Stress testing is the practice of assessing a bank's resilience to severe but plausible adverse scenarios. It sits at the centre of modern risk management because standard risk metrics — VaR, credit ratings, probability of default — are calibrated to normal market conditions and say very little about what happens when conditions become extreme. The 2008 financial crisis demonstrated this gap with devastating clarity: VaR models that had been validated against years of relatively calm data gave no warning of the losses that materialised when credit markets froze, correlation structures collapsed, and liquidity evaporated simultaneously.

Regulatory Stress Tests

The most visible stress tests are those run by regulators and applied to the largest banks. In the EU, the European Banking Authority (EBA) runs a biennial stress test of significant institutions, applying a common adverse scenario — developed in collaboration with the ESRB, the ECB, and national competent authorities — to each bank's balance sheet and trading book. The results are published bank by bank, providing transparency into how individual institutions would perform in a severe downturn. Banks that show capital ratios falling below certain thresholds may be required to raise capital or restrict dividends.

In the US, the Federal Reserve runs the Comprehensive Capital Analysis and Review (CCAR) and Dodd-Frank Act Stress Test (DFAST) processes, which impose similar requirements on US bank holding companies. In the UK, the Bank of England's Prudential Regulation Authority runs annual stress tests for the major UK banks, including both a cyclical scenario and an exploratory scenario designed to identify structural vulnerabilities. These regulatory stress tests have become a major driver of capital planning and balance sheet management at the banks subject to them.

For a capital markets bank, the regulatory stress test is operationally intensive. The trading book stress — applying large, correlated moves in equity markets, credit spreads, interest rates, FX, and commodity prices simultaneously — requires the risk infrastructure to be capable of repricing the entire portfolio under the scenario. The stress test also covers counterparty credit exposures, calculating what happens to CCR exposures if the scenario is severe enough to move them significantly.

Internal Stress Testing

Beyond regulatory stress tests, banks run their own internal stress testing programmes as part of their risk management and capital planning processes. Internal stress tests typically encompass three types of scenarios.

Historical Scenarios

Historical scenarios replay the market moves observed during past crises and apply them to the current portfolio. The most commonly used historical scenarios in capital markets stress testing include:

  • 2008 Global Financial Crisis: Credit spread widening, equity market collapse, interbank market freeze, sharp FX moves. The defining stress scenario for credit, rates, and equities desks.
  • 2020 COVID-19 Shock: Extreme equity volatility (VIX reaching 85), credit spread blowout, oil price collapse, dollar liquidity squeeze. Particularly relevant for cross-asset portfolios.
  • 2022 LDI Crisis: Rapid, large gilt yield rises driven by the UK mini-budget, creating margin calls for liability-driven investment funds and severe dislocation in long-dated sterling rates markets. Highly relevant for rates desks with UK gilt exposure.
  • 1998 LTCM / Russia Crisis: Extreme liquidity premium, correlation breakdown between previously stable relationships, forced deleveraging.
  • 2011 European Sovereign Debt Crisis: Peripheral sovereign spread widening, Italian and Spanish bond market stress, bank funding pressures.

Historical scenarios have the advantage of being grounded in actual market behaviour — they are not hypothetical constructs. Their limitation is that they replay the past: they cannot capture stress events that have not yet occurred, and a portfolio structured to have survived previous crises may be acutely vulnerable to a new type of stress.

Hypothetical Scenarios

Hypothetical scenarios are constructed by the risk management function to capture plausible future stress events that are not represented in historical data. Examples include: a sudden, sharp rise in US interest rates driven by a loss of confidence in US fiscal sustainability; a cyberattack on financial market infrastructure; a rapid transition to net zero causing stranded-asset losses in fossil fuel exposures; a geopolitical event causing severe commodity price dislocation.

Constructing hypothetical scenarios requires judgment about plausibility and severity. Scenarios that are too mild do not provide useful stress information; scenarios that are implausibly severe are dismissed by management as irrelevant. The governance of scenario selection — who decides which scenarios to run, how severe they should be, and how they should be updated — is itself a risk management question. Regulators expect the scenario library to be reviewed at least annually and updated to reflect changes in the macroeconomic environment and the bank's own risk profile.

Sensitivity Tests

Sensitivity tests apply a single-factor or limited-factor shock — a 100 basis point parallel shift in the interest rate curve, a 20% fall in equities, a 200 basis point widening in investment-grade credit spreads — without constructing a full macroeconomic narrative. They are used to assess the immediate P&L impact of specific market moves and to identify concentrated positions that would suffer disproportionate losses under particular scenarios.

Reverse Stress Testing

Reverse stress testing inverts the usual logic. Instead of asking "what losses would this scenario cause?", it asks "what scenario would cause losses large enough to threaten the bank's viability?" The output of a reverse stress test is a description of the scenario — the combination of market moves, credit events, and operational failures — that would breach regulatory capital thresholds or cause the bank to fail.

Reverse stress testing is required by the PRA and is a powerful tool for identifying concentration risks and tail risks that standard scenarios might miss. If the reverse stress test reveals that viability failure requires a relatively moderate scenario — one that is not implausibly extreme — this signals that the bank's capital buffers or its risk concentrations require attention.

ICAAP and ILAAP

Stress testing feeds directly into the Internal Capital Adequacy Assessment Process (ICAAP) and the Internal Liquidity Adequacy Assessment Process (ILAAP). The ICAAP is the bank's own assessment of the capital it needs to hold, taking into account all risks that are material to the business. Stress test results are a central input: the ICAAP must demonstrate that the bank would remain above minimum capital requirements even under the adverse scenarios it has tested.

The ILAAP performs the same function for liquidity: the bank must demonstrate that it holds sufficient liquid assets and has sufficient liquidity resilience to survive a severe liquidity stress — typically a combination of market-wide and firm-specific stress — over a defined survival horizon. Stress testing of the liquidity position — modelling the outflows that would occur under stress, the drawdown of committed facilities, and the ability to monetise the liquid asset buffer — is a core component of the ILAAP.

The PRA reviews the ICAAP and ILAAP annually and uses them to set the firm's Pillar 2 capital requirement and its individual liquidity guidance (ILG). A well-constructed stress testing programme, with plausible and severe scenarios, robust governance, and a clear link to capital planning decisions, supports a lower Pillar 2 requirement. A programme that is formulaic, uses mild scenarios, or is not genuinely used for management decision-making attracts additional PRA capital charges.