Financial risk methodologies have evolved over time in response to successive financial crises revealing their limitations. Value-at-Risk (VaR) represented an early quantitative approach, but its short-term horizon, reliance on historical data and assumption of relatively stable covariances made it poorly suited for periods of stress.
The Economic Capital framework integrated market, credit and operational risks, yet its one-year horizon did not adequately capture through-the-cycle consequences – a limitation exposed during the Global Financial Crisis (GFC). Its reliance on historical performance and point-in-time approach to stress testing contributed to financial institutions having their lowest capital cushions immediately before the crisis.
Long-term, forward-looking scenario analysis introduced a more comprehensive way to assess institutional resilience by incorporating macroeconomic variables into scenario projections. Conventional stress testing, however, still relies on a relatively small number of expert-selected scenarios, leaving it vulnerable to judgment and scenario-selection bias.
While capital and liquidity requirements proved adequate through the pandemic, the 2023 U.S. regional banking crisis exposed yet another weakness of traditional scenario analysis. It didn’t adequately capture critical combinations of interest-rate risk, balance-sheet exposures, liquidity, depositor behavior and social-network dynamics.
This crisis produced renewed calls for institution-specific reverse stress testing. Existing recovery and resolution regulation has largely been a compliance exercise rather than a source of critical early-warning signals and mitigation strategies. While it is often viewed as reverse scenario analysis, it looks a lot less like recovery (how to get out of crisis) and more like resolution (how to handle default).
Change in Approach
A more useful framework requires three linked steps: reverse stress testing asks how a bank reaches distress; recovery planning asks how it returns to viability; and resolution planning asks what happens if it cannot.
Alla Gil
The ECB’s 2026 reverse stress test represents an important step in this direction. Instead of asking, “Can the bank survive our scenario?”, it asked, “What combination of events could produce a specified capital loss – and can the bank explain the entire path there?”
The 110 participating banks were given a target of a 300-basis-point depletion in Common Equity Tier 1 (CET1) and asked to construct plausible, bank-specific geopolitical narratives capable of producing that outcome. Published findings focused on the quality of scenario discovery, cascading effects, governance and contingency planning rather than name-by-name capital resilience.
The ECB highlighted weaknesses in the granularity and sensitivity of risk assessments, consistency between scenario narratives and solvency/liquidity impacts, and practicality of mitigating actions. Bank-specific scenarios incorporated nonfinancial risks including cyberattacks, operational-risk events and third-party service disruptions. In other words, the exercise pushes stress testing away from a single macroeconomic story and toward an integrated map of how a bank can actually get into trouble.
This points toward the next generation of risk management: supplementing conventional stress testing with reverse stress testing and exhaustive scenario analysis that systematically explores a much broader range of interacting risk drivers.
Back-Test the Stress Test
To verify that the generated scenarios encompass unprecedented shocks, stress testing itself should be backtested. Using only information available before an unexpected market crash, the test should determine whether the generated scenario set included such a crash with a reasonable probability. When balance sheet segments are projected on all these scenarios, probabilities of the outcomes (e.g., institutions’ capital and liquidity ratios) are naturally obtained from their empirical distributions. Even a 1% or 0.5% probability of capital or liquidity deficiency warrants management attention and preparation.
Without full-range scenario analysis incorporating dynamic correlations and cascading effects, events such as the GFC, U.K. pension crisis and Silicon Valley Bank (SVB) failure appeared to be “black swans.” Yet exhaustive, path-dependent analysis of plausible combinations of shocks and their consequences can reveal the snowballing nature of such events. Their exact timing cannot be anticipated, but vulnerabilities can be discovered well in advance, creating time to develop mitigation strategies.
To address weaknesses identified by the ECB exercise, several improvements are needed.
Scenario detail must be adequate rather than merely dramatic. Multiple combinations of events that can trigger one another should be considered. A geopolitical shock in the Middle East may cause oil prices to surge, but other possible shocks and interactions matter as well. While early-2022 inflation did not directly cause the crypto crisis later that year, the combination of these two events without timely management action directly caused SVB’s failure.
When thousands of scenarios containing relevant macroeconomic and market drivers are generated across combinations of shocks and cascading effects, financial and behavioral KPIs can be projected consistently across them. This produces internally consistent distributions of capital, liquidity and profitability because they are constructed from the same underlying scenarios.
Thus, a 300-basis-point CET1 decline is not necessarily the right hurdle for every institution. A bank with abundant liquidity may be able to correct such a problem quickly. By contrast, a 250-basis-point CET1 decline combined with severe liquidity pressure could create a much more dangerous situation.
Consistency across scenarios and KPIs is therefore critical. Hurdles should reflect each institution’s risk capacity and inform its corresponding risk appetite.
From Scenarios to Mitigation
Full-range scenario analysis can also identify early-warning indicators (EWIs) showing when risk appetite limits approach risk capacity and should be reviewed. These indicators may vary significantly by institution and by the relevant portion of a KPI distribution. Risk drivers associated with outcomes near the center of a distribution may be entirely different from those affecting its tails.
Reverse stress testing asks how the bank reaches distress; recovery planning should return it to normal. Proactive mitigation should aim to prevent the need for compliance-driven resolution.
Such mitigation strategies cannot be developed around a single scenario. Multiple paths may lead to similar adverse outcomes, with each requiring a different response. Moreover, mitigating one scenario can inadvertently create a new concentration of hidden risk. Exhaustive analysis therefore must evaluate management actions’ effectiveness across the full range of scenarios.
Parting Thoughts
The suggested framework allows for consistent approach across existing regulations. Basel capital rules and FRTB (Fundamental Review of the Trading Book) provide standardized and model-based measures for specific risk types. ECB-style reverse stress testing searches beyond the comfortable expectations for the scenarios capable of breaking a bank’s assumptions.
Exhaustive scenario analysis then provides assurance that all plausible risks were considered and determines whether those vulnerabilities are mitigated by management actions. It enables organizations to identify the risks that matter before historical data, regulatory formulas or familiar scenarios make them obvious.
A sophisticated scenario engine is not enough if management cannot explain why a scenario is relevant, recognize its early-warning signals, execute credible actions and understand the side effects of those actions.
The organizing principle should be exhaustive scenario analysis: Do not try to predict the one crisis that will happen. Explore enough combinations to discover the different routes by which the same adverse KPI outcome can occur.
Alla Gil is co-founder and CEO of Straterix, which provides unique scenario tools for strategic planning and risk management. Prior to forming Straterix, Gil was the global head of Strategic Advisory at Goldman Sachs, Citigroup, and Nomura, where she advised financial institutions and corporations on stress testing, economic capital, ALM, long-term risk projections and optimal capital allocation.