Informing the positive neutral countercyclical capital buffer using stress test data

Prepared by Marianna Caccavaio, Carlo Cascini, Cyril Couaillier, Giorgia De Nora, Mara Pirovano and Frances Shaw

Published as part of the Macroprudential Bulletin 32, November 2025.

Early activation of the countercyclical capital buffer (CCyB), including adoption of a targeted “positive neutral” rate (PNR), has become an increasingly common practice within the euro area and beyond.^[1] To date national authorities have used a variety of approaches to calibrate the target positive neutral CCyB rate, i.e. the level at which it should be set in an environment where systemic risks are not yet elevated (see ECB/ESRB, 2025). In most cases, they have relied on more than one quantitative approach, reflecting the notion that policy decisions are more reliably informed when supported by complementary analytical methods. Several euro area countries (Estonia, Ireland, Latvia, Lithuania, Portugal, Slovenia and Spain) use stress-testing methodologies, typically involving scenarios with standard risk conditions to estimate the buffer required to cover capital shortfalls when cyclical systemic risks are not elevated. Other countries rely instead on analysing historical losses (Cyprus, Czech Republic, Estonia, Hungary and the Netherlands) and macroeconomic models (Czech Republic and Ireland). Target rates resulting from these calibrations range between 0.5% and 2%, reflecting differences in methodologies, country-specific characteristics and policymakers’ preferences.

This Focus Piece introduces a relatively novel approach for deriving a target positive neutral rate which links bank capital losses to macroeconomic variables using stress test data. First, elasticities linking bank capital depletion and stress test scenarios are estimated through multiple regression equations that account for combinations of different variable and fixed effects (bank-level and business model). For each exercise, changes in Common Equity Tier 1 (CET1) ratios for each bank across both baseline and adverse scenarios serve as the dependent variable. Explanatory variables include macroeconomic variables, bank-specific balance sheet characteristics, and the interactions between these.^[2] Second, the average elasticities of banks’ CET1 depletion obtained are then used to estimate the losses a bank would experience under a scenario in which cyclical systemic risks are neither subdued nor elevated; this is the relevant metric for calibrating the positive neutral CCyB rate. Finally, aggregating bank-level CET1 depletion projections at the system level provides a comprehensive view of the banking system’s loss absorption requirements under such a scenario.

This stress-test elasticities approach complements existing methodologies developed by the ECB to inform calibration of the target positive neutral CCyB rate for the euro area. Existing methodologies include the “losses-to-buffer” approach (see De Nora et al., 2025), which relies on historical bank losses, and the “risk-to-buffer” approach (see Couaillier and Scalone, 2024 and Herrera et al., 2025), which is based on an empirical macroeconomic model. The former methodology uses a location-scale quantile regression in a local projection framework to disentangle losses linked to cyclical risk from those driven by other factors, while accounting for the risk metrics used to calibrate other prudential requirements. The latter approach relies on a non-linear macroeconomic model that quantifies the impact of exogenous shocks on GDP under different levels of cyclical systemic risk and uses those occurring in a median risk environment to calibrate the positive neutral CCyB rate.

Existing approaches and the one proposed here complement each other in multiple ways. First, they capture different rationales for implementing the positive neutral CCyB rate (Table 1). While losses-to-buffer methodology calibrates the target positive neutral CCyB rate to cover losses arising from unobserved factors unrelated to domestic cyclical imbalances,^[3] the risk-to-buffer and stress test approaches aim to calibrate the target rate to cover losses that may materialise in a standard risk environment. Second, unlike risk-to-buffer, both the losses-to-buffer and the stress test approaches exploit the granularity of bank-level information but do not account for macroeconomic feedback effects. Third, while the methods differ in the degree to which they account for other capital requirements, the approaches are mutually complementary and help mitigate any limitations inherent in each one individually.

Relying on banks’ performance in past EU-wide stress tests to inform the calibration the target positive neutral CCyB is extremely flexible and informative. This approach could allow macroprudential authorities to more effectively leverage the granular results of EU-wide stress tests, which are approved by European and national authorities, as well as banks. Its flexibility lies in several key aspects. First, provided the country’s banking system is adequately represented,^[4] this granular approach enables complementary country-specific assessments, which combine a macroprudential perspective with the ability to account for heterogeneity across European banks. Second, it can map the effect of individual macroeconomic variables on the target positive neutral CCyB rate, allowing insights into which factors and scenarios can play a key role in calibrating the target rate. The ability to disentangle the impact of key drivers is particularly useful for policymakers and helps assess the scenarios the positive neutral CCyB is designed to address. Furthermore, EU-wide stress tests rely on assumptions and methodological constraints such as the constant balance sheet assumption, the hypothesis that banks do not adjust their total assets in response to stress and maintain credit supply even in times of declining aggregate credit demand and rising credit risk. This assumption provides useful information on the size of macroprudential buffers that, when released, would allow banks to absorb losses under the adverse scenario while keeping credit supply unchanged.

At the same time, the stress-test-elasticities approach comes with some caveats that must be carefully taken into account. First, the scenarios employed require thoughtful consideration to ensure they are conceptually aligned with the purpose of the positive neutral CCyB and meaningfully reflect country specificities and national policy preferences. Second, policymakers must consider that, while the results capture the impact of scenarios on individual banks, they do not account for potential interactions among banks. Finally, the approach does not explicitly account for the potential interactions between macro- and microprudential buffers informed by the EU-wide stress test results. It is therefore important to cross-check results against those obtained with other approaches.

Below, a target positive neutral CCyB rate for the euro area is simulated based on scenarios designed to capture losses occurring when cyclical systemic risks are neither subdued nor elevated.^[5] They are either constructed using historical data or tailored to reflect the macroeconomic dynamics observed in the aftermath of a financial crisis that materialised after a neutral phase of the financial cycle. As an example of how the approach works, scenarios are constructed by analysing the behaviour of macroeconomic variables (GDP, residential real estate prices, unemployment, interest rates and corporate bond spreads) three years after the onset of a crisis that began during a “standard” phase of the financial cycle. In particular, the macroeconomic variables used in the scenarios are derived from the median values of the historical distribution, ensuring consistency and stability over time.

To ensure robustness, crises are identified using two different approaches, resulting in two sets of scenarios. In the first approach, crises are identified by selecting events from the ESRB Crisis Database (see ECB/ESRB, 2017) consistent with a typical phase of the financial cycle. Selection is based on the systemic risk indicator (SRI), an indicator developed by the ECB to monitor changes in cyclical systemic risk; the SRI must have been between -0.2 and 0.2, the level of a tranquil period, when the crisis started.^[6] In the second approach, a crisis is identified by selecting periods in which GDP declines by at least 1% over two consecutive quarters, regardless of the financial cycle. In line with their objective, the three-year cumulative values for the macroeconomic variables selected in the scenarios are substantially lower than their historical averages across euro area countries, but less severe than those observed under adverse scenarios in past EU-wide stress tests (Chart 1).

Chart 1

The illustrative scenarios used to inform the positive neutral CCyB are significantly more severe than historical observations, but much less so than the EBA’s adverse scenarios

Difference between historical levels and the values used in the EBA’s adverse scenario

(percentages)

The simulated results are consistent with actual positive neutral CCyB rates and align well with estimates derived from other ECB approaches. Averaging across the two different approaches used to define the stress scenario, simulations suggest a euro area aggregate CET1 capital depletion consistent with target PNR rates adopted by countries with a positive neutral CCyB framework and with estimates from other ECB methodologies such as the risk-to-buffer and losses-to-buffer approaches (Chart 2). The positive neutral CCyB simulation using stress test elasticities is robust over time, as incorporating the latest results of the 2025 EU-wide stress test to derive elasticities does not materially affect the results. This stability is a desirable feature of calibration, as the positive neutral CCyB is not expected to be revised frequently. Among the variables included in the regressions, unemployment, real GDP growth and real estate prices emerge as the most influential drivers.^[7]

Chart 2

The stress-test-elasticity approach provides estimates in line with other ECB models and with existing positive neutral CCyB rates

ECB model-based target positive neutral CCyB rates for the euro area and positive neutral CCyB target rates announced in euro area countries

(percentages)

References

BCBS (2024), “Range of practices in implementing a positive neutral countercyclical capital buffer”.

Couaillier, C. and Scalone, V. (2024), “Risk-to buffer: setting cyclical and structural banks capital requirements through stress tests”, Working Paper Series, No 2966, ECB.

De Nora, G., Pereira, A., Pirovano, M. and Stammwitz, F. (2025), “From losses to buffer - calibrating the positive neutral CCyB rate in the euro area”, Working Paper Series, No 3061, ECB.

Detken, C., Hempell, H.S. and Pirovano, M. (2025), “Macroprudential and monetary policy interaction: the role of early activation of the countercyclical capital buffer”, Macroprudential bulletin 31, ECB.

ECB/ESRB (2017), “A new database for financial crises in European countries”.

ECB/ESRB (2025), “Using the countercyclical capital buffer to build resilience early in the cycle”.

Herrera, L., Pirovano, M. and Scalone, V. (2025), “From risk to buffer: calibrating the positive neutral CCyB rate in the euro area”, Working Paper Series, No 3075, ECB.

Lang, J.H., Izzo, C., Fahr, S. and Ruzicka, J. (2019), “Anticipating the bust: a new cyclical systemic risk indicator to assess the likelihood and severity of financial crises”, Occasional Paper Series, No 219, ECB.