Bank Profitability

19. Low bank profitability remains a persistent source of vulnerability (Figure 8). German banks have generally lower returns on total assets, risk-weighted assets, and equity (ROE) than the EU averages (Figure 28).¹¹ Cyclical and structural factors include low-interest rates affecting retail banks, constraints on fee and commission income generation amid customer risk aversion and limited customer experience with fee-based products, preference for savings rather than asset management products, strong bank competition, substantial staff base, and dense branch networks. Aggregate figures, however, mask significant heterogeneity among different bank types. Savings banks and cooperatives, sheltered from competition with a regional and household/SMEs focus, have performed better than private banks immersed in lengthy restructurings to address global financial crisis legacies, although the ROE of savings and commercial banks seems to have converged in 2021. Rising interest rates may also squeeze banks’ interest margins over the short-to-near term due to the growing duration mismatch between assets (mortgages) and liabilities (customer deposits).¹² Cooperatives and savings banks are likely to be more affected than commercial banks, given that commercial banks more commonly employ interest rate swaps to hedge against interest rate risks. Rising interest rates may also cool off housing demand and lead to a price correction in RRE. Savings and cooperative banks are also exposed to this risk, given the large share of housing loans in their asset portfolio, with real estate typically serving as collateral. Reportedly conservative collateral valuations are a mitigating factor of the risk. Other risks to profitability include low economic growth, underestimation of credit risk, competitive pressures from fintech and non-banks, slow IT innovation, and cyber risks.

View Full Size

Banks’ Business Model

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

• Download Figure
• Download figure as PowerPoint slide

Real Estate

20. Strong RRE price growth suggests price misalignments, which is higher in large cities. Standard indicators of overvaluation, e.g., the price-to-rent and price-to-income ratios as of end-2021 suggest a 21 percent and 37 percent deviation from long-run averages, respectively.¹³ Also, an econometric model employed by the FSAP at the country-level that takes account of real interest rates suggests RRE overvaluation of 10-15 percent as of 2021Q3, while a city-level panel data points to larger overvaluation estimates at end-2020.¹⁴ Bundesbank’s estimates, based on the latest data and several methodologies, suggest an overall overvaluation in RRE prices of 20-35 percent, with overvaluation in cities in the range of 15-40 percent in 2021.

21. Large CRE exposures and RRE price overvaluations suggest potential pockets of vulnerabilities in banks’ exposures to real estate (Figures 9, 30). In 2021Q3, outstanding bank loans for house purchases were 46 percent of GDP and about half of total loans to domestic enterprises and households. Housing loan growth in 2021Q3 was the highest in two decades. Banks’ exposures to CRE sectors (construction, housing corporations, and other real estate activities) was much smaller. Lending to CRE sectors, which peaked in 2019, remained strong in 2021Q3 although slowed marginally. The Bundesbank’s Bank Lending Survey (BBLS) did not indicate relaxation of lending standards at end-2021. However, the risks could be underestimated due to a lack of comprehensive borrower-based information (e.g., loan-to-value (LTV), debt-service-to-income (DSTI)). Immediate risks to the CRE sector may stem from sub-sectors (e.g., non-food retail and potentially hotel businesses), particularly with COVID-19 outlook uncertainty and rising interest rates in conjunction with higher inflationary pressures.

View Full Size

MFIs’ Exposure to Real Estate and Construction

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

• Download Figure
• Download figure as PowerPoint slide

22. Tail risks in real estate have increased since the onset of the pandemic, particularly in CRE. The FSAP’s property price-at-risk assessment pointed to increased tail risks in RRE and CRE in 2021Q3 compared to end-2019. Under the estimated conditional distributions, as of 2021Q3, the probability of a negative real price growth one year ahead increased to 2.2 and 66 percent (from 0.7 and 24 percent at end-2019) for RRE and CRE, respectively. Results also indicate a 5 percent chance that RRE and CRE prices could fall by 14 and 30 percent, respectively, under severe adverse scenarios in real cumulative terms over the medium-term (Figure 10). The use of price-at-risk model for CRE and RRE prices to supplement the Bundesbank’s toolkit to assess RE risks would help serve as an additional early warning indicator to help inform policy decisions.

View Full Size

Residential and Commercial Property Price-at-Risk 1/

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Source: IMF staff estimates and calculations.1/ RRE = residential real estate, CRE = commercial real estate, PaR = price-at-risk. Panel 1 shows the conditional one-year-ahead probability distributions of RRE price growth based on a parametric, t-skew density, fitted over quantile regression estimates in the period 2021:Q3 (blue line) and 2019:Q4 (orange line). Real house price volatility and worsening macro-financial conditions during COVID-19 largely account for the leftward shift in the distributions between 2019:Q4 and 2021:Q3. Figures are annualized growth rates. Panel 2 shows the point estimates of predicted RRE PaR as of 2021:Q3 in the short-term (one-year-ahead) and the medium-term (two- and three-years-ahead) projections at the 5th percentile (at compounded growth rates). Two- and three-year-ahead estimates show compounded figures. The counterfactual (“adverse scenario”) scenario is calibrated as a simultaneous 2 standard deviations shock to leverage (change in household debt-to-GDP, interest payments-to-disposable income), affordability measure (house price-to-gdp per capita ratio (misalignment)), and financial conditions index. Panel 3 shows the conditional one-year-ahead probability distributions of CRE price growth based on a parametric, t-skew density, fitted over quantile regression estimates in the period 2021:Q3 (blue line) and 2019:Q4 (orange line). Figures are annualized growth rates. Panel 4 shows the point estimates of predicted CRE PaR for 2021:Q3 in the short-term (one-year-ahead) and the medium-term (two- and three-years-ahead) projections at the 5th percentile (at compounded growth rates). Two- and three-year-ahead estimates show compounded figures. The counterfactual (“adverse scenario”) is calibrated as a simultaneous 2 standard deviations shock to employment.

• Download Figure
• Download figure as PowerPoint slide

23. Further closing data gaps including on lending standards are needed to monitor existing and evolving risks. In 2023, under the Financial Stability Data Collection Regulation (FinStabDEV), the Bundesbank will begin collecting data on housing loans to private households from all commercial lenders , including on LTV ratios and information on the amortization period and borrowers’ ability to service debt. A time series of lending standards to assess evolution over time should be helpful in informing policy decisions. Buoyant developments in the RRE market call for further data gaps closures to help identify and mitigate tail risk events. Data on transaction prices in the residential real estate markets need improvement, particularly at the regional level, to help improve data quality of the official RRE price indices.¹⁵

Macro-Financial Risks

24. The FSAP assessed the following key macro-financial risks in the adverse scenario, which combines several shocks (RAM: Table 6): an escalation of the war that could be associated with a Russian gas shut off, higher commodity prices, a global resurgence of COVID-19 with extended supply chain disruptions and de-anchoring of inflation expectations in the U.S. and advanced Europe, leading to rising core yields and risk premia.

• De-anchoring of inflation expectations in the U.S. and/or advanced European economies with the worsening of supply-demand imbalances, rising energy prices, and other risks related to the conflict in Ukraine, and higher nominal wage growth, leading to rising core yields and risk premia.

• Global resurgence of COVID-19 with extended supply chain disruptions could require costly containment efforts if protracted in time, causing greater scarring to the economic fabric and financial sector’s balance sheets, adding pressure on capital buffers and margins, and triggering credit tightening, an increase of zombie corporates, a wave of bankruptcies and higher NPLs.

25. The above-noted risks were reflected in the adverse scenario used in the stress testing exercise. The baseline scenario already includes elevated inflation from rising oil and gas prices due to the war-in-Ukraine. In the adverse scenario, COVID-19- and war-related constraints on the supply of manufactured parts and raw materials, including gas and oil, lead to higher commodity price hikes, higher-than-expected inflation, and a real GDP decline, including significant damage to the large manufacturing sector in Germany. Unexpected monetary tightening in the U.S. and the euro area and repositioning by market participants lead to tightening of financial conditions, a rise in risk premia, and assets market selloffs. Higher rates and supply constraints cause loss of confidence, a drop in demand, and recessionary pressures in Germany and lead to an increase in corporate and household NPLs. Moreover, the shift in market sentiment and lower investor confidence against some high-debt EA countries could raise sovereign yields, have knock-on effects on the broader financial sector, and affect German banks’ and other financial institutions’ holdings in these countries.

Bank Solvency Stress Testing

26. The FSAP banking solvency stress tests covered about 83 percent of the banking system’s assets and included banks of different sizes and business models. The FSAP conducted two separate stress tests using the same macroeconomic scenario and risk parameters—one for the SIs drawing on ECB supervisory data, and the other for LSIs analysis using Bundesbank supervisory data.¹⁶ The FSAP assessed banks’ resilience to (i) credit risk related to non-financial corporates, households, and sovereigns; (ii) interest rate and other market risks, particularly foreign exchange and sovereign risks, equity prices, and real estate prices; (iii) other income channels subject to modeling constraints and data availability.¹⁷

27. The FSAP assessed the banking system’s resilience to shocks against the World Economic Outlook’s (WEO) baseline scenario and a RAM-based adverse scenario (Appendix IA, STeM). The adverse scenario includes shocks noted above that result in a V-shape trough of 13.4 percent decline in real GDP relative to the baseline by 2024 (3 standard deviations from the baseline over two years, Figure 11).^18,19 The adverse scenario assumes that the output gap closes in 2025 and involves stagflation and monetary tightening followed by monetary policy loosening associated with a drop in demand (Figures 12, 31). House and stock prices decline by 23 percent and 29 percent, respectively, by 2024. Unemployment increases by 6.6 percentage points above the baseline and remains 0.8 percentage points above the baseline at the end of the projection horizon.

View Full Size

Macroeconomic Scenarios

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Note: The WEO figures corresponded to March WEO projections and differ from the latest updates presented in Table 4.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Global Macrofinancial Model Simulations

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

• Download Figure
• Download figure as PowerPoint slide

28. German SIs and LSIs are resilient to shocks under the baseline and adverse scenarios, thanks to the initial high capital buffers and low initial credit risk (Figures 13, 14). Under the baseline scenario, SIs’ aggregate capitalization remains high, driven by net income and a decline of credit risk despite some moderate near-term losses caused by the interest rate increase (which leads to interest rate and repricing risks). Under the adverse scenario, the aggregate common tier equity (CET1) capital ratio of SIs depletes by 5.2 percentage points by 2023 from 14 percent and increases in subsequent years. Because of strong funding shocks and a sharp recession, higher risk weighted assets (RWAs), interest rate, market, and to a lesser extent credit risks drive the decline in capital ratios.²⁰ Three banks fall below the hurdle rate in 2022-23, but capitalization remains above the minimum CET1 ratio and the aggregate capital shortfall remains small at 0.3 percent of GDP.²¹ The stress tests for LSIs show that, thanks to high overall capitalization, aggregate capital remains high under the adverse scenario. Overall, 8-9 very small banks (accounting for up to 1.6 percent of total LSIs assets) fall below the hurdle rate under the baseline scenario, and about 20 banks (accounting for about 3 percent of total LSI assets) fall below the hurdle rate under the adverse scenario.

View Full Size

Solvency Stress Tests for SIs

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Note: Net income (due to stress) is defined as the residual component of the net income that results from the stress test exercise (e.g., credit risk and interest rate risk.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Solvency Stress Tests for LSIs

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

• Download Figure
• Download figure as PowerPoint slide

Bank Liquidity Stress Testing

29. The banking system has significantly increased liquidity buffers since the 2016 FSAP (Figure 25). The weighted average liquidity coverage ratio (LCR) stands at about 160 percent, and the vast majority of banks have liquidity well in excess of the regulatory minimum of 100 percent and in excess of the NSFR of 100 percent (Figure 15). Other indicators (e.g., deposit-to-loan ratio and the ratio of liquid assets to total assets) also suggest ample liquidity in the system as a whole. Counter-balancing capacity (CBC) is robust and tilted to the highest quality assets, especially for SIs.

View Full Size

LCR and NSFR

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Sources: Bundesbank, ECB and IMF staff calculations

• Download Figure
• Download figure as PowerPoint slide

30. The FSAP stress tests confirm that the banking system appears to be resilient to liquidity shocks, especially at the conventional liquidity stress test horizon (Figure 16). Less than 5 percent of banks in the sample would become illiquid within a 1-month horizon, and about 15 percent (all small LSIs) at the 3-month horizon. Shortfalls at these horizons, at less than 0.1 percent and 0.5 percent, respectively, of the LSI sample’s assets would be manageable. Even at the 1-year horizon, shortfalls under the severe adverse scenario would be comparatively small, and the IPS schemes (whose support is not considered in the tests) could provide a second line of defense in such cases. In the LSI sample, only cooperatives and savings banks would account for additional shortfalls in the 3-month to 1-year horizons in the most adverse stress scenarios.

View Full Size

Cash Flow Analysis Results

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Sources: Bundesbank and IMF staff calculations

• Download Figure
• Download figure as PowerPoint slide

31. Under a severe adverse scenario, selected banks’ U.S. dollar exposures could pose a risk, which emphasizes the importance of existing swap lines among central banks (Figure 17). Only a few banks rely on the capacity to convert euro-denominated assets into U.S. dollars rapidly and at a low cost, a reliance that is reasonable in the highly liquid and efficient euro-dollar currency market. In the unlikely event of a disruption of access to U.S. dollar liquidity, a few banks with inadequate U.S. dollar liquidity buffers could face challenges under stressed conditions even within a few days. While U.S. dollar liquidity risk is relevant only for a limited number of banks, continued close monitoring of larger LSIs with significant foreign exchange exposures is warranted and take action as needed.

View Full Size

Cash Flow Analysis Results, FX

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Note: FX = foreign currency; CBC = counter-balancing capacity; CB = Central Bank; CQS = Credit Quality Step; LSIs = Less Significant Institutions; SIs = significant Institutions.

• Download Figure
• Download figure as PowerPoint slide

Interconnectedness Analysis and Contagion

32. The FSAP bank contagion analysis indicated that interbank contagion risks flow from SIs to LSIs and from LSIs as a group to SIs. The simulations assume that all banks fail one at a time and characterize the cascade of defaults in the interbank market due to credit and funding shocks. The contagion index measures the sum of losses in percent of the capital of all banks or groups of banks resulting from all or a subset of the 1317 simulations; the amplification index is the ratio of second and further rounds of losses to the losses incurred during the first round of contagion.²² Contagion risks are much less among LSIs and SIs (Figure 18). The results suggest the interbank market may be segmented among SIs and LSIs implying that failure of a SI (respectively LSI) generates limited direct contagion risks to other SIs (respectively other LSIs). The interbank market is concentrated, with a few large banks accounting for a significant share of the interbank lending volumes. As a result, a small number of banks account for most of the contagion losses. The analysis also found that while losses from the first round tend to dominate, as the severity of shocks increases the second and other rounds account for a higher share of the total losses, especially for simulations where a small bank is the trigger bank. Both SIs and LSIs are impacted by contagion losses as a share of their capital. Under the FSAP model’s hypothetical assumptions, credit cooperatives, savings banks, Landesbanken, and commercial banks are significantly exposed to contagion risks in the interbank market; the FSAP does not assess the likelihood that these risks would materialize.

View Full Size

Bank Contagion Analysis 1/

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Sources: Bundesbank and IMF staff calculations1/ Simulations 2, 5 and 8 respectively assume a 20 percent, 50 percent and 80 percent loss given default and haircuts on marketable securities in the contagion risk simulation.

• Download Figure
• Download figure as PowerPoint slide

33. The FSAP makes several recommendations to address pockets of vulnerabilities in the banking system. Stress test results showed that the overall banking system appears fairly resilient to severe shocks. At the same time, the high degree of uncertainty around the scenarios and the future recovery paths along with the uncertainties around the war-in-Ukraine warrant caution and some policy strengthening to ensure continued robustness of the financial system. The FSAP recommends that the authorities: (i) continue to closely monitor banks’ prudential ratios, particularly large SI commercial banks, and establish microprudential buffers (Pillar 2 guidance) for less capitalized banks as needed; and (ii) strengthen LSIs interest rate risk monitoring, including by gathering data on the remaining maturity of retail deposits, wholesale funding, and interest-bearing assets to perform top-down interest rate stress tests. The FSAP also recommends that data sharing between the Bundesbank and the ECB is strengthened and that risk monitoring and the analysis of domestic and cross-border interconnectedness continue to be strengthened, with a focus on key domestic interbank market institutions and other markets where exposures are located, as needed.

Corporate Sector

34. The pandemic caused a significant drop in enterprise sales and would have increased corporate debt at risk had it not been for timely measures to cushion the impact (text table). The sensitivity analysis suggests that, in the absence of support measures and even if listed firms had been able to offset the decline in sales by cutting production costs, the share of firms at risk would still have increased from 23 percent pre-pandemic to 60 percent, with debt-at-risk surging from 4 percent to 41 percent (Figure 19). Similarly, 38 percent of firms (up from 26 percent pre-pandemic) would have been unable to maintain positive cash balances in the absence of new borrowing and more than 5 percent of firms could have ended up with equity below zero, from none in the pre-pandemic period.

View Full Size

Corporate Sector Vulnerabilities—Sensitivity Analysis

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

• Download Figure
• Download figure as PowerPoint slide

Shock to listed companies' sales: Mid-2020 prediction & estimated outturn

Source: Datastream, Capital IQ and IMF Staff estimates

^1/Difference between analysts' January 2020 and une 2020 sales forecasts for individual firms for 2020.

Shock to listed companies' sales: Mid-2020 prediction & estimated outturn

Sector (# of listed firms in sample)	Share in 2019 total sales (percent)	Predicted shock to 2020 sales 1/ (percent)	Actual change in 2020 sales (percent)
Agriculture (2)	0.1	-3	15
Air transport (2)	2.2	-53	-62
Amusement and Recreation (6)	0.1	-55	-61
Business services (46)	2.5	-7	4
Communication (3)	7.1	-2	8
Construction	1.6	-19	-12
Electricity, gas, water supply (8)	7.8	-22	4
Hotels & restaurants (1)	0.0	-45	14
Manufacturing (145)	60.2	-19	-10
Mining excluding oil (3)	0.2	-13	-38
Other private services (7)	0.3	-11	-36
Social, health, education services (5)	3.0	-4	2
Transportation excl. air transport (6)	5.2	-21	-8
Wholesale & retail trade (19)	9.8	-11	-9
Total (263 listed companies)	100	-17	-8

Source: Datastream, Capital IQ and IMF Staff estimates

^1/Difference between analysts' January 2020 and une 2020 sales forecasts for individual firms for 2020.

View Table

Shock to listed companies' sales: Mid-2020 prediction & estimated outturn

Sector (# of listed firms in sample)	Share in 2019 total sales (percent)	Predicted shock to 2020 sales 1/ (percent)	Actual change in 2020 sales (percent)
Agriculture (2)	0.1	-3	15
Air transport (2)	2.2	-53	-62
Amusement and Recreation (6)	0.1	-55	-61
Business services (46)	2.5	-7	4
Communication (3)	7.1	-2	8
Construction	1.6	-19	-12
Electricity, gas, water supply (8)	7.8	-22	4
Hotels & restaurants (1)	0.0	-45	14
Manufacturing (145)	60.2	-19	-10
Mining excluding oil (3)	0.2	-13	-38
Other private services (7)	0.3	-11	-36
Social, health, education services (5)	3.0	-4	2
Transportation excl. air transport (6)	5.2	-21	-8
Wholesale & retail trade (19)	9.8	-11	-9
Total (263 listed companies)	100	-17	-8

Source: Datastream, Capital IQ and IMF Staff estimates

^1/Difference between analysts' January 2020 and une 2020 sales forecasts for individual firms for 2020.

35. Analysis of listed NFCs indicate a moderate increase of debt-at-risk under the adverse scenario.²³ Under the solvency stress test’s adverse scenario, based on end-2020 firm-level data, debt-at-risk would increase by six percentage points to 16 percent of total debt of listed NFCs (equivalent to about 2 percent of assets of banks in Germany), and the probability of default would increase by six basis points to 1.19 percent by 2023.²⁴ Reflecting the central role of economic growth in the profitability and viability of companies in the regression models, the share of debt in firms with an ICR<1 starts to rise in 2023 under the adverse scenario (Figure 20).

View Full Size

Corporate Sector and Dynamic Scenario-Based Regression Analysis

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Sources: Bundesbank and IMF staff calculations

• Download Figure
• Download figure as PowerPoint slide

Climate Transition Risks

36. Germany’s ambitious emissions reduction targets can have a non-negligible impact on the macroeconomy, which can have implications on the financial sector. In assessing the impact of Germany’s emissions targets on GDP and sectoral gross value-added, the FSAP analysis used a global computational general equilibrium (CGE) model (known as the “IMF-ENV” model) with 25 regions and 37 sectors. The simulations assume (i) other countries will meet their nationally determined contributions (NDCs), and (ii) carbon tax (or carbon pricing) is the instrument used to mitigate CO2 emissions. With these assumptions, the model suggests that reaching the 2030 emissions target would imply carbon pricing of USD320 per ton by 2030 and would adversely affect GDP.²⁵ The GDP impact in the simulations hinges on how carbon taxes are used (Figure 21).

View Full Size

Simulation of Macroeconomic Impact of Germany’s Mitigation Policy

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

• Download Figure
• Download figure as PowerPoint slide

• If used to lower the wage tax, GDP is estimated to decline by one percent by 2030 compared to the business-as-usual (BAU) scenario, which assumes no changes in climate policy and does not account for the impact of temperature increases.

• If used to provide a lump-sum rebate, GDP would decline by 1.6 percent because of the increase in workers’ wealth (and fall in labor supply).

• If used to pay off government debt, GDP would decline the next five years led by the decline in household consumption (driven by higher energy prices) but would recover with improved budget balances and their positive effect on national saving and investment.

The relatively benign impact on GDP masks sector heterogeneities. Sectors that emit large CO2 (e.g., coal, oil, gas power generation, the chemical and mining industries) are most affected while renewables (e.g., solar and wind) are to gain in value added.

37. The FSAP estimates indicate that expected losses to banks under the mitigation policy will rise slightly over time, but would still remain small at around 0.05 percent of total loans by 2030. ²⁶ Expected losses, while small, differ across bank types. Landesbanken and development banks are the most affected groups due mainly to their large exposure to the power and transport sectors (Figure 22). Savings banks and cooperative banks are the least affected groups due to their larger exposure to less carbon intensive sectors (e.g., real estate, construction).²⁷

View Full Size

German Banks’ Expected Credit Losses Under Climate Mitigation

Citation: IMF Staff Country Reports 2022, 231; 10.5089/9798400216886.002.A001

Note: The probability of default (PD) for each sector i used the coefficients estimated by the Bundesbank (PD_it = 0.664 ∗ PD_it−1 − 0.0509 ∗ ΔVA_it − 0.0267 ∗ ΔVA_it−1 − 0.0433 ∗ ΔEQP_it ΔVA_it and ΔEQP_it refer to the rates of change in scenario; see Bundesbank 2022). The expected loss of m-number of bank groups, consisting of n-number of sectors, is then calculated by aggregating changes in the probability of defaults for sectors multiplied by the loss-given default weighted by bank exposure to these sectors. The analysis used data on LGD by sector as of 2021Q3 provided by the Bundesbank.

• Download Figure
• Download figure as PowerPoint slide

38. The authorities need to build on the progress made by banks in enhancing their analysis, management, and disclosure of climate risks. According to the survey by BaFin in 2021 and a similar survey by the ECB in 2021, very few banks have conducted stress testing sustainability risks, although many banks are preparing for stress testing, and none of the institutions are close to fully aligning their practices with the supervisory expectations.²⁸ The disclosure of climate risks appears to be also somewhat limited. The FSAP underscores the importance for the authorities to work with the financial sector to expand its analytical capacity for assessing climate risks, enhance the resilience against unexpected future risks, and promote the disclosure of climate risks. The ECB is already conducting stress testing on SIs; the FSAP recommends BaFin to expand its stress test work to cover LSIs.

Insurance Sector

39. The effect of prolonged low-interest rates on the profitability and solvency of insurers, particularly those having portfolios of long-term policies with significant interest guarantees, has partly driven a search for yield. It has also led to the realization of gains on fixed-income securities to fund the expense of building up the Additional Provision to the Premium Reserve (Zinszusatzreserve – ZZR) to meet the guarantees. The pandemic has also increased the risk of credit downgrades. EIOPA’s January 2022 Risk Dashboard (RD) ranked macro risks as the highest category of risk currently facing the European insurance sector.

40. These risks are particularly acute for life insurers—exposed to the risks of offering significant guarantees on long-term policies. Guarantees on some policies are as high as four percent, and the term of policies can be very long; the weighted average years to maturity of German life insurers’ liabilities at the end of 2020 was 26.3 years. Furthermore, life insurers have been required since 2011 to build additional reserves under GAAP to reflect the potential cost of the guarantees in the low-interest rate environment. Together with prolonged low-interest rates, these guarantees are affecting the financial soundness of life insurers. Higher interest rates should ultimately enable insurers to generate more interest income to cover their guarantees, and the ZZR reserves set aside (for the policies whose guaranteed rate is below the reference rate) released. However, higher market interest rates might also reduce or eliminate the unrealized gains, making it more difficult for insurers to fund the expense of building up the ZZR. The 2021 EIOPA insurance stress test assessed the industry’s resilience to a prolonged COVID-19 scenario in a “lower for longer” interest rate environment from a solvency and a liquidity perspective. The exercise confirmed that the main vulnerabilities stem from market shocks but that insurers would be able to cope. The application of the Solvency II’s transitional provisions and management actions would enable an adequate solvency position to be maintained. The exercise did not show liquidity to be a significant vulnerability; the results for German participants appear to be consistent with the overall results, and results reported by insurers in their ORSAs and liquidity stress tests insurers performed at the request of BaFin.

Macroprudential Framework

41. Germany’s institutional framework is strong and operates effectively. The institutional arrangements in Germany are closely aligned with Fund guidance and broadly ensure the willingness and ability to act (IMF, 2014).²⁹ A macroprudential mandate is assigned to the Financial Stability Committee (FSC), a collegiate, high-level body comprised of voting members from the MoF, Bundesbank, and BaFin. The FSC chair (the State Secretary of the MOF) seeks consensus, striking an appropriate balance between the need to avoid delays and ensure different views are considered. BaFin ensures the ability to act as the National Designated Authority (NDA) for tools specified by European legislation (CRD/CRR) and is responsible for domestic macroprudential measures (e.g., residential real estate lending limits). The Bundesbank can exert considerable influence on the Committee’s proceedings by proposing resolutions for FSC action, and retains a veto on amendments to them and on their publication.

42. The FSC’s overall macroprudential instrument strategy should link macroprudential risks and mitigation measures. The FSC publishes, and periodically revises, its macroprudential strategy, which is good practice. However, the strategy should better articulate the link between the high-level principles governing the conduct of policy and the operational level of instrument setting. At the same time, an appropriate balance should be maintained between the provision of information guiding market participants and the need to react flexibly should new circumstances dictate. The FSC should also publish post-meeting records, providing timely and relevant information on the Committee’s activities, as in some other OECD jurisdictions, to promote the transparency and predictability of decision making.

Macroprudential Policy

43. The increases in the CCyB and the SSyRB for residential real estate exposures were appropriate, well-timed, and well-coordinated across the responsible agencies. The CCyB has been set to levels broadly consistent with the authorities’ buffer guide. Similar to 2019, the focus of concern is on potential build-up of vulnerability in the residential real estate sector. There is also evidence of compressed spreads in corporate debt markets, and some tilting in the composition of bank exposures towards riskier corporate borrowers. Weak profit margins limit banks’ ability to generate capital organically in the event of macroeconomic stress. The authorities should continue to use the flexibility afforded by the discretionary component of the CCyB framework to ensure the buffer is set to a level from which it can be materially reduced in the event of stress.

44. Additional macroprudential action to counter evolving vulnerabilities will be required. Accelerating growth in mortgage lending and house prices, coupled with evidence of over-valuations, make action on borrower-based measures (BBMs) increasingly urgent. This action would help bolster the resilience of borrowers’ balance sheets, complementing the actions taken to improve lenders’ resilience described above. Whereas BBMs have been activated in many OECD peers (Table 3), and have been recommended for activation in Germany by the ESRB (ESRB, 2022), legal concerns and ongoing data gaps appear to have fed into a reluctance to activate an LTV limit. Moreover, the legislative processes that is needed to implement past recommendations in full to augment borrower-based macroprudential tools and to close data gaps in residential real estate have proceeded exceptionally slowly. The FSAP recommends that:

• Legislated powers over as-yet unused borrower-based instruments in the Banking Act (Kreditwesengesetz) and related laws should be enhanced, and additional income-based tools introduced as soon as practical. In particular, provision should be made to allow existing borrower-based instruments, as well as similar additional instruments that may be introduced, to be activated prior to a material deterioration in lending standards.³⁰

• Guidance on residential real estate lending standards should be strengthened, especially in respect of the practice of lending against real estate transaction costs. The effectiveness of current guidance on lending standards risks is being limited by its lack of specificity, and the difficulties supervisors face in reviewing lending practices absent consistent reporting.

• Legally-binding borrower-based limits should be activated as soon as practical unless a material change of direction in real estate and credit markets is observed.

• The FSC should initiate the development of a communication strategy in support of the activation of borrower-based measures to help promote their acceptability.³¹

Banking Sector

45. The FSAP conducted a focused review of the regulation and supervision of Germany’s LSIs.³² As member of the EA, German banks’ regulation and supervision take places within the ECB’s Single Supervisory Mechanism (SSM). BaFin and the Bundesbank are responsible for the supervision of LSIs. The ECB exercises oversight over the functioning of the system.

46. The authorities made good progress on the implementation of the recommendations of the 2016 Basel Core Principles (BCP) assessment. The authorities fully rolled out the ECB/SSM approach to the Supervisory Review and Evaluation Program (SREP) to all LSIs in 2020 and made enhancements to liquidity and operational risks. However, there is scope for further strengthening institutional arrangements, the legal framework and approach to corporate governance, and some aspects of the supervisory framework and approach.

47. In response to the Wirecard fraud, the MoF initiated a reorganization program including several legal and structural reforms impacting BaFin. These reforms are aimed at strengthening BaFin’s ability to supervise complex credit institutions, undertake forensic audits, strengthen financial accounting oversight, the use of whistleblower information, consumer protection, and further digitalization of business processes. The (internal) powers of BaFin’s President were strengthened vis-a-vis the strategic direction, budget proposal, organization structure and the ability to give directions to BaFin’s Executive Directors. These reforms will enable BaFin to make earlier use of corrective and sanctioning powers and increase its effectiveness in dealing with lingering deficiencies of problem banks.

48. Streamlining the current reporting to the MoF into a more systematic approach with the responsibility elevated to the office/secretariat of BaFin’s president will strengthen BaFin’s operational independence. As observed in the 2016 assessment and the file review of the current assessment, the communication between BaFin and the MoF appears to go beyond the necessary oversight and financial stability responsibilities of the MoF.

49. Going forward, a joint BaFin and Bundesbank strategic agenda is critical to avoid duplicative or disjointed efforts to ensure that an effective banking supervision program is in place. This complementing strategic agenda should recognize the need for joint projects (including working towards shared data, systems, and tools) to ensure a fully cooperative partnership on the BaFin reform program.

50. Further enhancements are needed to align banks’ corporate governance framework with international best practices. The role of the supervisory board needs to be further strengthened in its oversight of the internal control functions (direct reporting lines of internal audit, compliance, and risk management) to ensure that critical checks and balances are in place to oversee and govern executive management activities. The current construct still leaves room for management boards to dominate essential control functions. The modernization of the corporate governance framework will require legal amendments.

51. The extent of reliance on external auditors and the ability of the authorities to conduct in-depth file reviews of auditors’ assessments should be reassessed. It is essential that supervisors be well placed to not only challenge the work of the external auditor but also to perform deep dives on the various risk areas when needed.

52. Supervisors, external auditors, and the industry would benefit from further guidance to complement Germany’s principles-based risk management guidelines (MaRisk) to clearly communicate supervisory expectations. Given the extensive nature of the EU regulations, EBA Guidelines (including Germany’s partial or non-compliance stance), as well as international standards, additional guidance for the industry, the external auditors, and supervisory staff is needed as MaRisk do not necessarily clearly outline supervisory expectations specific for banks in all areas.

53. Certain aspects of the overall supervisory framework (e.g., supervisory engagement and frequency of on-site inspections) need to be further strengthened. The German supervisory framework has been developed along the lines of the ECB LSI SREP methodology. Based on the SREP score and the impact, the Germany authorities determine both the minimum level of engagement (MEL) and the minimum frequency for the on-site inspection of LSIs. The MEL, which will be strengthened beginning with the 2022 supervisory cycle, currently does not include a minimum of engagement with the management function or internal control functions, even for the lowest risk assessment score. Further, the frequency of on-site inspections should be reassessed, given the need to ensure adequate coverage of both the riskiest LSIs and the highest impact LSIs. BaFin/Bundesbank will therefore need to reassess the adequacy of risk resources to ensure satisfactory coverage of on-site inspections, ability to undertake deep-dive reviews and ensure critical risk expertise.

54. The authorities have taken a competition and technology neutral approach (e.g., same activity, same risk, same regulation) to the regulation of fintech. This means for example that neobanks and online securities trading platforms have to comply with the same regulations as incumbent institutions and that crypto assets service providers need to be licensed according to the financial services they provide. While continuous monitoring is needed, the existing regulatory perimeter seems adequate, and the authorities are proactively monitoring and acting against unlicensed service providers. In the absence of a competition or market development mandate and regulatory experimentation clauses, BaFin has not setup a sandbox. Instead BaFin established in 2017 an “innovation hub” for market monitoring, internal coordination of regulatory and supervisory initiatives (hub and spoke model), and market outreach.

55. Fintech data collection, forward-looking and dynamic market monitoring, and financial stability analysis need further development. Bundesbank’s Statistics Directorate has run a pilot project to collect data on fintech. The first results of the pilot indicate that the size of the fintech sector across different financial sectors is still relatively small. However, given the rapid growth, a more structural approach to data collection, financial stability analysis, and a more dynamic forward-looking market monitoring are important while going forward to match the quickly developing monitoring needs.

Insurance Sector

56. There are no major shortcomings in the observance of Insurance Core Principles. Nevertheless, there is scope for improvement in the following four areas:

• BaFin and the MoF need to reduce the real and perceived risks to BaFin’s operational independence, as recommended in the previous FSAP.

• The authorities should strengthen and simplify the solvency framework through measures such as considering the impact of more extreme interest-increase scenarios on the funding of the Zinszusatzreserve, strengthening liquidity risk management reporting and stress testing requirements, and streamlining internal model approval and reporting requirements.

• BaFin should reassess the minimum and actual frequency of on-site inspections to ensure that they are sufficiently frequent to support robust identification and assessment of risks. Potential cost charges of an inspection to a particular insurer should not be a barrier to greater frequency of inspections. BaFin should also enhance its management reporting to facilitate systematic tracking and reporting of the timeliness of off-site and on-site supervisory activities and provide more comprehensive feedback to insurers on its supervisory findings, assessments, and concerns.

• To strengthen the resolution regime, particularly to facilitate the orderly exit or resolution of a failing large insurer or group, BaFin has extended the requirement for recovery plans to a total of 15 insurance groups in Germany (over 80 percent market coverage required by the EIOPA Opinion on the 2020 review of Solvency II). BaFin also continues to have resolution planning in place for the former globally systemically important insurers (G-SII) group. However, in addition, BaFin should have its powers strengthened and, for example, be able to require contingency plans and resolution plans. Such expansion should provide the tools required to exercise the resolution of an internationally active insurance group (as set out by the international association of insurance supervisors).

Financial Market Infrastructures

57. Clearstream Banking AG Frankfurt (CBF) is a Central Securities Depository (CSD) that provides the post-trade infrastructure for the German securities markets. It settles transactions for all German securities and futures markets and securities traded on various global trading platforms. It also acts as a custodian for securities accepted in Germany and issued by German and international issuers in the form of collective or individual certificates or registration rights. As of end-2020, the market value of outstanding securities held by CBF in collective safe custody was EUR 8.6 billion, and it served 286 participants, of which 110 were foreign. In addition to its function as a CSD and as a Securities Settlement System (SSS) operator, CBF is a licensed credit institution under the German Banking Act and authorized to provide banking-type ancillary services under the Central Securities Depositories Regulation (CSDR) of the EU. BaFin supervises CBF as the designated National Competent Authority (NCA), and the Bundesbank oversees its SSS operations. CBF’s banking services are supervised jointly by the Bundesbank and BaFin.

58. The FSAP assessment found CBF to be in observance of 20 of the 21 relevant principles of the PFMI and in broad observance of 1, albeit with room for improvement on certain critical areas. CBF was found to be of high market repute with a solid legal basis and a comprehensive approach to risk management. It has clear and transparent rules on all relevant processes, including for access to services and ensuring final settlement. In addition, CBF has a comprehensive default management framework, which is regularly tested. Discussions with market participants clearly indicated that CBF is a trusted infrastructure providing critical services for German securities markets and beyond. There is, however, room for improvement, particularly with respect to governance. The internal control functions of risk management, internal audit, and compliance are the responsibility of a single Executive Board (EB) member—the Chief Executive Officer—who is also responsible for global operations. The CEO is, therefore, concurrently the Chief Operating Officer and Chief Risk Officer, with there being no separation for the second and third lines of defense at the level of the EB. In addition, the Chair of the Risk Committee is the CRO of CBF’s ultimate parent, Deutsche Börse, and is therefore not independent. Finally, the performance of the Supervisory Board (SB) is not subject to periodic independent assessment. The FSAP recommends that CBF separate the responsibilities for internal audit from risk management and compliance at the level of the EB. In addition, CBF should appoint an independent member as Chair of the Risk Committee, as well as conduct periodic, independent reviews of the performance of its SB.

59. While CBF generally has a comprehensive and robust framework for identifying, monitoring, and mitigating a wide range of material risks, there are steps that can be taken to enhance risk management. For example, CBF relies on direct participant disclosures to gain information on its indirect participants. However, CBF could strengthen the criteria and thresholds it has in place which trigger such disclosures—e.g., by including system-level thresholds—, and work toward implementing direct monitoring of the risks posed to the infrastructure by the clients of its direct participants. In addition, despite its sophisticated collateral management system and processes, CBF could further mitigate credit risk by implementing ex-ante monitoring of collateral concentration limit breaches.