Interest rates in the euro area have been low for almost a decade (Brei et al. 2020, Fell et al. 2021). However, following the rise in global inflation, global central banks began tightening monetary policy in the first half of 2022. Despite Russia’s aggression in Ukraine blurring the economic outlook for the eurozone in 2022, expectations of a gradual normalization of monetary policy have begun to raise interest rates in the market.
Sudden changes in interest rates and inadequate management of interest rates can have an unintended effect on financial stability. Banks perform maturity transitions and thus, directly face interest rate risks that may affect their net worth (Driss et al. 2022) and profitability. These concerns have grown over the past few years, as banks in the eurozone have extended lending terms and increased their share of fixed-rate (mortgage) loans.
A parallel shift and steepening of the yield curve
To assess the weakness of the eurozone banking sector for interest rate changes in early 2022, we have looked at the impact of two stylized interest rate situations. The ECB’s macroeconomic forecast for December 2021 has created a baseline, with short-term and long-term interest rates slightly rising despite strong economic growth, a growing stock market and an almost stable spread, despite global supply constraints. There are two stylized situations built on this baseline but involving alternative evolution of interest rates and financial variables over a three year horizon. The first estimates a yield curve shift of approximately 150-250 basis points across maturity compared to the end level of 2021; The second is to tighten the yield curve, whose 10Y edge has increased by 200 basis points and its short edge is firmly anchored at the starting point level. Changes in interest rates have been accompanied by devaluation of equity and corporate bonds, especially in low-rated countries, leveling the premium of country-risk premiums on government bonds.
Two levels of analysis
We employed two ECB stress-testing frameworks, each creating a layer in two-layer analysis and aiming to deliver robust results while reducing model risk. The first framework assumes a constant balance sheet and relies on top-down models used by European banking authorities in the context of EU-wide stress tests integrated since 2011 to ensure the quality of bank estimates (Mirza and Dzhokowski 2017). These are reduction-form regression models that employ various estimation strategies and use macroeconomic, sectoral, bank- or debt-level data to arrive at estimates for credit, market risk parameters, and net interest income components. The second level offers a dynamic balance sheet perspective and relies on a workhorse macro-microbanking Euro Area Sector Stress Test (BEAST) model (Budnik et al. 2020). This semi-structured model includes the eurozone economy and representation of about 90 independent banks. The model is regularly used for both policy assessments (e.g. Budnik et al. 2021a) and macroprudential stress tests (e.g. Budnik et al. 2021b). In favor of our efforts to diversify model risk, BEAST often uses different datasets or experimental methods than top-down models.
The constant balance sheet method focuses on the provision of debt-loss, asset valuation in the bank’s trading books, and the effect of interest rates on net interest income. Bank assets and liabilities that mature or expand within the three-year horizon are replaced by similar financial instruments in terms of type, currency, geography, expiration date, credit quality and original maturity. The bank’s balance sheet contains wholesale and retail funds in various economic sectors in terms of loans, equity exposure and assets in terms of securities and liabilities.
The Dynamic Balance Sheet method considers a wide range of channels, excluding changes in the weight of risk, the components of the bank’s net interest income, and the reactions of pregnant banks. Bank balance sheets are presented with the same level of granularity as the top-down models. Part of the model equation adjusts macro-financial conditions to bank-level loan-loss provisioning parameters, weight of risk or cost of funds, and behavioral responses of other capture banks such as loan amount, interest rate, liability structure, and profit distribution.
Positive profit margins.
It turns out that a sharp rise in interest rates, reflected in parallel changes or the intensification of yield curves, could benefit banking sector profits in early 2022 (see Figure 1). The system-level return on assets (ROA) in the dynamic balance sheet system is largely due to the positive effect of interest rate changes on net interest income (NII) and a small increase in client revenue benefiting from high interest rate volatility. These are more than just compensation for the negative impact of the revaluation of fixed income securities and equities. The latter in material order is the negative impact of interest rates on credit risk and the ability of some borrowers to respect their obligations.
Figure 1 System-Wide Return on Assets (ROA), 2021-2024
Comments: The results presented are based on the dynamic balance sheet method.
… And rarely predict affected system-wide viability
The effect of affluence in the two stylized situations is negative but very inherent. The CET1 ratio drops from 15.5% in 2021 to 14.5% in 2024 in shift and stepping conditions compared to 14.7% at baseline. The panel to the right of Figure 2 breaks down the increasing change in CET1 ratio into CET1 capital loss which enters directly into the capital calculation due to revaluation loss and not through profit-loss, but a slight increase in effective risk weight, and mitigation effect of removal. In the constant balance sheet method and in the absence of the next two adjustments, the difference between the CET1 ratio and the baseline in the shift interest rate scenario goes to 1.4 percentage points and 0.5 percentage points for the stepping scene. Nevertheless, there is a strong (81%) correlation between bank-level liquidity outcomes between dynamic and constant balance sheet systems. The two approaches provide similar correlations to bank-level results for NII (85% correlation coefficient), devaluation loss (52%), and debt-loss provision (36%).
Figure 2 System-wide CET1 ratio, 2021-2024 (left); Increasing changes in CET1 ratio components (right)
Comments: The results presented are based on the dynamic balance sheet method.
Winners and losers
While the average impact on system viability is somewhat negative, there will be losers and winners between organizations. About half of the banks face losses and the other half benefit from similarly sharp rise in interest rates on dynamic and constant balance sheets (see Figure 3). Banks with relatively strong negative effects of high interest rates include development and promotional lenders, public banks and diversified lenders.
Figure 3 Difference between CET1 ratio between Alternative Scenario and Baseline for the Horizon End (2024) Constant Balance Sheet (left) and Dynamic Balance Sheet (Right) method
Comments: TD – Top-down ECB model and constant balance sheet method; BEAST – Dynamic balance sheet method.
What will we learn?
Overall, the euro area banking system is poised to face rising interest rates in early 2022, which has closed a decade of historically low interest rates. However, some banks appear to be somewhat at risk for a sharp revision of interest rates and are at risk of substantial capital losses and it is essential to ensure that it is closely monitored and prepared to deal with sharp changes in interest rates. The past few months have begun to test the firmness of this conclusion, with gradual increases in market interest rates, especially at the long end of the yield curve. The eurozone banking sector has so far avoided these increases against a much glomer economic outlook compared to our baseline, tainted by the Russian aggression on Ukraine. Above all, despite the limitations of how our stylized analysis can help in the current monetary policy discussion, it is reassuring that the resilience of the banking system does not seem to be at risk from the normalization of monetary policy.
Brei, M, C Borio and L Gambacorta (2020), “Bank Mediation When Interest Rates Are Too Low for Long Terms”, VoxEU.org, 7 February.
Dries, J, B Klaus, F Lenoci, and C Pancaro (2022), “Euro Area Bank Banking Book Interest Rate Risk Disclosure and Hedging”, Box 5, ECB Financial Stability ReviewMay.
Fell, J, T. Peltonen and R Portes (2021), “Macroprudential Policy Problems Emerging from a Low-Low Interest Rate Environment”, VoxEU.org, 2 June.
Budnik, K. M. Balatti, I. Dimitrov, J. Gross, M. Kleiman, T. Reichenbachs, F. Sanna, A. Sarchev, N. Sineko and M. Volk (2020), “Banking Euro Area Stress Test Model (BAST)”, ECB Working Paper No. 2469.
Budnik, K, I Dimitrov, C Giglio, J Gross, M Lampe, A Sarychev, M Tarbe, G Vagliano and M Volk (2021a) Paper Series, No. 258.
Budnick, K. L. Bocheri, M. Jankokova, M. Borsuk, J. Carmelavicius, I. Dimitrov, M. Lampe, G. Giraldo, G. Vagliano, J. Gross and M. Volk (2021b), 19) Epidemic “, ECB.
Mirza, H. and D. Jochowski (2017), “Assurance of Stress Test Quality from a Top-Down Perspective”, ECB macroprudential bulletin, Volume. 3.