The importance of being positive: costs and benefits of a positive neutral rate for the countercyclical capital buffer

by Luis Herrera, Valerio Scalone and Mara Pirovano

This article explores the benefits of a positive neutral rate for the countercyclical capital buffer (CCyB) and the conditions shaping the economic costs of its activation in a general equilibrium framework. The analysis shows that a gradual build-up of the buffer and favourable banking sector conditions (e.g. high profitability) limit these economic costs. Furthermore, a positive neutral CCyB rate ensures banking sector resilience in all phases of the financial cycle and improves macroprudential authorities’ ability to provide relief to the banking sector in the event of (potentially large) shocks, including those unrelated to the materialisation of domestic credit imbalances.

1 Introduction

The experience from the COVID-19 pandemic has sparked reflections on the role of releasable capital buffers and highlighted the desirability of enhancing their use. Health emergencies such as the pandemic, as well as natural disasters, wars and shocks arising from climate change, political events or technological disruptions, may happen at any stage of the financial cycle. These shocks can have negative repercussions on the banking sector and lead to disruptions in the financial intermediation function, which could amplify the initial impact. Given the concerns about the usability of structural buffers like the capital conservation buffer, releasable macroprudential buffers take on much greater significance in addressing such disruptions. Nonetheless, the CCyB, as the main releasable buffer, is currently tailored to mitigating cyclical systemic risks linked to the domestic credit cycle, with its accumulation contingent on indicators signalling excessive credit growth. This led to a limited build-up of releasable macroprudential buffers prior to the pandemic, constraining the extent of the macroprudential relief when the shock occurred.^[1]

In the aftermath of the pandemic, an increasing number of macroprudential authorities in the banking union have introduced a positive rate for the CCyB, even in the absence of elevated cyclical systemic risks (a positive neutral CCyB rate). These CCyB increases occurred late in the financial and economic cycle, with existing capital headroom and strong bank profitability seen as the key factors mitigating the procyclical effects of the increased capital requirements on lending (as theorised by Lang and Menno, 2023). At present, seven countries have either announced the introduction of a positive neutral CCyB rate or already have it in place, with target rates ranging from 1% to 2%.^[2] While the divergence of target rates across countries reflects the different methodologies used for its calibration, country-specific characteristics and policymakers’ preferences, the motivations for introducing a positive neutral CCyB rate are broadly shared across jurisdictions (Behn et al., 2023).

The introduction of a positive neutral CCyB rate reflects one of the main lessons learned from the pandemic, namely that adverse shocks with potentially disruptive implications for the financial system may occur in any phase of the financial cycle. Building up a CCyB through a positive neutral rate early in the financial cycle (i.e. when cyclical systemic risks are neither subdued nor elevated) allows for a more gradual increase in capital requirements over time, which has been found to lower implementation costs (Mendicino et al., 2020). In addition, it guarantees that releasable buffers are available to absorb losses stemming from shocks that could occur in any phase of the cycle or triggered by events unrelated to the domestic credit cycle. Furthermore, a positive neutral CCyB rate ensures that capital is available to cover potential unidentified risks, or risks that have not yet been captured by the established risk indicators owing to data lags. More generally, a greater amount of releasable buffers ensures that capital is available to absorb losses and allow credit institutions to fulfil their key economic functions in the event of (potentially large) adverse shocks (Behn et al., 2023).

A higher share of releasable macroprudential capital buffers also reduces concerns related to the effective usability of capital buffers, as banks may be reluctant to use available buffers in order not to breach regulatory thresholds (ECB, 2022 and references therein). Couaillier et al. (2024) find that, during the pandemic, euro area banks with little capital headroom above the regulatory buffers reduced corporate lending compared with other banks to avoid dipping into their buffers. Couaillier et al. (2022) show that the releases of prudential requirements during the pandemic were particularly effective in supporting bank credit supply for capital-constrained banks with little capital headroom above the combined buffer requirement (CBR). By increasing the distance to the CBR, releasable buffers should incentivise banks to use the released capital to support the real economy. Using the ECB’s stress test model, Borsuk et al. (2020) show that banks’ willingness to use capital buffers is reflected in higher lending, with positive effects on economic outcomes.

While the conditions supporting the build-up of the CCyB to address emerging cyclical systemic risks are, by now, well defined and broadly similar across banking union countries^[3], as yet no common understanding has been reached on those guiding the early build-up of the CCyB via a positive neutral rate. In most countries that have already implemented a positive neutral CCyB rate, its activation (or the replenishment of the buffer after a crisis) is conditional on (sufficiently) positive economic growth and the absence of financial imbalances (see Table 1 in Behn et al., 2023). The prevailing banking sector conditions, such as bank profitability, have also been important factors affecting the decisions to activate a positive neutral CCyB rate. Higher bank profitability dampens the potential cost of increasing capital requirements in terms of reduced lending, thereby limiting potential procyclical effects. Simulations based on the model developed by Lang and Menno (2023) show that the impact on lending from a 1 percentage point increase in macroprudential capital buffers is in fact smaller for more profitable banks. The impact is smallest for banks characterised by both high profitability and high voluntary capital buffers (see also ECB, 2023).

This article explores the benefits and costs of activating a positive neutral CCyB rate using a version of the dynamic stochastic general equilibrium (DSGE) model with three layers of default (the so-called “3D model”) developed by Mendicino et al. (2024). Specifically, we use a general equilibrium framework to assess the benefits of additional releasable buffers and the conditions shaping the cost of the activation of a positive neutral CCyB rate. To reflect the countercyclical nature of the CCyB, we focus on the short-term costs of its activation, taking account of the role played by bank profitability as in Lang and Menno (2023).^[4] To assess the benefits of a positive neutral CCyB rate, we examine how it helps to ensure banks’ resilience against shocks occurring in the early stages of the financial cycle. Furthermore, we examine how larger releasable capital buffers can provide relief to the banking sector in the event of (potentially large) shocks. We find that the implementation of a positive neutral CCyB rate can be associated with low economic costs, particularly if the buffer is built up gradually and during favourable periods (e.g. in times of high bank profitability). We also find that it (i) ensures banking sector resilience in all phases of the financial cycle and (ii) improves the macroprudential authorities’ ability to provide relief to the banking sector in the event of (potentially large) shocks, including when they are unrelated to the materialisation of domestic credit imbalances.

2 The model

The 3D model is a micro-founded DSGE model with financial frictions where borrowing households, entrepreneurs and banks may default on their liabilities (Figure 1). Borrowing households finance house purchases with bank loans and default on their mortgage loans when the value of the collateral is lower than the outstanding debt obligations. Entrepreneurs engage in capital investment, financing their capital purchases with entrepreneurial wealth and bank loans. Entrepreneurs default on their loans when the return on their investments is lower than the contractual debt obligations. The financial system is populated by two types of bank, one specialised in lending to households and one specialised in lending to entrepreneurs. Each type of bank raises equity from shareholders and takes deposits from saving households to finance their loan portfolio. Banks fail when the realised return on the loan portfolio is lower than their deposit repayment obligations.^[5]

The macroprudential authority sets structural and countercyclical capital requirements. In the model, capital regulation requires banks to hold capital in relation to the size of their loan portfolio. While structural capital requirements do not vary throughout the cycle, the CCyB rule implies that the CCyB is built up (released) when credit growth is positive (negative).^[6] To show the benefits of additional macroprudential space in the form of releasable buffers, we introduce a lower bound into the CCyB rule, which captures the fact that macroprudential authorities can only release a CCyB that has been built up in the past.^[7] Greater macroprudential space generated through a positive neutral CCyB activation implies that macroprudential authorities can release a larger amount of capital before reaching the lower bound.

To assess how banking sector conditions shape the transition costs of introducing a positive neutral CCyB rate, we allow for different costs of equity issuance in the model. One key channel for banks to accumulate equity is equity issuance. The higher the cost of equity is, the higher the cost of increasing capital requirements will be. In our version of the model, we introduce the possibility for saving households to directly issue additional equity to the banks, subject to a transaction cost.^[8] This cost determines the importance of the equity channel and, consequently, the costs that banks face when adjusting their equity to comply with capital requirements over the financial cycle. This approach allows us to explore how the costs of accumulating new bank capital can affect the costs of imposing higher capital requirements.

Figure 1

Stylised overview of the 3D model

3 Benefits and costs of a positive neutral CCyB rate

Favourable banking sector conditions, such as high profitability, reduce the cost of activating the positive neutral CCyB rate. In the short run, banks may comply with higher regulatory capital requirements in different ways, with varying implications for the average cost of their implementation. First, they can reduce credit supply, triggering an increase in lending spreads. Second, they can accumulate more capital via retained earnings. Third, they can issue new equity. When banks are profitable, they can more easily set aside capital via retained earnings. Chart 1 depicts the impact on GDP and total credit of a 100 basis point increase in capital requirements, phased in over four quarters, under different bank profitability conditions. These conditions are mapped to different levels of the equity issuance cost. In the baseline scenario, the equity issuance cost is calibrated to 1.5. In the high profitability scenario, we assume that the cost is three times smaller (0.5), so that bank equity can be more easily transferred from saving households to banks. In order to simulate a low profitability scenario, we set the equity issuance cost parameter to 10, so that capital transfers from savers are almost completely offset. In this case, the cost of capital is high, and banks will resort to more deleveraging to comply with the higher capital requirements. Compared with the baseline scenario^[9], higher bank profitability reduces the peak negative impact of higher capital requirements on GDP and lending by 0.05 and 0.6 percentage points respectively (difference between the blue and yellow lines in Chart 1). Conversely, lower bank profitability would increase activation costs in terms of GDP and total credit by 0.12 and 1.1 percentage points respectively compared with the baseline scenario.

Chart 1

High bank profitability reduces the cost of increasing capital requirements

Impact of a 100 basis point increase in capital requirements under different levels of bank profitability

(y-axis: percentage deviations from the steady state; x-axis: quarters)

An early and more gradual build-up of the CCyB over the financial cycle reduces the costs of build-up for the economy, especially when bank profitability is low. A gradual increase in capital requirements allows banks more time to accumulate retained earnings and build up buffers with a smaller reduction in the credit supply, resulting in less upward pressure on spreads and less bank deleveraging. To demonstrate this point, we examine the effects on lending and GDP of capital requirement activations that are phased in over different periods. Chart 2, panel a), shows the impact on GDP of a 250 basis point increase in capital requirements over four quarters (blue bars) and eight quarters (yellow bars) for different levels of bank profitability. Let us consider the scenario in which cyclical systemic risks remain at low or median levels in the first four quarters and increase to levels that call for the activation of a 2.5% CCyB rate in the second year. Without the possibility of setting a positive neutral CCyB rate, the macroprudential authority would activate a 2.5% CCyB rate only in the second year, phasing it in over four quarters. In the scenario where a positive neutral rate framework is adopted, an initial CCyB rate of, for example, 1% can be activated during the first year, followed by an additional 1.5% in the second year. In the absence of a positive neutral CCyB rate, banks would have less time to build up the total 250 basis point buffer. In the high bank profitability scenario, the costs of activation in cases where no positive neutral buffer has previously been built up would be equal to -0.15% of GDP compared with the case of non-activation. Where a 1% positive neutral rate is used, the costs of activation would be smaller (-0.09% of GDP at their peak) given the increased reliance on retained earnings to build up capital. The cost reduction resulting from a gradual activation is even greater in the case of the baseline profitability scenario, where the activation would imply a decrease in GDP of -0.29% compared with the case of non-activation (at the peak). In the scenario of a gradual activation, the peak effect on GDP would be smaller (-0.19%) as banks could generate more capital via retained earnings.

Chart 2

A gradual activation of the CCyB early in the cycle is less costly and allows the financial sector to remain resilient even in the early phase of the financial cycle.

a) Impact on GDP of a 250 basis point increase in capital requirements over four (standard) vs eight quarters (positive neutral rate) under different levels of bank profitability

(y-axis: percentage deviations from the steady state)

b) Effects on GDP of an adverse shock occurring in the early (t+4) or mature (t+8) phase of the financial cycle for different levels of CCyB available

(y-axis: percentage deviations from the steady state; x-axis: quarters)

An early and gradual build-up of the CCyB has the important benefit of ensuring the financial system remains resilient even if adverse shocks materialise early in the cycle. Let us consider again the scenario where cyclical systemic risks remain at low or median levels in the first four quarters and then increase to levels warranting the activation of a 2.5% CCyB rate in the second year. Chart 2, panel b), compares the impulse responses of GDP with an adverse financial shock under different scenarios. In the first scenario, the CCyB is built up when cyclical systemic risk increases, i.e. only in the second year (from time t+4 to t+8). In the second scenario, the buffer is built up via a positive neutral rate earlier in the cycle when risks are still low, i.e. from time t to t+4. In this context, two alternative risk materialisation episodes are simulated: (i) risks materialise at time t+8 when the build-up of the buffer is completed in both scenarios (blue line), and (ii) risks materialise at time t+4 when the buffer has been put in place via a positive neutral rate in one scenario (yellow line) and no buffer has been built up in the other scenario (red line). If a financial shock hits the economy in time t+4, when the capital buffer had not been previously built up, the shock will cause a loss in GDP of -1.5% (red line). If the same shock occurs at the end of the phasing-in period, the impact on GDP will be substantially smaller (-0.7%, blue line), as higher capital buffers lead to greater banking sector resilience. In the case of an early build-up of the CCyB via a positive neutral rate, the same financial shock hitting in t+4 would cause a loss in GDP of -1.1% (yellow line). The benefits of the positive neutral rate can be measured by comparing the recessions in the case when no buffer has been built up (red line) and the positive neutral rate case (yellow line). These benefits derive from the fact that an early build-up of the buffer makes banks relatively more resilient if a financial shock occurs early in the financial cycle. Banks would have a relatively lower probability of default compared with the non-activation scenario, keeping the risk level and default costs in the economy contained. This would also imply a smaller deceleration in credit growth in the case of activation of the positive neutral rate compared with the non-activation scenario. This is particularly relevant as shocks with potentially disruptive consequences for the banking sector, such as health emergencies, natural disasters, wars, and shocks arising from climate change, political events or technological disruptions, may occur at any stage of the cycle.

Furthermore, a positive neutral CCyB provides additional macroprudential space to support the economy in the event of severe shocks, helping to smooth their impact on lending and the economy. If risks materialise, a CCyB release allows banks to absorb losses and continue to provide lending to households and firms during economic downturns. Chart 3 shows the impulse responses of GDP, total credit and average bank capital requirements to a mix of recessionary shocks hitting the economy under three different scenarios.^[10] In the first scenario, no releasable capital buffers are in place (red line) and the shocks cause a 1.5% reduction in GDP compared with the starting point. In the second scenario, we assume that the macroprudential authority only has a 1% CCyB activated to address cyclical systemic risks. In response to the shock, the macroprudential authority fully releases the buffer in line with its CCyB policy rule. As a result, the recession is 0.3 percentage points smaller at its peak (yellow line) compared with the baseline. However, given the magnitude of the shock, a larger release of the CCyB would have been necessary. Since only 1% of the CCyB is available, the macroprudential authority hits the lower bound and is therefore constrained in its action (yellow line in Chart 3, panel c). In the third scenario, we assume that the macroprudential authority has a total 2% CCyB rate in place, i.e. a 1% CCyB activated to address cyclical systemic risk and a 1% positive neutral CCyB. In this case, the CCyB is bigger so a larger release can be made without hitting the lower bound (blue line). The increase in available macroprudential space, achieved by increasing the distance to the lower bound, reduces constraints on the macroprudential authority. This enables it to further dampen the reduction in lending and GDP, especially in the second and third years following the materialisation of the shock. During the third year, the countercyclical action of the release reduces the loss in GDP by 0.3 percentage points compared with the scenario where only a 1% CCyB rate is in place (i.e. the difference between the blue line and the red line). Conversely, no positive effect is observed in the case of a constrained release (yellow line overlapping with the red line). By increasing the macroprudential space available to support the banking sector in the event of unexpected and severe shocks, a positive neutral rate approach reduces the probability of hitting the lower bound and makes the countercyclical action of macroprudential authorities more effective.

Chart 3

A positive neutral CCyB rate provides macroprudential authorities with greater leeway to support the economy in the event of severe shocks.

Impact of a CCyB release: unconstrained (PNR case) and constrained (no PNR) release

(y-axis: percentage deviations from the steady state; x-axis: quarters)

4 Conclusions

The experience from the pandemic highlighted the desirability of holding a larger amount of releasable capital buffers to address shocks that could occur at any stage of the financial cycle. Since then, several macroprudential authorities in the banking union have set a positive rate for the CCyB in the absence of excessive credit developments. The main aim of this move is to facilitate a more gradual build-up of the buffer, providing macroprudential authorities with greater leeway to make capital available for use when needed by releasing the accumulated buffers in the event of adverse shocks. Favourable banking sector conditions, such as high bank profitability, are among the factors that support the implementation of a positive neutral CCyB rate, as they mitigate potential procyclical effects of the increased capital requirements on lending. This article explores the benefits of a positive neutral CCyB rate and the conditions shaping the costs of its activation within a general equilibrium framework.

In line with other studies (Lang and Menno, 2023), the results confirm that the economic costs of increasing capital requirements, including through setting a positive neutral CCyB rate, decrease when banking sector conditions are favourable. The more profitable banks are, the easier it is for them to satisfy the additional requirements with retained earnings, reducing their need for deleveraging. A positive neutral CCyB rate allows for a more gradual build-up of the buffer in the early stages of the financial cycle, thereby reducing economic costs compared with a situation where the CCyB is increased less gradually, such as when cyclical systemic risks warrant its swift activation. Furthermore, building up macroprudential space early in the cycle ensures that the macroprudential authority has capital buffers available for release in response to any adverse shocks unrelated to the domestic credit cycle in the early phases of the cycle.

Importantly, the additional macroprudential space provided by a positive neutral CCyB rate alleviates constraints faced by macroprudential authorities when releasing the CCyB in response to shocks. With more capital available for release, macroprudential authorities have more firepower to support the banking sector in the event of distress. This benefit is particularly significant in the face of large shocks.

References

Basel Committee on Banking Supervision (2010), “Guidance for national authorities operating the countercyclical capital buffer”, Bank for International Settlements, December.

Behn, M., Pereira, A., Pirovano, M. and Testa, A. (2023), “A positive neutral rate for the countercyclical capital buffer – state of play in the banking union”, Macroprudential Bulletin, No 21, ECB, Frankfurt am Main, April.

Borsuk, M., Budnik, K. and Volk, M. (2020), “Buffer use and lending impact”, Macroprudential Bulletin, Issue 11, ECB, Frankfurt am Main, October.

Clerc, L., Derviz, A., Mendicino, C., Moyen, S., Nikolov, K., Stracca, L., Suarez, J. and Vardoulakis, A.P. (2015), “Capital regulation in a macroeconomic model with three layers of default”, International Journal of Central Banking, Issue 40, pp. 9-63.

Couaillier, C., Reghezza, A., Rodriguez d’Acri, C. and Scopelliti, A. (2022), “How to release capital during a pandemic? Evidence from euro area banks”, Working Paper Series, No 2720, ECB, Frankfurt am Main, February.

Couaillier, C., Lo Duca, M., Reghezza, A. and Rodriguez d’Acri, C. (2024), “Caution: do not cross! Distance to Regulatory Capital Buffers and Corporate Lending in a Downturn”, Journal of Money, Credit and Banking, February 2024.

European Central Bank (2022), “Annex 2: Enhancing macroprudential space in the banking union”, Report from the Drafting Team of the Steering Committee of the Macroprudential Forum, March.

European Systemic Risk Board (2018), “The ESRB handbook on operationalising macroprudential policy in the banking sector”, December.

Mendicino, C., Nikolov, K., Scalone, V. and Supera, D. (2024), “Inflationary shocks, Monetary Policy and the Financial Stability Trade-offs”, mimeo.

Mendicino, C., Nikolov, K., Suarez, J. and Supera, D. (2020), “Bank capital in the short and in the long run”, Journal of Monetary Economics, Vol. 115, pp. 64-79.

Lang, J. and Menno, D. (2023), “The state-dependent impact of changes in bank capital requirements”, Working Paper Series, No 2828, ECB, Frankfurt am Main, February.

Annex

The model is calibrated by matching steady-state values with macro-financial variables, following the strategy presented in Mendicino et al. (2020).