Next-Generation System-Wide Liquidity Stress Testing

The global financial crisis has shown that neglecting liquidity risk comes at a substantial price.

This column presents a new framework to run system-wide, balance sheet data–based liquidity stress tests. The liquidity framework includes a module to simulate the impact of bank-run type scenarios, a module to assess risks arising from maturity transformation and rollover risks, and a framework to link liquidity and solvency risks.

Bank liquidity was traditionally viewed as of equal importance to solvency. Liquidity risks are inherent in maturity transformation, ie the usual long-term maturity profile of banks’ assets and short-term maturities of liabilities. Banks have commonly relied on retail deposits, and, to some degree, on long-term wholesale funding as supposedly stable sources of funding. Yet, attention to liquidity risk diminished in recent decades and was symbolised by the absence of consideration of liquidity risk in the 1988 Basel I framework (Goodhart 2008).

The global financial crisis has clearly shown that neglecting liquidity risk comes at a substantial price. Over the last decade, large banks became increasingly reliant on short-term wholesale funding (especially in interbanking markets) to finance their rapid asset growth. At the same time, funding from non-deposit sources (such as commercial paper placed with money-market mutual funds) soared. With the unfolding of the global financial crisis, when uncertainties about the solvency of certain banks emerged, various types of wholesale funding market segments froze, resulting in funding or liquidity challenges for many banks. In the light of this experience, there is now a widespread consensus that banks’ extensive reliance on deep and broad unsecured money markets is to be avoided (and in current market conditions there is no appetite for that anyway). Creating substantial liquidity buffers across the board is the explicit aim of a number of regulatory responses to the crisis, such as the Committee of European Banking Supervisors Guidelines on liquidity buffers (CEBS 2009) as well as the forthcoming Basel III liquidity standards, the Liquidity Coverage Ratio and the Net Stable Funding Ratio.

Literature on liquidity stress testing

One of the early adopters of cash flow–based liquidity stress testing in recent years has been the Austrian National Bank (OeNB 2008), or more recently Schmitz (2009 and 2010), whose work has heavily influenced the European approach as well (eg ECB 2008, and the first EBA liquidity risk assessment 2011).

Van den End (2008) at the Dutch Central Bank developed a stress-testing model that tries to endogenise market and funding liquidity risk by including feedback effects that capture both behavioural and reputational effects. Wong and Hui (2009) from the Hong Kong Monetary Authority sought to explicitly capture the link between default risk and deposit outflows. An attempt to (fully) integrate (funding) liquidity risks and solvency risk is the Risk Assessment Model for Systemic Institutions, developed by the Bank of England (Aikman et al 2009).

At the IMF, in the context of Financial Sector Assessment Programs, stress tests, liquidity tests were originally centred on Čihák (2007) using bank balance sheet data to perform bank run–type stress tests on a bank-by-bank level while a recent chapter of the Global Financial Stability Report (IMF 2011) focused on systemic liquidity and Barnhill and Schumacher (2011) develop an empirical model that seeks to link solvency risk and liquidity risks.

On the regulatory side, substantial microprudential efforts were undertaken to contain liquidity risks on a bank-by-bank level. In 2008, the Basel Committee of Banking Supervision (BCBS 2008) and the Committee of European Banking Supervisors (CEBS 2008) published qualitative regulation for sound liquidity risk management. An overhaul of the quantitative regulatory framework followed in December 2010 (BCBS 2010), when the Committee introduced two measures to contain short-term vulnerabilities on the one hand and excessive maturity mismatch on the other.

Finally, in the industry bank-level tests are centred on maturity-mismatch approaches, sometimes complemented by stochastic value-at-risk components for those funding sources for which sufficient histories of high frequency data is available.

Our approach

The liquidity stress-testing framework presented in Schmieder et al (2012) was developed in the context of recent Financial Sector Assessment Programs1 and IMF technical assistance especially in Eastern Europe, extending the seminal work of Čihák (2007), and drawing upon work at the Austrian National Bank (OeNB 2008). It was initiated in a complementary solvency stress-testing tool by Schmieder et al (2010).

While developing the framework, five key facts were accounted for:

  • The availability of data varies widely;
  • Liquidity risk has several dimensions and assessing banks’ resilience vis-à-vis funding risks requires multidimensional analysis;
  • Designing and calibrating scenarios is more challenging than for solvency risks, mainly as liquidity crises are relatively rare and originate from different sources;
  • There is a close link between solvency and liquidity risks;
  • And while the paper and tool present some economic benchmark scenarios, these scenarios as well as economic and behavioural assumptions used for the tests should depend on bank- and country-specific circumstances, and current circumstances (iethe level of stress), among others.

The answer to these multiple dimensions is a framework that is an Excel-based, easy-to-use balance sheet–type liquidity stress-testing tool that allows running bottom-up tests for hundreds of banks.

First, the tool can be used to run some basic tests in circumstances where data is very limited to broad asset and liability items. Likewise, a cash flow–based module allows running detailed liquidity analysis like those carried out by banks for the internal purposes but again can be adapted to a more limited data environment. Second, the framework includes three broad dimensions that allow for complementary views on liquidity risks, including the link to solvency risks. Third, the paper provides benchmark scenarios based on historical evidence on the one hand and common scenarios used by IMF missions on the other. Fourth, the framework allows assessing the link between liquidity and solvency, albeit additional effort is needed in this context, including work that captures dynamic aspects of this relationship and spillover effects such as dynamically examining the link from liquidity to solvency concerns. As such, the framework is meant to provide users with the possibility to run a wide range of idiosyncratic and market-wide liquidity stress tests for individual banks and entire banking systems within a relatively short period of time, but can also be used for monitoring purposes.

It is vital to bear in mind that the key benefit of the approach is to benchmark banks against one another, ie to run peer comparisons and thereby assess their relative vulnerability to different shocks. Whether and how a shock materialises depends on the various factors, with behavioural aspects increasingly playing an essential role.2 Hence, it is also acknowledged that regular liquidity stress testing is not a panacea for a qualitative judgement by policymakers in order to complement findings even from well-designed liquidity stress tests. To be effective, rigorous quantitative liquidity risk management must be integrated with the qualitative requirements defined by the Basel Committee of Banking Supervision (2008) and by the Committee of European Banking Supervisors (2008). We also stress that scenarios should be interpreted as tools to condense a wealth of information and assumptions (much rather than a single benchmark against which resilience is established). To avoid a false sense of security, we recommend running a whole range of them with varying degrees of severity (including reverse stress tests). That allows, both supervisors and banks to get a better understanding of a key concept in liquidity risk supervision/management – their respective liquidity risk tolerance.

While cash flow–data reporting, for instance, will become mandatory in the European Capital Requirements Directive IV regulation given the Basel III implementation, it is (for now) still rarely available at regulatory/ supervisory institutions so we follow a two-pronged approach, distinguishing between implied cash-flow tests and a ‘real’ cash-flow approach3, thereby seeking to lift liquidity tests to a next-generation level.

The framework consists of three elements:

  • Stress testing funding liquidity based on an implied cash-flow approach, with two different components: (a) a tool to simulate bank run–type scenarios while accounting for fire sales of liquid assets and/ or central bank liquidity provision subject to eligible collateral and haircuts;4 and (b) a liquidity gap analysis module that matches assets and liabilities for different maturity buckets under different stress assumptions, including rollover risk; the tool also allows for calculating (simplified)5 Basel III liquidity ratios.
  • Cash flow-based liquidity tests – running this module ideally requires detailed data on contractual cash flows for different maturity buckets and behavioural data based on banks’ financial/funding plans.6 The calibrated scenarios then denote rollover assumptions for contractual cash outflows and cash inflows. The former focus on funding risk and the latter take into account the banks’ objective to maintain its franchise value even under stress. Accordingly, the module allows for an intuitive view of each banks’ liquidity risk–bearing capacity in the form of the cumulated counterbalancing capacity at the end of each maturity bucket. In addition to stress testing, the module is also meant to be used for liquidity monitoring purposes, for which behavioural cash flows are particularly informative.
  • Tests linking solvency and liquidity risk – the tool allows linking liquidity and solvency risk from three complementary perspectives. First, the module allows simulating the increase in funding costs from a change in solvency, indicated by a change in a bank’s (implied) rating. Second, the tool enables simulating the partial or full closure of funding markets (both long and short-term) depending on the level of capitalisation with or without considering solvency stress. Third, it allows examining the potential impact of concentration in funding and a name crisis (eg, from parent banks) on banks’ liquidity positions.


One of the main contributions of this project consists in providing input templates for cash flow–based tests that could also serve regulators/supervisors as a first step towards fully fledged cash-flow analysis based on a regular data collection from banks. Once available, the cash-flow module allows simulating detailed funding structures of single banks, which enables to draw some broader conclusions for the system-wide situation of banks and potential contagion effects, respectively. Moreover, the presented tool allows for easy peer comparisons that should always play an important role for liquidity stress tests and can readily reveal vulnerabilities. Finally, this project contributes to existing work on liquidity by modelling the link to solvency stress (tests) explicitly.

Future research will focus on better understanding the link between banks’ solvency and liquidity strains. Both are inherently interrelated and standalone stress tests that only examine either solvency or liquidity stress testing potentially risk producing downward-biased results. For example, a bank’s severe funding strain could swiftly mutate into solvency concerns with the market putting pressure on the bank to increase its capital. The focus here has been predominantly to analyse the link from solvency to stress testing but the feedback loop can also originate with liquidity.

Disclaimer: The views expressed in this column are those of the author(s) and do not necessarily represent those of the IMF/OeNB or IMF/OeNB policy. Working Papers describe research in progress by the author(s) and are published to elicit comments and to further debate.


Aikman, David, Piergiorgio Alessandri, Bruno Eklund, Prasanna Gai, Sujit Kapadia, Elizabeth Martin, Nada Mora, Gabriel Sterne and Matthew Willison (2009), Funding liquidity risk in a quantitative model of systemic stability, Bank of England Working Paper no. 372.

Barnhill, Theodore Jr and Liliana Schumacher (2011), “Modeling Correlated Systemic Liquidity and Solvency Risks in a Financial Environment with Incomplete Information,” IMF Working Paper 263.

Basel Committee on Banking Supervision (BCBS) (2010), Basel III: International framework for liquidity risk measurement, standards and monitoring, December.

Čihák, Martin (2007), “Introduction to Applied Stress Testing”, IMF Working Paper 59.

Committee of European Banking Supervisors (2008), CEBS’s technical advice on liquidity risk management (2nd part), London.

Committee of European Banking Supervisors (2009), Guidelines on liquidity buffers and survival periods, London.

European Central Bank (2008), “EU banks liquidity stress tests and contingency funding plans”, Eurosystem.

Goodhart, Charles (2008), “Liquidity risk management”, Special issue on liquidity No. 11, Banque de France Financial Stability Review, February.

IMF (2011), ‘How to address the systemic part of liquidity risk?’ Chapter 2, Global Financial Stability Report, April.

Oesterreichische Nationalbank (OeNB, Austrian Nationalbank), 2008, Financial Stability Report 16. December.

Schmieder, Christian, Heiko Hesse, Benjamin Neudorfer, Claus Puhr, and Stefan W Schmitz (2012), “Next Generation System-Wide Liquidity Stress Testing,” IMF Working Paper 3.

Schmieder, Christian, Puhr, Claus and Maher Hasan (2011), “Next Generation Balance Sheet Stress Testing”, IMF Working Paper no. 11/83.

Schmitz, Stefan W (2009), “Central banks and systemic liquidity risk: the role of stress testing”, presented at the 6th Annual RiskCapital Conference, Brussels, June 2009.

Schmitz, Stefan W (2010), “Liquidity Regulation and Stress Tests”, presented at RiskMinds, Geneva, December.

Van den End, Jan Willem (2008), “Liquidity Stress Tester: A macro model for stress-testing banks’ liquidity risk”, Dutch National Bank Working Paper No. 175, May.

Wong, Eric and Cho-Hoi Hui (2009), “A liquidity risk stress- testing framework with interaction between market and credit risks,” Hong Kong Monetary Authority Working Paper 06/ 2009.