commentators as one of the key factors contributing to the financial crisis. However, Basel I and Basel II focused on capital only, with no internationally agreed (Moody's 2011) quantitative standard for liquidity. In December 2010 the Basel Committee on Banking Supervision published the final form of a set of reforms to strengthen liquidity risk management by international active banks (the 2010 Liquidity Paper). The liquidity paper is intended to address concerns highlighted in the Economic crisis, where a lack of liquidity and inadequate liquidity risk management operated together to amplify difficulties caused by credit losses and due to the interconnectedness of markets affected all (Moody's 2011) markets with subsequent dire consequences. The Basel iii revises proposals set out in the initial framework for improving liquidity risk management and controlling liquidity risk exposures set out in the Committee paper adopted in September 2008.
3. The Basel Accord and Ratio's Whilst the problem of solvency was at the core of the financial crisis between 2007- 2009, it demonstrated that illiquidity can amplify the depth of such a crisis'. A bank can face impending illiquidity of two kinds: (i) Market Illiquidity which occurs when banks cannot sell assets without realising large losses and (ii) Funding liquidity when banks that rely on short-term funding cannot refinance long maturity assets (ESFRC 2011). If banks hold enough highly liquid assets and do not place heavy reliance on short-term funding, the contagious effects of capital deficit will be lessened. Market discipline cannot be relied upon to resolve this externally; it however could be addressed by increasing capital requirements. However, the costs to the banking system would be reduced by employing liquidity requirements along with less stringent capital requirements. The Basel Committee has evoked two requirements that must be satisfied by banks regarding maturity transformation. The liquidity coverage ratio (LCR) is designed to promote short term liquidity resilience which compares the stock of high quality liquid assets held by a bank to its net cash outflows (Moody's 2011) during a hypothetical 30-day severe stress scenario. The liquidity ratio will be set at a minimum of 100%, requiring high calibre liquid assets to fully cover the net cash outflows in such a scenario, and the liquidity coverage ratio must be maintained at all times. The Net Stable Funding Ratio (NSFR) refers to a ratio between availability of stable funding relative to the need created by long-term assets. The NSFR limits the degree of maturity transformation of banks, and therefore enhances funding liquidity.
4. Weaknesses of Basel Accord Standards Both sets of ratios are based on a complex set of weighing factors, which could be specified in a simpler manor. Instead of a variety of weighted factors, liquidity requirements could be in the form of a minimum ratio of cash and other highly liquid and riskless assets to total (ESFRC 2011) assets instead of the LCR and a simple measure of maturity mismatch instead of the NSFR. These requirements should be applicable under normal economic conditions; however in a period of a weak economic climate could these conditions be relaxed. Basel iii definition of ' high quality liquid assets in the context of the LCR ratio consist of cash and high quality government debt plus discounted proportions of high quality corporate and covered bonds. There is a risk that the 'high quality assets' standard is too conservative to the end that it could create a shortage of liquid assets or significant concentration risks (Ref). Thus it is more restrictive than the standards central banks typically maintain for collateral eligibility under the liquid facilities that serve as a key area to the banking system. Basel iii new liquidity standards should be an addition to firm level risk management and micro-prudential regulation, if combined with micro-prudential regulation and improved supervision. By raising liquidity buffers and reducing mismatches the new standards will indirectly address systemic liquidity risk as it will reduce possibility that banks will have a simultaneous requirement for liquidity. However policymakers will need to ensure that the weights and factors in the calibration of such ratios do not fully restrict banks (ESRFC 2003) in their ability to undertake maturity transformation or in the ability of money markets to act as a buffer for the financial institutions to manage their short term liquidity needs. If the standardization is too restrictive it may encourage migration of some banking activities into less regulated practices including towards shadow banks thus potentially accentuating rather than alleviating systemic risk. The Basel iii standards could therefore extend the quantitative liquidity requirements to less regulated institutions. A framework that is too rigid may force banks to take risks to reach compliance, resulting in a high correlation amongst particular assets and concentrations in some of them. Consequently, the LCR ratio may inevitably tip towards high holdings in eligible liquid assets that could effectively reduce liquidity during a systemic crisis. Also, by applying unvarying quantitative standards across countries may not be suitable as a number of countries may not have the markets to extend term funding for banks given the absence of a bond market in a domestic currency, which would accordingly require banks to be subject to exchange rate risks. An analysis on the NSFR by (OECD 2010) finds that the ratio would not have indicated problems in the banks that ultimately failed due to poor liquidity management and overuse of short term wholesale funding. Therefore the NSFR appears to have several limitations and should not be used as an appropriate technique to mitigate liquidity risk. For Basel iii to be effective liquidity requirements will need to be set at a high level for all institutions, resulting in a prohibitive cost to the real economy; otherwise the possibility will always exist that a (OECD 2010) systemic liquidity event will exhaust all available liquidity. In such circumstances central bank support is warranted to ensure that systemic liquidity shortages to not morph into large scale solvency problems. A problem so far has been the lack of analysis of a uniform measure of liquidity risk and to the extent to which an institution contributes to this risk.
5. Liquidity Risk Measurements & Metrics Including Value-at-Risk (VaR) The analysis of liquidity requires bank management to identify measure and monitor its positions on an on-going basis as well as to examine how funding requirements are likely to evolve under various scenarios' including adverse conditions (Cross 2010). However, 'liquidity is difficult to define and even more difficult to measure' (Persaud 2007), due to the underlying variables driving the exposure can be dynamic and unpredictable. Until recently, managing and measuring liquidity risk was rarely seen as a high priority by most banks and financial institutions. Furthermore, no agreement has existed in the international community on the proper measurement of liquidity; hence there was not an integrated measurement tool to cover all dimensions of liquidity risk available to financial institutions. As to liquidity risk metrics in use, it is considered necessary to distinguish between analytical approaches such as VaR, that are focused on assessing potential effects on profitability, liquidity risk models and measures which aims at assessing cash flow projections of assets and liabilities, or the inability to conduct business as a result of a lack or a reduction of secured and unsecured funding capacities and/or liquid assets. Banks generally apply a variety of measurement techniques dependent on the specific type of risk that they want to assess, (e.g. funding liquidity risk, market liquidity risk etc.) Where institutions have adopted quantitative analyses for the assessment of liquidity risk, this approach has tended to be a deterministic (Cross 2010) one, such as static maturity ladders however; in such cases distributions for determining risk exposures are not utilised as scenario analysis is based on user-defined assumptions and resulting estimates therefore produce only a single view of the future. Therefore a more effective alternative is a stochastic approach which has been proven effective for both market and credit risk management. In this framework, (Cross 2010) the future values of risk factors are calculated under a variety of randomly generated scenarios thus producing probability distributions. See Appendix (1) for Stochastic Approaches Thus in reality most markets are less than (Cross 2003) perfectly liquid. If regulators in countries required banks to use VaR models for risk quantification processes the results from such models would produce inaccurate results as (i) there is no estimate of tail risks and losses (ii) difficulties in identifying the non-linear pay-offs characteristics of many complex and structured products (iii) no consistent method of aggregating risks across different asset classes, (iv) concentration on the distribution of portfolio value changes resulting from movements in the mid-price of each asset and (v) separate modelling of asset prices and portfolio size amongst others. Bangia et al (1999) cites that VaR methodology does not distinguish between market risk and liquidity risk, because historical market prices are supposed to embrace latent liquidity effects. Severe critique has been made regarding VaR as a measurement of liquidity risk; whilst it isn't completely appropriate it does still give an insight into the level of risk of an institution. Hence where VaR is insufficient, through the use of stress testing it becomes an adequate compliment (Kotz & Gerhrig 2010). Where VaR reflects price behaviour in everyday markets stress testing simulates portfolio performance during abnormal market periods. The CGFS (2005) cites that stress testing is increasingly viewed as a complement to the previously defined VaR rather than as a supplement. Generally two types of stress testing are differentiated, the Scenario tests where the source and the financial risk parameters that are affected by the shock are well defined, and the Sensitivity test in which neither the shock nor the parameters are defined. The BCBS (2008) strongly recommends that regular stress testing of banks is implemented as it can be helpful in detecting liquidity risk and checking if the current exposure remains in accordance with the banks established risk tolerance. VaR models assumes model conditions as to the unwinding of the position with one trade at a predetermined price equal to the current quoted mid-price, within a fixed period of time and no consideration of the size of the position. Liquidity in the market is connected to a variety of factors (Cross 2003) including the relative size, frequency, traded volumes, and the credit worthiness of the issuer amongst others, thus in order to account for these variables the standard VaR will require an adjustment to incorporate market liquidity and transaction costs into the VaR framework. See Appendix (2) for VaR calculations.
6. Conclusion It is unlikely that there is a single and uniformly best measure of liquidity risk considering the differing natures of financial institutions and their respective funding arrangements. However analysis finds that standard VaR methodology is an inadequate measure of liquidity risk as it does not distinguish between market and liquidity risk and does not take into account the level of risk within a particular institution. Adjusted VaR methods coupled with stress testing have proven to be a compliment which will incorporate liquidity risk into the computation. Other measurement methods such as the SRL model has the benefit of using daily market data and standard risk management methods to interpret individual contributions to systemic risk into a macro-prudential measure. The SRL can produce opportune and forward looking measures of risk of simultaneous liquidity shortfalls in financial institutions (IMF 2011). Alternatively or as a compliment to the SRL the ST framework could be implemented, as with other stress testing techniques it captures systemic solvency risk by assessing the vulnerabilities of institutions to a common macro-financial shock, and then adds this to risk of liquidity shortfalls ad assesses transmission of liquidity risk to the rest of the system through their exposures to the interbank market (2011).
References Bangia A., Diebold F.H., Schuermann T., Stronghair J.D (1999), "Modelling Liquidity Risk with Implications for Traditional Market Risk Measurement and Management", Working Paper, pp. 99-106, Wharton School, Philadelphia. Bardenhewer, M (2007) "Modelling Non-maturing Products", in Matz L., Neu P., Liquidity Risk. Measurement and Management. A Practitioner's Guide to Global Best Practices, John Wiley & Sons, Chichester. Barrel, R et al (2009) Optimal regulation of bank capital and liquidity: how to calibrate new international standards [Online] Available At: http://www.fsa.gov.uk/pubs/occpapers/op38.pdf [Last Accessed 29 April 2011]. Basel Committee on Banking Supervision, A Framework for Measuring and Managing Liquidity, September 1992. Basel Committee on Banking Supervision, Sound Practices for Managing Liquidity Risk in Banking Organisations, February 2000. Basel Committee on Banking Supervision, The management of liquidity risk in financial groups, Bank for International Settlement, May 2006. Basel Committee on Banking Supervision (2011), Principles for Sound Liquidity Risk Management and Supervision. Blanco, C (2010) Financial Liquidity Adequacy [Online] Available at: http://www.blackswanrisk.com/pdf/June04RiskDesk.pdf [Last Accessed 28 April 2011]. Cross, A (2010). The new capital and liquidity proposals- Implications for Banks and their Supervisors [Online] Available at: http://siteresources.worldbank.org/FINANCIALSECTOR/Resources/Session4AndrewCross.pdf [Last Accessed 23 May April 2011]. Dowd K., (2002) Measuring Market Risk, John Wiley & Sons, Chichester. ESFRC (2011) Basel III The need for simplicity in capital and liquidity requirements [Online] Available at: http://www2.lse.ac.uk/fmg/events/conferences/2011/financialRegulation_24Jan2011/ESFRC_Statement.pdf [Last Accessed 01 May 2011]. IMF(2011) How to address the systemic part of liquidity risk [Online] Available at: http://www.imf.org/external/pubs/ft/gfsr/2011/01/pdf/chap2.pdf [Last Accessed 23 May 2011]. Moody's Analytics (2011). Basel III New Capital and Liquidity Standards FAQ's [Online] Available at: http://fermat.eu/downloads/basel-iii-faq.pdf [Last Accessed 1 May 2011]. Nier, T & Tiesset, M (2005). Liquidity, Banking Regulation and the Macroeconomy [Online] Available at: http://www.bis.org/bcbs/events/rtf05AspachsNierTiesset.pdf [Last Accessed 01 May 2011]. Otker,R, Pazarbasioglu, C (2010), Impact of Regulatory reforms on large and complex financial institutions, Staff position Note no 2010/16 (Washington: International Monetary Fund, November). Reserve Bank of New Zealand (2009) Capital adequacy ratios for banks simplified explanation and example of calculation [Online] Available at: http://pages.stern.nyu.edu/~igiddy/articles/capital_adequacy_calculation.pdf [Last Accessed 26 April 2011]. Wignall, A & Atkinson, P (2010). Thinking beyond Basel III: Necessary solutions for capital and liquidity [Online] OEDC Journal: Financial Market Trends Available at: http://hb.betterregulation.com/external/OECD%20-%20Thinking%20beyond%20Basel%20II%20%20Necessary%20Solutions%20for%20Capital%20and%20Liquidity.pdf [Last Accessed 1 May 2011]. Woschnagg, E (2007) ICAAP Implementation in Austrian Banks [Online] Available at: http://www.oenb.at/en/img/fsr_16_special_topics_02_tcm16-95421.pdf [Last Accessed 29 April 2011]. Vento, G & La Ganga, P (2009) Bank liquidity risk management and supervision: Which lessons from the recent market turmoil? [Online] Available at: http://www.eurojournals.com/jmib_10_06.pdf [Last Accessed 1 May 2011]. Appendix 1 This approach is expressed in a formula using Cash Flow at Risk (CFaR) as a measure of the maximum expected loss expected as a deviation from the mean, with a confidence interval alpha for a defined holding period: Where ÃŽÂ± is the confidence interval in which the cash flow at risk will not be exceeded by the maximum loss, CF is the cash flow with left tail confidence interval alpha, and is the cash flow in the reference case (typically the mean of the stochastic distribution). From CFaR, a further risk indicator can be drawn, namely the Liquidity at Risk (LaR) which can be defined as the proportion of the available liquidity that remains with the firm after CFaR has been entirely subtracted from the formula: Where available liquidity is defined as the amount of liquidity can be raised with the level of risk aversion of the bank is willing to endure. Another facet of liquidity risk is market risk, it should be noted that in the standard VaR models are typically based on the assumption of normal markets, and also assumes that any quantities of securities can be traded without influencing markets prices