## EXECUTIVE SUMMARY — 1deuea2022010

## Source details

**Canonical URL:** [EXECUTIVE SUMMARY — 1deuea2022010](https://www.imf.org/-/media/files/publications/cr/2022/english/1deuea2022010.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2022/english/1deuea2022010.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2022/english/1deuea2022010.pdf.json)

---

### Key findings on profitability and resilience
- Reported returns in 2020:
  - ROA = 0.13 percent; RORWA = 0.34 percent.
  - EU averages in 2020: ROA = 0.15 percent; RORWA = 0.43 percent.
  - Reported ROE in 2020: 1.95 percent (EU average = 2.3 percent).
- Bank cost of capital (sample of large European banks): in the range of 8-12 percent.
- German banks rank more favorably in risk-adjusted terms because lower profitability is partially compensated by lower volatility of returns, but on measures such as returns on assets, equity, and risk-weighted assets, German banks on aggregate rank among the least profitable in Europe.
- Low profitability undermines resilience by:
  - Hindering the build-up of capital.
  - Increasing likelihood that banks use capital rather than earnings to absorb losses in a shock.
  - Constraining ability to fund growth through retained earnings, raise capital and debt more cheaply, and implement recovery plans.

### Drivers of low profitability
- Cyclical factors:
  - Low-interest rates and flat yield curves have put downward pressure on loan and deposit interest rates, squeezing profit margins.
  - Retail banks are more susceptible due to maturity transformation via deposit-taking and lending.
- Structural factors:
  - Three-pillar industry structure with barriers to entry and partial market segmentation (commercial banks; public sector banks—savings banks “Sparkassen” and Landesbanken; cooperative banks) affects competition and business models.
  - Savings and cooperative banks have mandates to maximize welfare of stakeholders rather than profits.
  - High competition and high costs erode profits.
  - Limited ability to diversify revenues through fees and commissions due to customer risk-aversion and low experience with fee pricing.
- Heterogeneity across bank groups:
  - Commercial banks (particularly large private banks) have reported low returns since the Global Financial Crisis.
  - Savings and cooperative banks have reported relatively stable profitability, funding largely through retail deposits, enabling market share gains and loan volume growth that partially offset declining interest rates.

### Banking system structure and scale (select statistics)
- German MFIs’ assets increased from about EUR 7.5 to EUR 9.4 trillion in 2010-21Q3; relative to GDP these declined from 291 to 263 percent of GDP.
- Banking sector composition (end-2020):
  - Total credit institutions = 1,408.
  - Total assets = 9,207 EUR bn; Percent of GDP = 273; Number of branches = 24,060.
  - Commercial banks: 164 institutions, holding about 43 percent of assets.
    - Big banks: 3 banks, 2,749 EUR bn (30 percent of total assets), 82 percent of GDP.
    - Regional and other commercial banks: 139 institutions, 1,094 EUR bn (12 percent).
    - Branches of foreign banks: 22, 123 EUR bn (1 percent).
  - Savings banks: 377 institutions, 1,407 EUR bn (15 percent).
  - Credit cooperatives: 814 institutions, 1,030 EUR bn (11 percent).

### Profitability dynamics, costs, and revenues
- Net interest income (NII) reduction is the main source of profit pressure for retail banks.
- German banks’ NII relative to total income broadly on par with EU average, but NII per unit of assets is lower than the European average.
- NFCI-to-assets ratios largely remained stable; a May 2021 Federal German Court of Justice ruling may have reduced NFCI generation.
- Impairment charges from NPLs put little pressure because aggregate NPL ratios remain low by EU standards.
- Cost efficiency:
  - Cost-to-income ratio in Germany = 72 percent in 2020.
  - EU average cost-to-income ratio = 63 percent in 2020.
  - Cost-to-asset ratio for German banks = 1.3 percent in 2020.
  - European average cost-to-asset ratio = 1.2 percent in 2020.
- Aggregate underperformance largely reflects low revenues: declining NII and stagnant NFCI.

### Bank performance by type and real estate exposures
- Savings and cooperative banks generated about 90 percent of the industry’s profits in 2020; their joint share in industry profits increased from 30 percent in 2006 to over 90 percent in 2020.
- Joint market share of savings and cooperative banks in private sector loans increased from about 43 percent to 52 percent during 2010-21Q3.
- Between end-2008 and 2021Q3:
  - Savings and cooperative banks increased total assets by 45 percent and risk-weighted assets by 43 percent.
  - Credit cooperative banks increased total assets by 77 percent and risk-weighted assets by 68 percent.
- By 2021Q3, lending to domestic households and private enterprises grew 4.2 percent year-on-year, of which about 3.2 percentage points were attributed to savings and cooperative banks.
- Housing loans grew by 7.3 percent year-on-year, to which savings and cooperative banks contributed 4.8 percentage points.
- For savings banks, housing loans increased from 29 to 35 percent of assets between 2008-21Q3; lending to households increased from 21 to 24 percent of assets; lending to corporates increased from 7 to 11 percent of assets.
- For credit cooperatives, housing loans increased from 29 to 36 percent of assets between 2008-21Q3; loans to households increased from 23 to 26 percent of assets; loans to corporates increased from 6 to 10 percent of assets.
- Savings banks’ lending to commercial real estate sectors increased from 9.1 to 14.1 percent of assets during 2008-21Q3.
- Credit cooperatives’ lending to commercial real estate sectors increased from 5.4 to 11.8 percent of assets during 2008-21Q3.

### Real estate and housing price dynamics
- Country-wide owner-occupied housing prices increased by some 60 percent in real and over 90 percent in nominal terms during 2010-21.
- In the largest 7 cities, house prices increased by 143 percent during 2010-21.
- Drivers: supply shortages, income growth, strong economic outlook, immigration and urbanization, construction sector labor shortages, limited land availability, shortages of construction materials.
- Consequence: housing less affordable; elevated exposures raise vulnerability to RRE/CRE price correction.

### Empirical analysis and regression findings
- Samples used:
  - European SIs: 112 SSM-supervised institutions operating in 19 countries (twenty German SIs contributed about 17 percent to sample assets).
  - German bank sample: 1,358 banks as of mid-2021.
- Data period: bank-level dataset covering 2000-20; many regressions focus on 2008-20.
- Estimation method: Arellano-Bover/Blundell-Bond linear dynamic panel-data estimator with robust standard errors; Arellano-Bond and OLS used for robustness checks.
- Key regression results (European SIs):
  - Larger banks with lower NPLs, higher capitalization, and better cost efficiency tend to be more profitable.
  - Bank size positively correlated with profitability (statistically significant with ROAA as dependent variable).
  - Higher capitalization (tangible common equity-to-tangible assets) positively correlated with profitability.
  - Better asset quality (lower NPL ratios) associated with higher profitability.
  - Better cost efficiency (lower cost-to-income ratio) correlates with higher profitability.
  - Business model variables not statistically significant in most specifications.
  - Number of bank branches per capita found negatively correlated with ROAA; bank concentration positively correlated with profitability (coefficients not robust across specifications).
- Key regression results (German bank sample):
  - Smaller, better-capitalized banks with better cost efficiency are likely to be more profitable.
  - Bank size negatively correlated with profitability (coefficients significant for savings banks and credit cooperatives sub-samples).
  - Cost-to-income ratio negatively correlated with profitability.
  - Real GDP growth and monetary policy stance (2-year government debt yield) strongly correlated with higher bank profitability.
  - Greater bank concentration (market share of largest 5 institutions) positively correlated with profitability.
  - Tangible common equity and equity-to-assets coefficients not statistically significant across specifications.
- Risk-adjusted measures:
  - Sharpe ratio and Modigliani-Modigliani measures indicate German banks’ long-term profitability averages fall within low percentiles of European comparator samples (e.g., 20th percentile for average ROE; Sharpe ratios within the 24th percentile).
  - Credit cooperatives’ profitability in risk-adjusted terms is above the median of the European sample (60th percentile).
  - German SIs’ average weighted risk-adjusted ROE over 2001-20 equivalent to the 28th percentile of the European SI sample.

### Recent policy and industry responses
- Cost-reduction measures:
  - Substantial reductions in branch networks and employee headcount.
  - Consolidation mostly within savings and cooperative banks; since the GFC the number of credit institutions declined by about a quarter (over 500 institutions); institutions per million inhabitants declined from 20 to 15.
  - Digitalization investments increased during COVID-19; expected short-run cost increases with medium-term benefits.
- Revenue-side actions:
  - Greater pass-through of negative interest rates to clients has helped slow NII decline.
  - Efforts to increase fee and commission income face persistent challenges.
- Restructuring of major private banks:
  - Between 2006 and 2020, balance sheet size reduced by 13 percent for Deutsche Bank and by 17 percent for Commerzbank.
  - Deutsche Bank’s headcount declined by about 17 percent over the last 10 years.
  - Branch networks shrunk by almost 40 percent for each institution between 2010-20.

### Risks, vulnerabilities, and regulatory responses
- Downside risks include:
  - Geopolitical shocks (e.g., war in Ukraine) affecting GDP growth via higher energy prices, tighter financial conditions, supply chain disruptions, and rising inflation.
  - Potential COVID-19 resurgences, lockdowns, and sectoral shocks (hotels, retail) increasing NPLs.
  - Real estate price correction in RRE and CRE combined with higher unemployment could force mortgages into negative equity and raise NPLs.
  - Faster transmission of rising interest rates into liabilities than assets could keep profitability low in the short term.
  - Slow IT innovation and fintech competition; increased cyber risk (EBA survey Fall 2021: 55 percent expected increase in operational risk; 90 percent expected cyber/data security issues).
  - Possibility of underestimation of credit risks given loan loss provisions largely reversed in 2021 to pre-pandemic levels.
- Regulatory measures:
  - BaFin set a countercyclical capital buffer of 0.75 percent of risk-weighted assets on domestic exposures and introduced a sectoral systemic risk buffer of 2.0 percent of risk-weighted assets on loans secured by residential real estate. The rates are currently at zero percent. Institutions have to meet the additional capital requirements by February 1, 2023.
  - Phasing out of ECB support measures such as TLTRO III may impact profitability.

### Policy recommendations (selected and timing)
- Continue to collect and ensure sufficient granularity of data to assess bank exposures to risks in RRE and CRE; assess potential effects of RRE/CRE price correction and interest rate adjustment regularly, including via stress tests.
  - Timing: ST
  - Authorities: Bundesbank, BaFin, ECB
- Examine scope for cost and inefficiencies reduction, including through further consolidation and optimization of the extensive branch network.
  - Timing: MT
  - Authorities: Banks
- Analyze benefits of revenue diversification and implement measures to boost non-interest revenues.
  - Timing: MT
  - Authorities: Banks
- Step up digitalization and automation efforts to face fintech competition.
  - Timing: MT
  - Authorities: Banks
- Additional recommended measures:
  - Headcount and branch network optimization combined with digitalization.
  - Greater process automation and product standardization while managing adverse selection risks.
  - Support market-led consolidation to attain economies of scale and reduce obstacles to cross-border consolidation.

*Source: EXECUTIVE SUMMARY — 1deuea2022010 (Technical note prepared by Alla Myrvoda and Dan Cheng).*

### EXECUTIVE SUMMARY __________________________________________________________________________ 5

### EXECUTIVE SUMMARY

### Key findings on profitability and resilience
- German bank profitability is low by international standards:
  - Reported returns in 2020: ROA = 0.13 percent; RORWA = 0.34 percent.
  - EU averages in 2020: ROA = 0.15 percent; RORWA = 0.43 percent.
  - Reported ROE in 2020: 1.95 percent (EU average = 2.3 percent).
  - Bank cost of capital (sample of large European banks): in the range of 8-12 percent.
- German banks rank more favorably in risk-adjusted terms because lower profitability is partially compensated by lower volatility of returns, but on measures such as returns on assets, equity, and risk-weighted assets, German banks on aggregate rank among the least profitable in Europe.
- Low profitability undermines bank resilience by:
  - Hindering the build-up of capital.
  - Increasing the likelihood that banks use capital rather than earnings to absorb losses in a shock.
  - Constraining banks’ ability to fund growth through retained earnings, raise capital and debt more cheaply, and implement recovery plans.

### Drivers of low profitability
- Cyclical factors:
  - Economic growth and interest rates are vital. Low-interest rates and flat yield curves have put downward pressure on loan and deposit interest rates, squeezing profit margins.
  - Retail banks are more susceptible to interest rate changes because their business models rely on maturity transformation through deposit-taking and lending.
- Structural factors:
  - A complex three-pillar industry structure with barriers to entry and partial market segmentation (commercial banks; public sector banks—savings banks “Sparkassen” and Landesbanken; cooperative banks) affects competition and business models.
  - Savings and cooperative banks have an explicit mandate to maximize welfare of stakeholders rather than profits, which affects incentives and retained earnings usage.
  - High competition and high costs have been eroding profits.
  - Limited ability to diversify revenues through fees and commissions due to customer risk-aversion and low experience with fee pricing.
- Heterogeneity across bank groups:
  - Commercial banks (particularly large private banks) have reported low returns since the Global Financial Crisis; large private banks underwent costly restructurings with headcount and branch reductions, de-risking, and digitalization.
  - Savings and cooperative banks have reported relatively stable profitability, funding operations largely through retail deposits, enabling market share gains and loan volume growth that partially offset declining interest rates.

### Recent policy and industry responses
- Cost-reduction measures implemented:
  - Substantial reductions in branch networks and employee headcount.
  - Consolidation mostly within savings and cooperative banks.
  - Ongoing, but slow, digitalization and fintech adoption.
- Revenue-side actions:
  - Greater pass-through of negative interest rates to clients has helped slow net interest income decline.
  - Efforts to increase fee and commission income face persistent challenges.

### Risks and vulnerabilities
- Geopolitical risks (e.g., war in Ukraine) may affect banks via second-round effects on the economy, default rates, and inflation.
- Potential COVID-19 outbreaks and lockdowns could lead to deteriorating economic conditions, insolvencies, losses, and profit declines.
- In a rising interest rate environment, low profitability could persist in the short term because of faster transmission of interest rates into liabilities than assets.
- Slow IT innovation and digitalization and increased fintech competition could disadvantage traditional banks.
- A real estate price correction (residential and commercial) could significantly impact profitability due to large bank exposures.
- Recommendation to continue collecting granular data on bank exposures (notably RRE and CRE) and to regularly examine potential effects of risk materialization via stress tests.

### Banking system structure and scale (select statistics)
- German MFIs’ assets increased from about EUR 7.5 to EUR 9.4 trillion in 2010-21Q3; relative to GDP these declined from 291 to 263 percent of GDP over this period.
- The banking sector comprises about 1,408 credit institutions (end-2020), structured along three pillars:
  - Commercial banks: 164 institutions, holding about 43 percent of assets at end-2020.
    - Big banks: 3 banks, 2,749 EUR bn (30 percent of total assets), 82 percent of GDP (table context).
    - Regional and other commercial banks: 139 institutions, 1,094 EUR bn (12 percent).
    - Branches of foreign banks: 22, 123 EUR bn (1 percent).
  - Public sector banks (Landesbanken and savings banks): e.g., savings banks: 377 institutions, 1,407 EUR bn (15 percent).
  - Credit cooperatives: 814 institutions, 1,030 EUR bn (11 percent).
- Table 2 (selection): All categories of banks: 1,408 reporting institutions; Total assets = 9,207 EUR bn; Percent of GDP = 273; Number of branches = 24,060 (as of 2020).

### Empirical analysis and heterogeneity
- The note analyzes drivers of profitability using two samples:
  - Single Supervisory Mechanism (SSM)-supervised European banks (European significant institutions).
  - German Monetary and Financial Institutions (MFIs) bank sample.
- Findings highlight significant heterogeneity: loan volume growth boosted profits of savings and cooperative banks; commercial banks, especially large private banks, show prolonged low returns.

### Policy recommendations (Table 1: Germany — Key Recommendations)
- Continue to collect and ensure sufficient granularity of data to assess bank exposures to risks in RRE and CRE. Potential effects of risk materialization on banks, particularly of RRE/CRE price correction and interest rate adjustment, should continue to be assessed in depth regularly, including via stress tests.
  - Timing: ST
  - Authorities: Bundesbank, BaFin, ECB
- Examine the scope for cost and inefficiencies reduction, including through further consolidation in the banking system by way of mergers and optimization of the extensive branch network.
  - Timing: MT
  - Authorities: Banks
- Analyze the benefits of revenue diversification and implement measures to boost non-interest revenues.
  - Timing: MT
  - Authorities: Banks
- Step up digitalization and automation efforts to face competition from fintech.
  - Timing: MT
  - Authorities: Banks
- Note: ST=short-term, MT=medium-term, RRE=residential real estate, CRE=commercial real estate.

*Source: EXECUTIVE SUMMARY — 1deuea2022010 (Technical note prepared by Alla Myrvoda and Dan Cheng).*

### 8.      In part, this reflects the high competition in the German banking sector. With over

### 8.      In part, this reflects the high competition in the German banking sector. With over 

### Banking sector structure and competition
- Over 1,400 banks operate in Germany, creating fierce competition.
- Traditional bank business models rely on a fairly extensive branch network with a significant employee base.
- Branch penetration declined from 0.48 to 0.29 branches per thousand of inhabitants during 2010-20.
- Bank employees per capita declined from about 8.3 to 6.9 employees per thousand of inhabitants during 2010-20.
- The number of credit institutions and employees remains high relative to some peers.
- A complex tiered structure of the banking system increases fragmentation and inhibits attainment of economies of scale.
- Credit cooperatives and savings banks pursue mandates to maximize welfare of stakeholders rather than profits.

### Profitability drivers and pressures
- Low interest rates and flat yield curves, together with strong competition, have squeezed bank profit margins and put significant pressure on banks’ interest income.
- Net interest income (NII) reduction is the main source of profit pressure for retail banks due to heavy reliance on loans, particularly mortgages.
- On aggregate, German banks’ NII relative to total income has remained broadly on par with the EU average, but NII per unit of assets is lower than the European average.
- German banks’ ability to generate fee and commission income has been hampered by customer risk aversion and limited experience with fee pricing, exacerbated by strong competition.
  - German retail customers are risk-averse, prefer in-person interaction, and opt for saving over investment products.
  - Internet banking penetration is low (though increasing), and credit card use is limited relative to other countries.
  - NFCI-to-assets ratios have largely remained stable.
  - A Federal German Court of Justice ruling in May 2021 against banks’ approach to raising service fees without customers’ explicit consent may have reduced NFCI generation.
- Impairment charges from Non-Performing Loans (NPLs) put little pressure on profitability because aggregate NPL ratios remain low by EU standards.
- Cost efficiency:
  - Cost-to-income ratio in Germany was 72 percent in 2020.
  - EU average cost-to-income ratio was 63 percent in 2020.
  - Cost-to-asset ratio for German banks was 1.3 percent in 2020.
  - European average cost-to-asset ratio was 1.2 percent in 2020.
- Aggregate underperformance largely reflects low revenues: declining NII and stagnant NFCI.

### Bank performance by type of credit institution
- Aggregate system-wide profitability masks large heterogeneity across institution types.
- Savings and cooperative banks have continued to outperform commercial banks; private banks have been substantially outperformed by savings and cooperative banks.
- Factors supporting savings and cooperative banks’ performance:
  - Ownership structure and mandates differ from private banks; they prioritize stakeholder welfare over profit maximization.
  - Operate on a regional principle and typically do not compete with institutions of the same pillar, reducing within-pillar competition.
  - Share platforms and services across the pillars, improving fixed cost management and enabling smaller banks to operate at lower costs.
  - Cross-guarantee networks provide safety nets for savers.
  - Sector-wide pooling of resources (e.g., DekaBank and DZ bank) limits costs.
- Commercial banks face more intense competition both within their pillar and across pillars.

### Restructuring of major private banks (Deutsche Bank and Commerzbank)
- Private commercial banks, especially Deutsche Bank and Commerzbank, suffered large losses during the GFC from relatively risky investments and faced fines and settlements.
- Strategic restructuring measures focused on headcount and branch reduction, digitalization, and de-risking.
  - Between 2006 and 2020, balance sheet size reduced by 13 percent for Deutsche Bank and by 17 percent for Commerzbank.
  - Deutsche Bank’s headcount declined by about 17 percent over the last 10 years.
  - Branch networks shrunk by almost 40 percent for each institution between 2010-20 (domestically and internationally).
- Business models shifted toward more traditional domestic banking with greater reliance on customer deposits and loans; derivatives and securities portfolios declined substantially while loans remained broadly stable.
- Cost-reduction measures are expected to be reflected in profitability over the medium-term.

### Savings and cooperative banks: lending growth and exposures
- After the GFC, savings and cooperative banks increased lending to support recovery and gained market share in lending to the private sector.
- Distribution of assets by bank type changed little, but distribution of loans to the private sector changed noticeably.
  - Joint market share of savings and cooperative banks in private sector loans increased from about 43 percent to 52 percent during 2010-21Q3.
- Between end-2008 and 2021Q3:
  - Savings and cooperative banks increased total assets by 45 percent and risk-weighted assets by 43 percent.
  - Credit cooperative banks increased total assets by 77 percent and risk-weighted assets by 68 percent.
- As a result:
  - Ratio of risk-weighted-to-total assets remained broadly stable for savings banks.
  - Ratio slightly increased in credit cooperative banks.
  - Commercial banks substantially reduced their risk-weighted-to-total-assets ratios as part of deleveraging.
- Elevated loan volume growth helped savings and cooperative banks generate additional income streams offsetting declining interest rates.
  - By 2021Q3, lending to domestic households and private enterprises grew 4.2 percent year-on-year, of which about 3.2 percentage points were attributed to savings and cooperative banks.
  - Housing loans grew by 7.3 percent year-on-year, to which savings and cooperative banks contributed 4.8 percentage points.
- Increased exposure to real estate makes savings and cooperative banks highly exposed to real estate price correction.

*Source: INTERNATIONAL MONETARY FUND.*

### 24.      Accelerated lending increased the exposure of savings   and cooperatives banks to real

### 24.      Accelerated lending increased the exposure of savings   and cooperatives banks to real estate

### Exposure to residential and commercial real estate
- For savings banks, housing loans increased from 29 to 35 percent of assets between 2008-21Q3.
  - Lending to households increased from 21 to 24 percent of assets.
  - Lending to corporates increased from 7 to 11 percent of assets.
- For credit cooperatives, housing loans increased from 29 to 36 percent of assets between 2008-21Q3.
  - Loans to households increased from 23 to 26 percent of assets.
  - Loans to corporates increased from 6 to 10 percent of assets.
- Savings banks’ lending to commercial real estate sectors (construction, housing enterprises, other real estate enterprises) increased from 9.1 to 14.1 percent of assets during 2008-21Q3.
- Credit cooperatives’ lending to the commercial real estate sectors increased from 5.4 to 11.8 percent of assets during 2008-21Q3.
- Most of the commercial real estate increase was attributed to housing enterprises.

### Drivers of loan growth and housing price dynamics
- During 2010-21, country-wide owner-occupied housing prices increased by some 60 percent in real and over 90 percent in nominal terms.
- In the largest 7 cities, house prices increased by 143 percent during 2010-21.
- Primary drivers cited:
  - Supply shortages.
  - Income growth over the last decade supporting demand.
  - Strong economic outlook reinforcing demand for real estate.
  - Immigration and urbanization increasing urban demand.
  - Housing supply constraints from construction sector labor shortages, limited land availability in larger cities, and shortages of construction materials due to supply disruptions.
- Consequences:
  - Surging housing backlogs.
  - Housing has become increasingly less affordable as income has not kept up with rising residential real estate (RRE) prices.

### Asset productivity and revenue composition
- Savings and cooperative banks generate higher income per unit of assets than commercial banks:
  - Both NII- and net commission income (NCI)-to-assets ratios of savings and cooperative banks are greater than those reported by commercial banks.
- Since the GFC, NII earnings of savings and cooperative banks have fallen more than those of commercial banks due to business model concentration on traditional maturity transformation (deposit-taking and lending).
- Efforts to diversify revenues through fee- and commission-based products produced subdued results: net commission income has grown in line with assets for most bank types.
- Savings and cooperative banks focus more on loans with customer deposits as the main funding source (asset and liability composition noted for 2021Q3).

### Profitability and cost dynamics
- Savings and cooperative banks generated about 90 percent of the industry’s profits in 2020.
  - Their joint share in industry profits increased from 30 percent in 2006 to over 90 percent in 2020.
- Elevated loan volume growth helped offset declining interest rates and maintain cost-to-income ratios for savings and cooperative banks.
  - Costs relative to income increased much less for savings and cooperative banks than for commercial banks.
  - Cost-to-income ratios for commercial banks, particularly big banks, "exploded" over the last few years.
  - While savings and cooperative banks can still cover their costs with NII, this income source is insufficient for commercial banks.
- Costs per unit of assets are high for savings and cooperative banks, reflecting a high-cost base with extensive branch networks and substantial employee presence.
  - Credit cooperatives and savings banks reduced costs per assets by 0.4 and 0.3 percentage points between 2013-20, respectively.

### Risk-adjusted profitability and benchmarking
- Three risk-adjusted profitability measures used:
  - Sharpe ratio (Sharpe, 1966): 퐸퐸(푟푟푖푖) − 푟푟푓푓 / 훿훿푖푖, where 푟푟푖푖 = return equity of bank i, 푟푟푓푓 = one-year German government bond rate, 훿훿푖푖 = standard deviation of return on equity of bank i.
  - Risk-adjusted performance (Modigliani and Modigliani, 1997): 푅푅푅푅푅푅푖푖 = (훿훿푚푚/훿훿푖푖) * (푟푟푖푖 − 푟푟푓푓) + 푟푟푓푓, where 훿훿푚푚 = standard deviation of return on equity of benchmark European comparator sample.
  - Return on risk-weighted assets.
- Return on risk-weighted assets places German banks as weak performers relative to many European peers.
- Adjusting for volatility, German banks’ risk-adjusted profitability appears more favorable:
  - In terms of average ROE, German banks’ long-term profitability averages fall within the 20th percentile of the European comparator sample.
  - Estimated Sharpe ratios of German banks are within the 24th percentile of the European bank sample.
  - Credit cooperatives’ profitability in risk-adjusted terms is above the median of the European sample (60th percentile), attributed to higher average weighted returns and low volatility.
- German SSM-supervised significant institutions (SIs):
  - Average weighted risk-adjusted ROE of German SIs over 2001-20 was equivalent to the 28th percentile of the European SI sample.
  - Risk-adjusted Sharpe and Modigliani ratios for German SIs improved to the 35th percentile but remained below the EU average and median.

### Regression analysis: data, scope, and methodology
- Samples:
  - German MFIs: includes all German MFIs with available data (both LSIs supervised by BaFin/Deutsche Bundesbank and German SIs under SSM).
  - European SIs: banks classified as SIs.
- Data sources and period:
  - Bank-level dataset covering 2000-20; many regressions focus on 2008-20 due to data limitations prior to 2008.
  - Balance sheet and income statement data from FitchConnect and S&P Capital IQ.
  - Macroeconomic series from IMF’s World Economic Outlook and Haver Analytics.
  - Structural banking data from ECB’s Banking Structural Statistical Indicators.
- Estimation method:
  - Arellano-Bover/Blundell-Bond linear dynamic panel-data estimator with robust standard errors.
  - Model specification: 푌푘,푗,푡 = 훿푌푘,푗,푡−1 + 휑푘 + 휙′푋푘,푗,푡 + 휃′퐼푗,푡 + 휆′푀푗,푡 + 휀푘,푗,푡.
    - 푌푘,푗,푡 is the headline profitability measure for bank k, country j, time t.
    - Bank-fixed effects (휑푘) and time-varying bank-specific indicators (푋푘,푗,푡).
    - Industry structural indicators (퐼푗,푡) include industry concentration and measures of excess capacity.
    - Macroeconomic/cyclical variables (푀푗,푡) include GDP growth and short-term interest rates.
- Bank-specific indicators included:
  - Bank size: natural logarithm of total assets (and natural logarithm of number of employees for robustness).
  - Solvency: Tier 1 capital, equity-to-assets ratios, tangible common equity-to-total assets.
  - Asset quality: non-performing loans ratios (NPLs).
  - Cost efficiency: cost-to-income ratio.
  - Diversification: ratio of noninterest income-to-total income.
  - Business model proxies: loan-to-asset and deposit-to-asset ratios, share of wholesale funding, deposit and asset growth rates, income diversification measure.

*Source: IMF staff estimates, Bundesbank, ECB, FitchConnect, S&P Capital IQ, Haver Analytics, and IMF text contained in content unit 1deuea2022010.*

### 39.      Banking industry structural indicators capture competition and excess capacity.

### 39.      Banking industry structural indicators capture competition and excess capacity.

### Indicators and measurement
- Banking sector capacity is measured by branches per capita and banking sector employees per capita.
- Banking sector concentration, or competition, is measured by the Herfindhal index and the share of the largest 5 banks in total assets.
- The numerator of the tangible common equity-to-total asset ratio includes only high quality, loss-absorbing capital, and assets in the denominator are not risk-adjusted, which is relevant for cross-country comparison due to different risk weights for similar exposures across banks and countries (Detragiache et al, 2018).
- Economic activity is measured by the real GDP growth.
- Policy rates account for the monetary policy stance, where 3-months EURIBOR is used as a proxy. Due to EURIBOR remaining close to zero since the GFC, yield on a 2-year German government bond is used in the German bank sample as a proxy for monetary policy stance.
- The analysis controls for broader financial conditions using the Financial Conditions Index (FCI).

### European Significant Institution (SI) bank sample: scope and composition
- Sample is based on 112 SSM-supervised institutions operating in 19 countries.
- The twenty German SIs included in the sample contributed about 17 percent to the total sample assets.
- Note: Calculations based on a sample of 113 institutions classified as Significant Institutions (SI) as of January 1, 2021; two SIs (1 established in Germany and 1 in Luxembourg) are missing from the sample. Three institutions were excluded from the analysis due to data limitations in the data sources used, including one German institution.

### European SI empirical findings: profitability and business model
- Average returns of German SIs lag European peers; capacity to generate revenue falls short amid high-cost inefficiencies.
- On aggregate, overall income of German SIs is well-diversified (measured by non-interest operating income in percent of total income), more so than the income streams of several other European peers.
- Relative to assets, the non-interest income of German SIs is broadly similar to the sample average.
- Interest income of German SIs is subdued and falls short of European peers, reflecting a low share of loans in total assets.
- Conclusion: German SIs’ cost base is too high for the revenues generated.

### European SI regression results: drivers of profitability (summary)
- Larger banks with lower NPLs, higher capitalization, and better cost efficiency tend to be more profitable.
- Bank size (natural log of total assets) is positively correlated with bank profitability; coefficients statistically significant with ROAA as dependent variable but not with other profitability measures.
- Higher capitalization (tangible common equity-to-tangible assets) is positively correlated with higher profitability; results statistically significant across most model specifications. Interpretation cautioned: could reflect lower funding costs for better-capitalized banks or profitable banks recapitalizing faster via retained earnings.
- Better asset quality (lower NPL ratios) is associated with higher profitability; statistically significant in most specifications.
- Better cost efficiency (lower cost-to-income ratio) correlates with higher profitability.
- Business model variables (income diversification, share of wholesale funding) are not statistically significant in most specifications.
- Structural indicators: number of bank branches per capita found negatively correlated with ROAA; bank concentration (share of largest five banks) positively correlated with profitability. However, coefficients on structural indicators were neither robust nor statistically significant across different model specifications.
- Estimation approach: Arellano-Bover/Blundell-Bond linear dynamic panel-data estimation; Arellano-Bond with robust clustered standard errors produced similar results.

### European SI sample statistics and visuals (as presented)
- Figures illustrate: Net Fees & Commissions (In percent of average total assets), Income Diversification (In percent), Gross loans (In percent of total assets), Net Interest Income (In percent of average total assets) across countries including Germany, Austria, Spain, Finland, Belgium, Portugal, France, Greece, Italy, Netherlands, Luxembourg.
- Figure 14 highlights that returns of German SIs lag peers both relative to total assets and to risk-weighted assets; balance sheet composition comparable to SIs in other countries; an elevated cost base relative to income depletes revenues of German SIs.

### German bank sample: scope
- Sample based on 1,358 banks operating in Germany as of mid-2021.
- Includes commercial banks, savings banks, credit cooperatives, building and loan associations, mortgage banks, Landesbanken, a regional institution of credit cooperatives, and banks with special, development and other central support tasks.
- Data availability varies by bank and year.

### German bank empirical findings: profitability drivers
- Regression analysis suggests smaller, better-capitalized banks operating with better cost efficiency are likely to be more profitable.
- Bank size (total assets and number of employees) is negatively correlated with bank profitability; estimated coefficients not statistically significant in most regression specifications.
- Size becomes more relevant for savings banks and credit cooperatives: coefficients become significant when sample is restricted to these bank types, reflecting observed trend that smaller banks (savings banks and credit cooperatives) reported higher profitability ratios than larger banks.
- Cost efficiency (cost-to-income ratio) is negatively correlated with profitability.
- Asset quality (NPL ratio) negatively correlated with profitability but not statistically significant in the German bank sample.
- Capitalization (tangible common equity-to-tangible total assets and equity-to-assets) coefficients are not statistically significant across model specifications.

### German bank regression results: macro and industry influences
- Real GDP growth and monetary policy stance (2-year government debt yield) are both strongly correlated with higher bank profitability in most regression specifications.
- Positive correlation interpretation: stronger economic growth raises loan demand and borrowers’ ability to service debt; higher interest rates raise interest margins.
- The correlation with the 2-year government bond yield appears largely driven by credit cooperatives (statistically significant coefficient for credit cooperatives sub-sample).
- Greater bank concentration (market share of largest 5 institutions) is positively correlated with profitability.
- Alternative specifications: using EURIBOR or short-term interest rates produced similar results; yield curve steepness and FCI were not statistically significant in alternative specifications.
- Given less persistence in profitability in the German bank sample, robustness checks used OLS estimators and excluding 2020 (COVID-19 year) did not alter results.

### Estimation notes and sample sizes
- European SI estimates reported using Arellano-Bover/Blundell-Bond linear dynamic panel-data estimation; Arellano-Bond with robust clustered standard errors also used.
- German bank sample estimates reported using Arellano-Bond linear dynamic panel-data estimation with robust standard errors adjusted for clustering by bank; OLS used for robustness.
- Reported observation counts and numbers of banks vary by specification (examples in tables): European SI observations e.g., 1,135; German sample observations e.g., 7,350 and number of banks e.g., 1,237 in table summaries.

*Source: IMF staff analysis as presented in the chapter "Banking industry structural indicators capture competition and excess capacity."*

### 48.      Over the past few years, German banks have taken several steps to cut costs and

### 1deuea2022010 - 48.

### Cost cutting, revenue diversification, and consolidation
- Over the past few years, German banks have taken several steps to cut costs and increase revenues.
- Customer preference for brick-and-mortar models historically slowed aggressive cost cuts, but urbanization, migration, and COVID-19 accelerated shifts toward cheaper and faster digital services and asset management products, supporting potential future noninterest fee earnings.
- Since the GFC, the number of credit institutions declined by about a quarter (over 500 institutions).
- Institutions per million inhabitants declined from 20 to 15.
- In end-2020, Germany had about 0.28 branches and 6.9 bank employees per thousand inhabitants.
- Savings and cooperative banks rely on resources-sharing (IT support, customer services, audit functions), mortgage origination software, sector-wide insurers, asset managers, DekaBank and DZ Bank, and cross-guarantee networks to reduce costs and achieve economies of scale.

### Branch network, headcount, and operational changes
- Banks have substantially reduced branch networks and employee headcount; some operate co-branded branches to reduce maintenance and fixed costs.
- Digital transformation investments increased during COVID-19, expected to raise costs in the short run but generate medium-term benefits.

### Digitalization, fintech, and cyber risk
- Banks are utilizing technologies including digital wallets, cloud computing, big data analytics, and artificial intelligence, but progress is limited relative to other European peers.
- COVID-19 accelerated IT upgrades and digital adoption; customers increasingly compare products and pricing across institutions.
- The German fintech sector has grown dynamically; fintechs encroach on banks’ businesses (e.g., payments), heightening competition.
- Digitalization increases cyber risk: an EBA survey (Fall of 2021) reported 55 percent of interviewed banks expected an increase in operational risk, specifically cyber risk and data security issues (90 percent), followed by conduct and legal risk (42 percent).

### Interest rates, deposit behavior, and net interest income
- Greater pass-through of negative interest rates to clients has helped slow net interest income decline, though delays occurred amid competition and fear of losing market share.
- Interest rates on new deposits of non-financial corporations (NFCs) (up to 1 year) declined to -0.54 percent by end-2021.
- Reduction in deposit interest expenses has not fully offset declining interest revenues, resulting in shrinking interest margins.
- Faster transmission of rising interest rates into liabilities than assets could keep low banking profitability persisting over the short-term.
- Banks have been increasingly funded by sight deposits; longer-term deposits declined in appeal amid low interest rates.
- Commercial banks more commonly use interest rate swaps to hedge interest rate risk; credit cooperatives and savings banks are more exposed.

### Real estate exposures and credit risks
- Divergent dynamics in real estate: residential prices climbed through 2021 while some commercial property prices declined; office prices registered a small increase in 2021Q4; retail prices continued to decline.
- Immediate risks are concentrated in non-food retail and hotels.
- Large exposures of many German banks to CRE mean continued CRE price declines would likely reduce collateral values and translate into higher NPL ratios.
- High residential property valuations and estimates of RRE overvaluation suggest loan collateral may be overvalued, particularly in larger cities.
- A real estate price correction combined with a substantial increase in the unemployment rate could force mortgages into negative equity and raise NPLs, with savings and cooperative banks more exposed due to larger shares of housing loans.
- Reportedly conservative collateral valuations, extended fixed-rate mortgage periods, and full recourse on residential mortgages mitigate borrower default risk.

### Profitability risks, provisioning, and risk-taking incentives
- Risks and vulnerabilities are tilted to the downside despite potential profitability gains from private banks’ restructuring.
- Downside risks include another COVID outbreak, worsening economic developments, monetary policy adjustments, potential collateral overvaluation and price correction of residential and commercial real estate, and possible underestimation of credit risk.
- A potential resurgence of COVID-19 could lead to losses and weaker profitability, with further asset quality deterioration in sectors like hotels and restaurants.
- Low profitability may encourage risk-taking (search for yield), greater loan volume growth, accumulation of less liquid loans, and lengthening loan maturities, increasing liquidity and maturity risk.
- By end-2021, the Bundesbank’s survey indicated, on average, no changes in banks’ loan approval criteria, though pockets of vulnerability may exist.
- Loan loss provisions increased in 2020 then largely reversed in 2021, returning to pre-pandemic levels; relatively low provisions amid rising vulnerabilities raise questions of potential underestimation of credit risks.

### Regulatory responses and buffers
- In response to rising vulnerabilities (including residential real estate), BaFin set a countercyclical capital buffer of 0.75 percent of risk-weighted assets on domestic exposures and introduced a sectoral systemic risk buffer of 2.0 percent of risk-weighted assets on loans secured by residential real estate. The rates are currently at zero percent. Institutions have to meet the additional capital requirements by February 1, 2023.
- Phasing out of ECB support measures such as TLTRO III may also impact bank profitability.

### Demographics and structural shifts in demand
- Conservative preferences and an ageing population have historically favored brick-and-mortar models, but urbanization and migration shifted population growth to larger cities where customers prefer cheaper and faster digital services.
- Continued migration of clients to populated areas may force smaller rural institutions out of business.
- Over the long term, declining population and rising old-age dependency ratios may slow demand for financial services, curtail balance sheet growth, and reduce income generation.

*Source: IMF content unit 1deuea2022010 - 48.*

### 67.      Sustainability risks –   namely environmental, social, and governance –   could pose

### 67.      Sustainability risks –   namely environmental, social, and governance –   could pose

### Sustainability risks: scope, channels, and preparedness
- A BaFin survey suggests that the majority of German entities recognize the need of a comprehensive appraoch to potential sustainability risks, with climate and environmental risks receiving marginally more attention and accounting for both physical and transition risks.
- Climate risks for banks could stem from:
  - potential losses in the value of existing assets due to extreme weather events (physical risks); and
  - changing patterns of consumption and investment arising in connection with the transition to a climate-neutral economy (transition risks). 46
- Most institutions, however, are yet to incorporate fully climate risks into their business practices. 47
- BaFin’s report referenced: “Germany’s financial sector and the issue of sustainability risks: A status survey conducted by BaFin”.
- ECB milestone: the “ECB Guide on climate-related and environmental risks” (published 27 November 2020) set minimum disclosure expectations; a 2020 ECB stocktaking found virtually none of the institutions in scope would meet that minimum, while a subsequent assessment (2022) revealed clear progress toward the goal. 47

### Geopolitical risks and transmission to banks
- Geopolitical developments, notably the war in Ukraine, could delay economic recovery and pose risks to German banks.
- Direct exposures of German banks to Russia and Ukraine are limited, but the war is likely to exert a material drag on GDP growth through:
  - higher energy prices,
  - tighter financial conditions,
  - elevated uncertainty,
  - disruptions in supply chains, including energy supply, and
  - rising inflation.
- Potential bank impacts include weaker clients’ ability to service debt, leading to higher NPLs and lower profitability.

### Conclusions on profitability and resilience
- The structurally low profitability of the German banking sector remains a concern for financial stability.
- Risk-adjusted profitability measures suggest German banks’ performance appears more favorable because low profitability is partially compensated by lower volatility and risk-taking, but most banks fall below the comparator group averages and medians.
- Using other profitability measures (returns on assets, equity, and risk-weighted assets), German banks rank among the lowest in Europe (Figure 1 referenced).
- Low profitability undermines banking sector resilience because banks may be more likely to use capital rather than earnings to absorb losses in case of a shock. 69

### Risks to savings and cooperative banks; data needs
- Rising risks and vulnerabilities challenge the sustainability of savings and cooperative banks’ business models and call for greater risk monitoring. 70
- Past revenue support from increased lending—particularly for residential and commercial real estate—helped these banks offset declining interest rates.
- Going forward, rising risks—particularly those stemming from real estate—challenge the sustainability of such business models.
- Other rising risks include interest rate changes, cyber risks, and geopolitical developments.
- Data collection efforts should continue, including granular information on:
  - exposures to residential real estate, particularly in markets with greater estimated overvaluation (such as larger cities);
  - exposures to commercial real estate sub-sectors (such as office, logistics, food/non-food retail properties).
- The impact of risk materialization—such as price correction in RRE/CRE and interest rate normalization—on banks should be examined closely.

### Policy recommendations to restore sustainable profitability and resilience
- Implement a combination of cost-reducing and income-generative measures to develop sustainable business models; cost-reduction measures alone have limits and may not offset declining revenues. 71
- Scale-up profitability measures:
  - Headcount and branch network optimization combined with digitalization to reduce cost inefficiencies and create new business opportunities. 72
  - Review scope for greater process automation and product standardization to cut costs and face competition from fintech; assess replacing tailored with more standardized products while ensuring such measures do not worsen adverse selection problems. 73
  - Income generation through revenue diversification and greater reliance on fees to bolster profitability; low-interest rates have prompted more customer interest in fee-based products, which could boost banks’ profits as interest rate stabilization reduces gains on securities. 74
- Support market-led consolidation to attain economies of scale; continued consolidation resulting in fewer but more profitable institutions is crucial. According to the EBA questionnaire, around 60 percent of surveyed European banks reported, considering or have considered, mergers and acquisitions. Obstacles to cross-border consolidation should be reduced where possible (differences in reporting requirements, insolvency schemes, taxation regimes, AML/CFT, data privacy laws, etc.). 75

*Source: 1deuea2022010 - 67.      Sustainability risks –   namely environmental, social, and governance –   could pose (IMF PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2022/english/1deuea2022010.pdf_