Liquidity management has to be considered under a broader view than simple liquidity gaps, which are necessary but not sufficient. The framework for addressing liquidity management, notably under stressed conditions, is now expanding. Several publications address the issue. They include the FSA report of 2009 [31], the draft document of BCBS in 2008 [6]. The publication in 2003 on "liquidity black holes" [62] addressed the difficulty earlier.

Liquid Assets

Banks publish the level of liquid assets serving as a cushion against market disruption. In order to be meaningful, the amount of liquid assets should be related to the periodical funding over the next period, allowing one to find out for how long the bank can sustain a market disruption. A review of annual reports for 2006 showed that banks published periods of one month up to one year. The financial crisis showed that most banks could not sustain such periods of liquidity crunch. This period relies on many assumptions with respect to projected net outflows. It should be calculated based on stressed assumptions, for example in inter-bank lending and borrowing, to be meaningful.

Borrowing long to hold short-term assets is uneconomical. A more appropriate source of financing liquid assets consists of using repos, which allow the bank to isolate the cost of funding liquid assets from the cost of funding of the bank. The repo cost is cheaper than the bank's funding cost. Furthermore, the repo cost matches the reset dates of the interest rate of liquid assets. The technique is used for financing hold-to-maturity securities.

Liquidity Crises and Stress Test Scenarios

An unexpected release of information might cause normal funding sources to cut down their credit lines. A block in the national or global financial markets could result in a system-wide "run on deposits"; a liquidity crunch caused by some failures; fear of contagion effect to other financial institutions causing an increase of risk aversion and reluctance to lend. There are several examples of such crises:

• the Russian debt crisis

• the "Y2K" fear of information systems failure

• the 9/11 crisis, and, of course

• the sub-prime crisis of August 2007, which showed that a freeze of the most liquid markets could effectively happen.

For addressing such liquidity crises, stress testing is the appropriate methodology. Stress testing consist of simulating what could happen in a worst case event. The possible origins of a crisis and the historical events provide some examples of such factors. In the event of a system-wide crisis, the unique safety cushion against market disruptions is liquid assets because they always provide a source of funding. Liquid assets are also the source of repo transactions, widely used in the financial system, though which securities are used for collateral-based financing.

Note that liquidity crises might be triggered for a single bank, and not necessarily be system-wide. Examples of such events include unexpected and sudden losses for a single bank that threatens its solvency, disrupting the willingness to lend by other financial players. A downgrade of the bank's rating can also have major effects on both the asset and the liability sides. The cost of funding will increase and the bank might become not eligible for further lending by other institutions which follow lending rules based on a minimum ratings. The same cause can trigger margin calls, through which the bank is supposed to post more collateral in its debt or, alternatively reduce its debt. On the asset side, the bank might become ineligible for some deals when a minimum rating is required, such as providing guarantees. This illustrates the complexity of liquidity stress testing. Credit risk, solvency and liquidity issues are intertwined. A comprehensive stress test would include all such effects on liquidity, combining various scenarios and cumulating their effects on liquidity.

The Royal Bank of Scotland Statement on stress testing (Annual Report, 2006) is an illustration of how banks can define stress testing for liquidity.

The maintenance of high-quality credit ratings is recognized as an important component in the management of the Group's liquidity risk. Credit ratings affect the Group's ability to raise, and the cost of raising, funds from the wholesale market and the need to provide collateral in respect, for example, of changes in the mark-to-market value of derivative transactions.

Given its strong credit ratings, the impact of a single notch downgrade would, if it occurred, be expected to have a relatively small impact on the Group's economic access to liquidity. More severe downgrades could have a progressively greater impact but have an increasingly lower probability of occurrence.

As part of stress testing of its access to sufficient liquidity, the Group regularly evaluates the potential impact of a range of levels of downgrades in its credit ratings and carries out stress tests of other relevant scenarios and sensitivity analyses.

Royal Bank of Scotland

Such conservative rules did not prevent the RBS from being short of liquidity and from failure triggering government actions.

Contingent plans serve for assessing the ability to generate cash flows to meet loan commitments and deposit withdrawals, or margin calls on collaterals, under extreme circumstances. In order to provide some practical inputs to such plans, banks have to consider the size of their liquid asset portfolio, balance sheet and funding limits, maximum repos that can be contracted, the diversification of sources of funding and the levels of unsecured funding. Once these are identified, a bank can further diversify sources of funding, increase the base for repo financing, etc. The recent reports on sound liquidity management practices and the FSA document provide a number of criteria for addressing such contingency plans.

< Prev   CONTENTS   Next >