Liquidity Risk Management

Relevance of liquidity risk 

One debilitating consequence of the recent subprime crisis was a sharp decrease in market liquidity; as banks sought to bolster cash reserves, money market rates increased and trading volumes decreased. This lack of liquidity – and the vast sums of sovereign cash injected into markets to alleviate the problem – has brought the associated risks to the forefront of regulatory authorities’ priorities, and to the attention of the public in general. 

A synopsis

A significant rise in subprime-class mortgage defaults in America led to a large number of loan write offs, ultimately resulting in several banks’ insolvency. The impact of these write-offs wasn’t confined to American financial markets: incorporation of these assets into asset-backed securities, together with the global sale of these wrapped securities, meant that few global economies were left unscathed.

As markets for asset-backed securities dried up, the resulting illiquidity forced investment banks and hedge funds to book significant losses (indeed, the crisis has resulted in the downfall of numerous hedge funds). A general flight to safety ensued, with credit risks being deserted for the safe haven of government securities. Caution amongst the banks over future losses – both internally and externally – caused money markets to dry up and rates to increase. The effects of such liquidity crises should not be underestimated: a downward spiral consisting of a loss in creditworthiness and growing liquidity problems was a sufficiently strong pairing to cause a complete loss of public confidence in UK mortgage bank Northern Rock plc; the resulting run on the bank and government rescue represented an early nadir in the subprime crisis.

The correlation and globalisation of the financial world made a primarily American problem into a world one. In Germany, for instance, the Landesbanken (the regional state banks) were particularly exposed to the American subprime markets and several banks have received varying levels of government support: SachsenLB and IKB Deutsche Industriebank both suffered fundamental problems (eventually leading to both being bought out) whilst BayernLB and WestLB received financial support from state and central governments together with state guarantees for their creditworthiness. Despite ECB intervention, the lack of money-market liquidity also caused problems for several German banks. Hypo Real Estate suffered from severe liquidity problems due to Irish subsidiary Depfa’s inability to finance liabilities through borrowing.

Affected countries generally responded through government and central bank action. Governments responded to credit downturns and liquidity issues through the introduction of economic stimulus packages, aiding the real economy. Central banks took two approaches. Firstly hundreds of billions of Euros, US Dollars and Pounds Sterling were injected into stagnated markets, ‘jumpstarting’ the lending and borrowing services vital to financial institutions. Secondly, fears over banks’ creditworthiness were assuaged though state guarantees. The immediate consequences of these activities resulted in the effective nationalisation of several banks (in particular, the UK’s Northern Rock, Fannie Mae and Freddie Mac in the US, and the aforementioned Hypo Real Estate plus IKB in Germany). In the UK, the Government is now the majority owner of both RBS and Lloyds Banking Group – as Lloyds TSB, the latter was forced to take over the ailing HBOS, much to the chagrin of existing shareholders. The borrowing restrictions and later losses in the equity markets have had a substantial impact on the real economy, leading to recession in many countries.

Closer and more rigorous supervision

A consequence of the recent crisis is closer supervision and a tighter regulatory regime to be imposed upon the banks and financial markets by Government-sponsored regulatory authorities.

The USA (presented in March 2009) and UK have presented first drafts of updated regulatory requirements and there is a general push from other countries to update and revise their own regulatory requirements.

In general, a trend away from standardised models towards institution-specific liquidity-risk models can be observed. One example is the opening clause of the LiqV (“Liquiditätsverordnung”) liquidity regulations in Germany, allowing the use of internal liquidity models. The September 2008 revision of the Basel II accord (International Convergence of Capital Measurement and Capital Standards) emphasises the importance of liquidity risk measurement. In particular, the following requirements are stressed:

• Group-wide consideration of liquidity risk, involving all subsidiaries and related companies, regardless of the consolidation according to accounting standards.
• Incorporation of FX risk into the stress-test scenarios to evaluate difficulties such as currency conversion.
• Definition of stress scenarios; incorporating intrinsic, idiosyncratic aspects of the institution’s risk and also system-wide crises into account.

In November 2008 the FSA (the UK regulatory authority) specified new regulatory requirements in a series of consulation papers (CP08/22, CP09/13 and CP09/14); these requirements were primarily concerned with the measurement and management of liquidity risk. The papers detail new requirements such as the Individual Liquidity Adequacy Standards (ILAS) or the Liquidity Reporting. It is likely that these documents will influence any updates to the regulatory requirements in Germany and other European countries. Crucial points of ILAS are:

• multidimensional breakdown of contracts (for example, currency, asset type or time buckets)
• at least 3 main stress-test scenarios (idiosyncratic-institution stress, market-wide stress or a combination of both) evaluated across 10 key risk drivers
• interpretation of results in a coherent manner
• new reporting regime: granular, frequent (daily, weekly, monthly, quarterly) and partially automated – the “Enhanced Mismatch Report” has to be submitted weekly (with the ability to report daily) in an automated process.

Recent updates of the MaRisk (regulatory requirements in Germany) in summer 2009 reveal the lessons learned through the financial crisis. The following innovations can be found:

• Specification of the stress scenarios that, as a minimum, have to be considered in the treatment of liquidity risk.
• Updated requirements for the provision of liquidity reserves.
• Separate analysis of liquidity per currency.

In general, the updated MaRisk requirements (regarding the coverage and the degree of specification) are significantly less stringent than those released by the FSA.

Recent developments have focused on the applied models and scenarios, the re-definition of banks’ roles, and the improvement (primarily developments from software providers) in the measurement of liquidity risks.

What is liquidity?

The term liquidity is used in the financial world in different contexts:

• liquidity as a measure of the saleability of securities such as bonds or shares
• liquidity as a description of the financial solvency of individual institutions
• liquidity as a level of market activity
• liquidity as unhindered cash flows within an economy.

The primary objective of liquidity risk management remains the same: to ensure financial solvency at all times.

To successfully manage this risk, one should consider all relevant factors: from the business structure which determines liquidity needs, the analysis of markets (market price and market liquidity), and finally the necessary level of funding diversification.This makes liquidity management a very complex and comprehensive topic.

Within the risk management, liquidity risk takes on a unique position; since, unlike other types of risk (market risk, credit risk, operational risk etc) it cannot be covered entirely by regulatory capital requirements. Of equal importance to the amount of capital available is the manner in which it is invested (for instance, the structure of the assets). In this sense a bank, invested heavily in government bonds, is significantly more liquid than a similar bank with investments in 10-year customer loans.

Liquidity management

An important figure in liquidity management is the liquidity gap analysis. This analyses maturity mismatches, displaying the cash flows and placing them in sets (or ‘buckets’) according to their timings. The liquidity required at any one particular time can then be determined.

When considering liquidity risk one can consider (dynamic) or discard (static) new business or deal extensions; both are useful. Static representation of portfolios highlights issues such as a lack of funding on three-month money market credits – such funding problems would not be shown with a dynamic view due to the prolongation assumptions. The dynamic presentation is more realistic: here, expected new business is included when considering payment obligations and claims.

To ensure institutions are not left exposed by cashflow timings the so-called counter-balancing capacity is calculated. This is the liquidity that can be made available through certain actions (sales, engagement in repos etc.). The counter-balancing capacity is composed of potential inflows, e.g. from sales/lending of securities; the filing of securities at the central banks; the health of collateral credits in funds, and finally out of new public issues / term borrowing at money markets etc.

Results and conclusions from liquidity gap and counter-balancing capacity analyses are further strengthened by stress-testing economic scenarios. Defining these stress-test scenarios and the corresponding early warning indicators, together with the modelling of non-deterministic cash flows (current accounts, loan commitments, contracts with rights of cancellations, options, shares etc.) is a significant challenge.

The following approaches are taken when modeling non-deterministic products:

• Simple and pragmatic. This relies on observed historical values, e.g. average value over past years. Risks are then assessed using semi-heuristic forecasts.
• Econometric. Models cash flows with established time-series models. Forecasts are created through statistical analysis of results. 
• Simulation-based. Econometric and financial motivations are harnessed to create Monte-Carlo simulations, through which risks are assessed.

All approaches hold advantages and disadvantages (accuracy versus speed, complexity versus stability) and the most suitable method will depend on available resources.

Incorporated into the liquidity risk analyses required by regulators are regulations (such LiqV and MaRisk in Germany) requiring analysis of funding sources. The main aim is to detect any potential concentration risks within funding sources.

Liquidity gap reports as well as counterbalancing capacities are solely based on the expected cash flows of transactions. Other liquidity analyses are desirable and advisable; for instance the potential costs arising from any reduction in liquidity gaps or the impact new business would have on bank P&L. Incorporation of these considerations allows the assimilation of liquidity risk into a wider, general risk framework within a bank; such an assimilation is not possible through simple cash-flow analysis.
 
Advanced approaches

The basic methods outlined above are typically based on simple assumptions and do not venture far into the stochastic modelling world of mathematical finance. (For instance, mean values of credit line draws are used instead of mean reverting random walks.)

More complex stochastic approaches such as Liquidity at Risk (maximum liquidity gap within a certain time frame and for a given confidence level) or Liquidity Value at Risk (maximum cost of liquidity under certain assumptions) do exist. However these are not broadly accepted, due to issues with complexity or more fundamental problems concerning the variables modelled. For instance, a stochastic approach to LaR traces the liquidity gap but not the causes for gaps. In contrast, a more econometric-inspired, stress-test argument developed from the fundamental causes of liquidity gaps is more intuitive and perhaps more parsimonious. The stochastic approach to LaR calculation is also hindered due to the relevance of statistical outliers (suggesting extreme value statistics would be useful). Academically, this stochastic approach is well developed; however, practitioners remain cautious or unwilling and practical implementations are few.

Liquidity Risk Management - What we can do for you

Our experts can help you quantify and manage your liquidity risk. How advanced and up to date is your liquidity risk management? Which stress scenarios have to be considered? How should the non-deterministic products be modelled? How do you calculate the Liquidity at Risk?

Utilising our comprehensive experiences with liquidity risk management we offer:

• brief checks of your models and processes by skilled experts
• design and development of models for internal management and controlling, as well as for regulatory reporting
• project execution from one source: starting from the selection of appropriate software through system integration to the definition of management reports.

Benefit from our extensive experience in liquidity risk management.

We would be delighted to provide further information on our approach, together with our references.

Please contact:

Dr Stefan Hengstmann
Tel.: +852 3711 5800

d-fine Zurich