## 3. Higher for Longer: What Are the Macrofinancial Risks?

## Source details

**Canonical URL:** [3. Higher for Longer: What Are the Macrofinancial Risks?](https://www.imf.org/-/media/files/publications/reo/mcd-cca/2023/october/english/ch3.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/mcd-cca/2023/october/english/ch3.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/mcd-cca/2023/october/english/ch3.pdf.json)

---

### Overview
- Core inflation remains above central bank targets in many ME&CA countries, prompting the possibility of a prolonged period of tighter monetary policy.
- A higher-for-longer interest rate environment could produce unintended consequences for financial systems, including credit quality deterioration, liquidity strains, and capital losses from mark-to-market on government bond holdings.
- Assessment finds regional banking systems broadly resilient under adverse scenarios, but with important pockets of vulnerability, especially among state-owned banks.

### Factors that Could Exacerbate Financial Stability Risks
- Reliance on foreign funding:
  - Nonresident deposits and other foreign liabilities increase vulnerability to sudden shifts in investor sentiment.
  - Countries with greater dependence on external funding include GCC (Bahrain, Qatar) and CCA (Georgia).
  - Formal deposit insurance varies across the region; some countries lack it.
  - Large shares of government deposits and government ownership of major banks (Azerbaijan, Egypt, Saudi Arabia) can mitigate deposit-run risks in some cases.
- Sovereign-bank nexus:
  - Elevated bank holdings of domestic sovereign debt (Algeria, Egypt, Pakistan) create interest rate and credit spillover channels from sovereigns to banks.
  - Higher sovereign borrowing costs and worsening sovereign credit could prompt domestic banks to increase holdings of government debt, crowding out private sector credit.
  - Credit to the private sector is often lower where banks are more exposed to sovereign debt.
- Corporate credit quality and bank buffers:
  - State-owned bank performance remains well below prepandemic levels in MENA EM&MIs and Pakistan and, to a lesser extent, in the GCC.
  - CCA bank profitability has surged above prepandemic trends in part because of inflows from Russia.
  - Capital ratios remain well above regulatory minimums across ME&CA; nonperforming loan ratios are mostly contained but elevated for state-owned banks in MENA EM&MIs and Pakistan.
  - Banks have ample liquidity buffers, strengthened by higher oil prices in oil-exporting countries.
- Hidden vulnerabilities:
  - Accounting rules, regulatory treatments, and concentration in certain asset classes (for example, government bonds) can mask exposures and losses.
  - Reliance on foreign funding sources is a key vulnerability; funding can evaporate rapidly, as shown by recent stress in some advanced economies.

### Corporate Sector Stress Test: Key Findings
- Definition and scope:
  - Focus on “zombie” firms (average interest coverage ratio over two years below 2.5 and receiving subsidized lending).
- Share of zombie firms and balance-sheet metrics:
  - About 12 percent in 2022.
  - Zombie firms’ median leverage (total liabilities to total assets) at the end of 2022: 40 percent versus 20 percent for other firms.
  - At the end of 2022, zombie firms held about 12 percent of corporate debt.
- Adverse scenario calibration:
  - Effective interest rate increased sequentially by 100 basis points per year, reaching an average of more than 8 percent by the end of 2024, about 2 percentage points higher than prepandemic levels.
  - Sector-specific profitability shocks calibrated from post–global financial crisis earnings, with most sectors experiencing double-digit negative returns.
- Stress test outcomes:
  - Firm profitability could decline, on average, to about 3 percent in 2024, below the prepandemic level of 5 percent.
  - Median interest coverage ratio estimated to decline from 3.5 to 1.5 at the end of 2024.
  - Share of debt at risk of default increases from about 12 percent of total debt in 2022 to almost 30 percent by 2024.
  - Sectoral vulnerabilities: transportation, capital goods, and food and beverages most affected; the food and beverages sector shows a particularly sharp increase in median share of zombie firm debt by 2024.

### Banking Sector Stress Test: Scenarios and Results
- Scenarios simulated amid higher-for-longer interest rates:
  1. Liquidity shock through deposit outflows (baseline).
     - Simulated withdrawals: 20 percent of resident deposits; 30 percent of foreign deposits and wholesale funding.
  2. Liquidity shock + 200 basis point increase in interest rates.
  3. Corporate sector stress mapped to higher provisioning needs.
  4. Combined scenario: higher interest rates + corporate sector stress + liquidity shock.
- General findings:
  - Banking systems would be resilient to individual stress scenarios, but vulnerabilities emerge under combined shocks.
  - GCC and MENA EM&MIs and Pakistan: resilient to individual scenarios but could be tested by combined shock.
  - CCA banks: negligible losses across scenarios due to large cash buffers and high profitability, but exposed to foreign exchange risks (high dollarization, unhedged FX borrower exposures, FX funding stress).
  - State-owned banks are more vulnerable than privately owned banks in MENA EM&MIs and Pakistan and to a lesser extent in the GCC, reflecting lower profitability and higher securities holdings.
- Specific outcomes:
  - Liquidity stress scenario: losses relatively small.
  - Liquidity + higher rates: much larger losses, driven by unrealized capital losses on fixed-income securities, especially long-duration securities (Jordan, Morocco, Saudi Arabia). Countries with lower ex ante capital buffers more exposed (Egypt, Morocco).
  - Corporate sector stress: banks have ample buffers in many cases (for example, Kuwait due to relatively high provisioning; low private sector exposure in Egypt, Pakistan).
  - Combined scenario: largest capital losses.
    - Privately owned banks experience losses of 16.7 percent in MENA EM&MIs and Pakistan and 11.8 percent in the GCC.
    - State-owned banks experience losses about twice as large as privately owned banks.
  - Across all scenarios, CCA losses are negligible, but external shocks could trigger significant vulnerabilities through FX channels.

### Box: Banking Sector Stress Test for the Caucasus and Central Asia (CCA)
- Context and added assumptions:
  - Inflows from Russia normalize, and profitability returns to its prepandemic average amid a normalization of net foreign currency gains.
  - An adverse external shock triggers a rise by 150 percent in sovereign spreads.
  - Exchange rates depreciate by 30 percent, stressing unhedged corporate and household borrowers.
- Key quantitative results (losses and capital impacts):
  - Liquidity stress scenario: Losses of between 4 and 7 percent of regulatory capital.
  - Corporate stress with large currency depreciation: Losses rise to 7.3 percent of regulatory capital.
  - Combined liquidity and corporate stress: Losses as large as 15.3 percent of regulatory capital.
  - Undercapitalization and aggregate capital ratio effects:
    - 1.8 percent of banks (weighted by assets) would become undercapitalized.
    - Capital ratios would decline by 2.6 percentage points in aggregate, from 17.4 percent to 14.8 percent.
  - Corporate stress scenario (aggregate Tier 1 impacts): Aggregate Tier 1 capital ratio would drop by 1.2 percentage points.
  - Profitability shock without currency depreciation: Without currency depreciation, the decline would be negligible even if profitability returns to prepandemic averages (a 44 percent decline from current levels).
- Vulnerabilities and drivers:
  - Dollarization and unhedged foreign exchange borrowers are key drivers of losses.
  - CCA banks resilient to small shocks due to high profitability and cash buffers; main vulnerabilities stem from a surge in sovereign spreads and sizable currency depreciations.
- Scope and data:
  - Countries included: Georgia and Kazakhstan.
  - Data sources: Fitch Connect; IMF, Financial Soundness Indicators database; and IMF staff calculations.

### Excess Losses: Concentration and Bank Characteristics
- Losses measured in excess of net income as a fraction of Tier 1 regulatory capital.
- Loss concentration higher for banks with:
  - Relatively illiquid balance sheets.
  - Low profitability (RoA defined as net income over total assets).
  - Low provisioning levels.
  - Higher leverage.
  - Higher within-country market shares.
- Banks in countries with greater duration of outstanding sovereign bonds face much higher sensitivity to interest rate increases.
- State-owned banks (SOB defined as banks with at least 50 percent government ownership) are separately assessed.
- Bank undercapitalization and capital ratio impacts:
  - Few banks would become undercapitalized in the combined scenario, but capital buffers would be significantly eroded, especially in countries that begin with relatively low capital ratios (Egypt, Morocco, Pakistan).
  - "18 percent of banks in MENA EM&MIs and Pakistan would become undercapitalized in the combined scenario."
  - All banks in the GCC and the CCA would remain above minimum requirements in the combined scenario.
  - Imposing country-specific minimum capital requirements increases the share of undercapitalized banks slightly in the GCC and MENA EM&MIs and Pakistan.
  - Aggregate Tier 1 capital ratios would decline by:
    - "2.6 percentage points for privately owned banks in MENA EM&MIs and Pakistan."
    - "4.5 percentage points for state-owned banks in MENA EM&MIs and Pakistan."
    - In the GCC, ratios decline by "3.7 percentage points" for state-owned banks and "2.0 percentage points" for privately owned banks.
  - Basel III minimum capital requirement referenced: "4.5 percent common equity Tier 1 plus 1.5 percent additional Tier 1 capital and a 2.5 percent capital conservation buffer."
  - Most countries included in the stress testing exercise require minimum Tier 1 capital higher than "8.5 percent."

### Impact on Credit Provision and Output
- Historical relationship:
  - A "1 percentage point decline in capital ratios" has been historically correlated with a contraction in real credit, reaching "1.2 percent after eight quarters."
- Using the combined scenario estimates:
  - Real credit could contract by "4.3 percent in MENA EM&MIs and Pakistan" over a two-year horizon.
  - Real credit could contract by "3.2 percent in the GCC" over a two-year horizon.
- Macro model results on output:
  - Median output loss from the decline in lending is estimated to be about "0.5 percent" in MENA EM&MIs and Pakistan, and in the GCC.
  - At the 95th percentile of estimated output losses:
    - There could be a "1.5 percent output contraction in MENA EM&MIs and Pakistan" over two years.
    - There could be a "0.9 percent contraction in the GCC" over two years.
- Caveats:
  - Banking stress test does not explicitly model interbank linkages due to lack of data; estimates should be interpreted as lower bounds.
  - Amplification could occur through nonbank financial institutions, though expected to be limited because they have a low market share in most countries of the region.

### Macroprudential Frameworks: Current Stance and Gaps
- Use of macroprudential policies in ME&CA has generally been slow despite broad-based tools being available:
  - Almost all countries have broad-based tools covering capital buffers (e.g., countercyclical capital buffer), but most have left the setting at "zero since inception."
  - Most countries have implemented some borrower-based tools for the household sector (for example, caps on debt-service-to-income ratios).
  - Tools to guard against vulnerabilities and elevated credit risk in the nonfinancial corporate sector have generally been less used.
  - Some GCC countries (for example, Saudi Arabia) have not used measures to reduce banks’ foreign currency liquidity risks.
  - Some MENA countries (Algeria, Lebanon, Morocco, Tunisia) have taken fewer actions to reduce risks from domestic systemically important financial institutions.
- Regional variation in macroprudential action may reflect different risk profiles but can leave key gaps in addressing vulnerabilities.

### Policies to Safeguard Financial Stability (Recommendations)
- Strengthen and develop macroprudential frameworks:
  - In MENA and Pakistan: ramp up the use of broad-based macroprudential tools such as the countercyclical capital buffer.
  - Countries with elevated corporate debt-at-risk or prevalent zombie firms (for example, Kuwait, Jordan, United Arab Emirates) should consider borrower-based tools such as caps on debt-service-to-income ratios and loan-to-value ratios.
  - Implement measures targeting large domestic systemically important institutions (e.g., increased capital surcharges), especially in MENA EM&MIs.
  - In the GCC: guard against unexpected liquidity stress related to foreign liabilities; consider tools that account for concentrated nonresident deposit bases in liquidity coverage and net stable funding ratios; consider enhanced macroprudential foreign exchange measures such as reserve requirements.
  - In the CCA: continue ongoing macroprudential measures to build resilience across credit cycles, incentivize de-dollarization, and reduce foreign exchange mismatches.
  - Monitor links between the banking sector and nonbank financial institutions where risks may be migrating.
- Address vulnerabilities from the sovereign-bank nexus:
  - Near term: preserve bank capital, conduct bank stress tests considering multiple nexus channels, pay attention to bank asset classification, provisions, and exposures to interest rate and liquidity risks; where risks are elevated, restrict profit distribution plans as first-line defense.
  - Medium term: in countries with limited fiscal space, pursue macroeconomic policies that strengthen debt sustainability to contain government financing needs and limit deterioration of the sovereign-bank nexus.
  - Where bank holdings of sovereign bonds exceed concentration limits (for example, Egypt, Pakistan), consider gradual measures including imposing capital surcharges on banks’ sovereign bond holdings above certain thresholds.
  - Across all countries: foster a deep and diversified investor base, build adequate buffers at state-owned banks, provide clear mandates, and align supervisory tools such as stress tests with banks’ unique risk profiles.
- Communication and crisis preparedness:
  - Clear communication of central bank objectives and policy functions is crucial to avoid unnecessary uncertainty.
  - Improve communication on macroprudential frameworks (for example, by issuing a financial stability report).
  - If monetary policy is adjusted for financial stability purposes, clearly communicate the intent to return inflation to target as soon as possible once stress lessens.
  - Prepare emergency liquidity tools:
    - In countries where central banks lack explicit authority to provide emergency liquidity assistance (Algeria, Morocco, Oman), governments should prioritize establishing a clear framework for dealing with liquidity distress.
    - In countries where central bank laws allow emergency liquidity but lack operational directives (Egypt, Jordan in MENA; Georgia, Tajikistan in CCA), provide specific instructions and internal guidelines for the use of emergency liquidity facilities, including foreign exchange liquidity.
  - Design liquidity support to address liquidity, not solvency; interventions should have well-defined end dates, be parsimonious, and be appropriately priced to minimize moral hazard and avoid conflicting with monetary tightening.
  - Ensure liquidity support and crisis management tools comply with Islamic banking rules where applicable.
- Strengthen resolution frameworks:
  - Enhance insolvency procedures to address legacy nonperforming loans.
  - Structure resolution regimes to enable swift resolution of nonperforming loans to prevent evergreening and the emergence of zombie firms.

*Source: IMF Regional Economic Outlook—Middle East and Central Asia, October 2023 (chapter 3).*

### 3. Higher for Longer: What Are

### 3. Higher for Longer: What Are the Macrofinancial Risks?

### Overview
- Core inflation remains above central bank targets in many ME&CA countries, prompting the possibility of a prolonged period of tighter monetary policy.
- A higher-for-longer interest rate environment could produce unintended consequences for financial systems, including credit quality deterioration, liquidity strains, and capital losses from mark-to-market on government bond holdings.
- The chapter assesses banking sector resilience to credit and liquidity risks in a higher-for-longer environment and finds regional banking systems broadly resilient under adverse scenarios, but with important pockets of vulnerability, especially among state-owned banks.

### Factors that Could Exacerbate Financial Stability Risks
- Reliance on foreign funding:
  - Nonresident deposits and other foreign liabilities increase vulnerability to sudden shifts in investor sentiment.
  - Countries with greater dependence on external funding include GCC (Bahrain, Qatar) and CCA (Georgia).
  - Formal deposit insurance varies across the region; some countries lack it.
  - Large shares of government deposits and government ownership of major banks (Azerbaijan, Egypt, Saudi Arabia) can mitigate deposit-run risks in some cases.
- Sovereign-bank nexus:
  - Elevated bank holdings of domestic sovereign debt (Algeria, Egypt, Pakistan) create interest rate and credit spillover channels from sovereigns to banks.
  - Higher sovereign borrowing costs and worsening sovereign credit could prompt domestic banks to increase holdings of government debt, crowding out private sector credit.
  - Credit to the private sector is often lower where banks are more exposed to sovereign debt.
- Corporate credit quality and bank buffers:
  - State-owned bank performance remains well below prepandemic levels in MENA EM&MIs and Pakistan and, to a lesser extent, in the GCC.
  - CCA bank profitability has surged above prepandemic trends in part because of inflows from Russia.
  - Capital ratios remain well above regulatory minimums across ME&CA; nonperforming loan ratios are mostly contained but elevated for state-owned banks in MENA EM&MIs and Pakistan.
  - Banks have ample liquidity buffers, strengthened by higher oil prices in oil-exporting countries.
- Hidden vulnerabilities:
  - Accounting rules, regulatory treatments, and concentration in certain asset classes (for example, government bonds) can mask exposures and losses.
  - Reliance on foreign funding sources is a key vulnerability; funding can evaporate rapidly, as shown by recent stress in some advanced economies.

### Corporate Sector Stress Test: Key Findings
- Focus on “zombie” firms (average interest coverage ratio over two years below 2.5 and receiving subsidized lending).
- Share of zombie firms:
  - About 12 percent in 2022.
  - Zombie firms’ median leverage (total liabilities to total assets) at the end of 2022: 40 percent versus 20 percent for other firms.
  - At the end of 2022, zombie firms held about 12 percent of corporate debt.
- Adverse scenario calibration:
  - Effective interest rate increased sequentially by 100 basis points per year, reaching an average of more than 8 percent by the end of 2024, about 2 percentage points higher than prepandemic levels.
  - Sector-specific profitability shocks calibrated from post–global financial crisis earnings, with most sectors experiencing double-digit negative returns.
- Stress test outcomes:
  - Firm profitability could decline, on average, to about 3 percent in 2024, below the prepandemic level of 5 percent.
  - Median interest coverage ratio estimated to decline from 3.5 to 1.5 at the end of 2024.
  - Share of debt at risk of default increases from about 12 percent of total debt in 2022 to almost 30 percent by 2024.
  - Sectoral vulnerabilities: transportation, capital goods, and food and beverages most affected; the food and beverages sector shows a particularly sharp increase in median share of zombie firm debt by 2024.

### Banking Sector Stress Test: Scenarios and Results
- Four stress scenarios simulated amid higher-for-longer interest rates:
  1. Liquidity shock through deposit outflows (baseline).
     - Simulated withdrawals: 20 percent of resident deposits; 30 percent of foreign deposits and wholesale funding.
  2. Liquidity shock + 200 basis point increase in interest rates.
  3. Corporate sector stress mapped to higher provisioning needs.
  4. Combined scenario: higher interest rates + corporate sector stress + liquidity shock.
- General findings:
  - Banking systems would be resilient to individual stress scenarios, but vulnerabilities emerge under combined shocks.
  - GCC and MENA EM&MIs and Pakistan: resilient to individual scenarios but could be tested by combined shock.
  - CCA banks: negligible losses across scenarios due to large cash buffers and high profitability, but exposed to foreign exchange risks (high dollarization, unhedged FX borrower exposures, FX funding stress).
  - State-owned banks are more vulnerable than privately owned banks in MENA EM&MIs and Pakistan and to a lesser extent in the GCC, reflecting lower profitability and higher securities holdings.
- Specific outcomes:
  - Liquidity stress scenario: losses relatively small.
  - Liquidity + higher rates: much larger losses, driven by unrealized capital losses on fixed-income securities, especially long-duration securities (Jordan, Morocco, Saudi Arabia). Countries with lower ex ante capital buffers more exposed (Egypt, Morocco).
  - Corporate sector stress: banks have ample buffers in many cases (for example, Kuwait due to relatively high provisioning; low private sector exposure in Egypt, Pakistan).
  - Combined scenario: largest capital losses.
    - Privately owned banks experience losses of 16.7 percent in MENA EM&MIs and Pakistan and 11.8 percent in the GCC.
    - State-owned banks experience losses about twice as large as privately owned banks.
  - Across all scenarios, CCA losses are negligible, but external shocks could trigger significant vulnerabilities through FX channels.

### Policy Recommendations and Mitigation Measures
- Strengthen macroprudential frameworks to address systemic vulnerabilities and hidden exposures.
- Contain vulnerabilities stemming from the sovereign-bank nexus to prevent adverse feedback loops that could erode bank capital and crowd out private lending.
- Enhance clear and timely communication by authorities to reduce market uncertainty and prevent abrupt funding runs.
- Establish emergency liquidity tools to stem systemic financial stress, including measures to address uneven deposit insurance coverage and rapid foreign funding outflows.
- Develop resolution regimes to reduce the buildup of zombie firms and limit the long-term drag on productivity and credit allocation.

*Source: IMF Regional Economic Outlook—Middle East and Central Asia, October 2023 (chapter 3).*

### 1. Excess Losses

### Excess Losses

### Excess losses by bank characteristics
- Losses are measured in excess of net income as a fraction of Tier 1 regulatory capital.
- Loss concentration:
  - Higher for banks with relatively illiquid balance sheets.
  - Higher for banks with low profitability (RoA defined as net income over total assets).
  - Higher for banks with low provisioning levels.
  - Higher for banks with higher leverage.
  - Larger losses concentrated in banks with higher within-country market shares.
- Banks in countries with greater duration of outstanding sovereign bonds face much higher sensitivity to interest rate increases.
- State-owned banks (SOB defined as banks with at least 50 percent government ownership) are separately assessed.

### Bank undercapitalization and impact on capital ratios
- Few banks would become undercapitalized in the combined scenario, but capital buffers would be significantly eroded, especially in countries that begin with relatively low capital ratios (Egypt, Morocco, Pakistan).
- Relative to Basel III minimum capital requirements:
  - "18 percent of banks in MENA EM&MIs and Pakistan would become undercapitalized in the combined scenario."
  - All banks in the GCC and the CCA would remain above minimum requirements in the combined scenario.
- Imposing country-specific minimum capital requirements increases the share of undercapitalized banks slightly in the GCC and MENA EM&MIs and Pakistan.
- Aggregate Tier 1 capital ratios would decline by:
  - "2.6 percentage points for privately owned banks in MENA EM&MIs and Pakistan."
  - "4.5 percentage points for state-owned banks in MENA EM&MIs and Pakistan."
  - In the GCC, ratios decline by "3.7 percentage points" for state-owned banks and "2.0 percentage points" for privately owned banks.
- Basel III minimum capital requirement as referenced: "4.5 percent common equity Tier 1 plus 1.5 percent additional Tier 1 capital and a 2.5 percent capital conservation buffer."
- Most countries included in the stress testing exercise require minimum Tier 1 capital higher than "8.5 percent."

### Impact on credit provision and output
- Historical relationship (local projection approach, Jordà 2005):
  - A "1 percentage point decline in capital ratios" has been historically correlated with a contraction in real credit, reaching "1.2 percent after eight quarters."
- Using the combined scenario estimates:
  - Real credit could contract by "4.3 percent in MENA EM&MIs and Pakistan" over a two-year horizon.
  - Real credit could contract by "3.2 percent in the GCC" over a two-year horizon.
- Macroeconomic model results on output:
  - The decline in bank lending in line with the combined scenario could translate into output losses of a magnitude similar to past credit downturns in MENA EM&MIs and Pakistan, and the GCC.
  - The median output loss from the decline in lending is estimated to be about "0.5 percent" in MENA EM&MIs and Pakistan, and in the GCC.
  - At the 95th percentile of estimated output losses across countries following the credit contraction in the combined scenario:
    - There could be a "1.5 percent output contraction in MENA EM&MIs and Pakistan" over two years.
    - There could be a "0.9 percent contraction in the GCC" over two years.
- Caveats:
  - The banking stress test does not explicitly model interbank linkages due to lack of data; estimates should be interpreted as lower bounds.
  - Amplification could occur through nonbank financial institutions, though expected to be limited because they have a low market share in most countries of the region.

### Macroprudential frameworks: current stance and gaps
- Use of macroprudential policies in ME&CA has generally been slow despite broad-based tools being available:
  - Almost all countries in the region have some form of broad-based tool available covering capital buffers (e.g., countercyclical capital buffer), but most have left the setting at "zero since inception."
  - Most countries have implemented some borrower-based tools for the household sector (for example, caps on debt-service-to-income ratios).
  - Tools to guard against vulnerabilities and elevated credit risk in the nonfinancial corporate sector have generally been less used.
  - Some GCC countries (for example, Saudi Arabia) have not used measures to reduce banks’ foreign currency liquidity risks.
  - Some MENA countries (Algeria, Lebanon, Morocco, Tunisia) have taken fewer actions to reduce risks from domestic systemically important financial institutions.
- Regional variation in macroprudential action may reflect different risk profiles but can leave key gaps in addressing vulnerabilities.

### Policies to safeguard financial stability (recommendations)
- Strengthen and develop macroprudential frameworks:
  - In MENA and Pakistan: ramp up the use of broad-based macroprudential tools such as the countercyclical capital buffer to prevent sharp credit contractions during downturns.
  - Countries with elevated corporate debt-at-risk or prevalent zombie firms (for example, Kuwait, Jordan, United Arab Emirates) should consider borrower-based tools such as caps on debt-service-to-income ratios and loan-to-value ratios.
  - Implement measures targeting large domestic systemically important institutions (e.g., increased capital surcharges), especially in MENA EM&MIs where implementation is lagging.
  - In the GCC: guard against unexpected liquidity stress related to foreign liabilities; consider tools that account for concentrated nonresident deposit bases in liquidity coverage and net stable funding ratios; consider enhanced macroprudential foreign exchange measures such as reserve requirements.
  - In the CCA: continue ongoing macroprudential measures to build resilience across credit cycles, incentivize de-dollarization, and reduce foreign exchange mismatches.
  - Monitor links between the banking sector and nonbank financial institutions where risks may be migrating.
- Address vulnerabilities from the sovereign-bank nexus:
  - Near term: preserve bank capital to absorb losses in countries with elevated interest rate risk (for example, Jordan); conduct bank stress tests considering multiple nexus channels; central banks should pay attention to bank asset classification, provisions, and exposures to interest rate and liquidity risks; where risks are elevated, restrict profit distribution plans as first-line defense.
  - Medium term: in countries with limited fiscal space, pursue macroeconomic policies that strengthen debt sustainability to contain government financing needs and limit deterioration of the sovereign-bank nexus.
  - Where bank holdings of sovereign bonds exceed concentration limits (for example, Egypt, Pakistan), consider gradual measures to lessen the nexus, including imposing capital surcharges on banks’ sovereign bond holdings above certain thresholds, phased in appropriately and conditional on appropriate macroeconomic policies.
  - Across all countries: foster a deep and diversified investor base, build adequate buffers at state-owned banks, provide clear mandates, and align supervisory tools such as stress tests with banks’ unique risk profiles.
- Communication and crisis preparedness:
  - Clear communication of central bank objectives and policy functions is crucial to avoid unnecessary uncertainty.
  - Some countries would benefit from improved communication on macroprudential frameworks (for example, by issuing a financial stability report).
  - If monetary policy is adjusted for financial stability purposes, clearly communicate the intent to return inflation to target as soon as possible once stress lessens.
  - Prepare emergency liquidity tools:
    - In countries where central banks lack explicit authority to provide emergency liquidity assistance (Algeria, Morocco, Oman), governments should prioritize establishing a clear framework for dealing with liquidity distress.
    - In countries where central bank laws allow emergency liquidity but lack operational directives (Egypt, Jordan in MENA; Georgia, Tajikistan in CCA), provide specific instructions and internal guidelines for the use of emergency liquidity facilities, including foreign exchange liquidity.
  - Design liquidity support to address liquidity, not solvency; interventions should have well-defined end dates, be parsimonious, and be appropriately priced to minimize moral hazard and avoid conflicting with monetary tightening.
  - Ensure liquidity support and crisis management tools comply with Islamic banking rules where applicable.
- Strengthen resolution frameworks to address solvency concerns:
  - Enhance insolvency procedures to address legacy nonperforming loans.
  - Structure resolution regimes to enable swift resolution of nonperforming loans to prevent evergreening, the buildup of impaired legacy assets, and the emergence of zombie firms that would weigh on aggregate productivity growth.

*Source: IMF staff calculations and analysis in "Excess Losses" (chapter excerpt).*

### References

### References

### Selected citations
- Acharya, Viral, Matteo Crosignani, Tim Eisert, and Sascha Steffen. 2022. “Zombie Lending: Theoretical, International, and Historical Perspectives.” Annual Review of Financial Economics 14 (1): 21–38.
- Adams, Mark, Hanife Yesim Aydin, Hee Kyong Chon, Anastasiia Morozova, and Ebru Sonbul Iskender. 2022. “Regulating, Supervising, and Handling Distress in Public Banks.” IMF Departmental Paper 22/010, International Monetary Fund, Washington, DC.
- Copestake, Alex, Divya Kirti, and Yang Liu. Forthcoming. “Banks’ Joint Exposure to Market and Run Risk.” IMF Working Paper, International Monetary Fund, Washington, DC.
- Drechsler, Itamar, Alexi Savov, and Philipp Schnabl. 2021. “Banking on Deposits: Maturity Transformation without Interest Rate Risk.” The Journal of Finance 76 (3): 1091–143.
- Damodaran, Aswath. 2023. “Ratings, Interest Coverage Ratios, and Default Spreads.” https://pages.stern.nyu.edu/~adamodar/
- International Monetary Fund (IMF). 2019. “Kuwait: Financial Sector Assessment Program—Financial System Stability Assessment.” IMF Country Report 19/96, Washington, DC.
- International Monetary Fund (IMF). 2021. “Georgia: Financial Sector Assessment Program—Financial System Stability Assessment.” IMF Country Report 21/216, Washington, DC.
- International Monetary Fund (IMF). 2023. “Jordan: Financial Sector Assessment Program—Financial System Stability Assessment.” IMF Country Report 23/140, Washington, DC.
- International Monetary Fund (IMF). Forthcoming. “Stress Testing in Times of Increasing Interest Rates and Beyond.” IMF Department Note, Washington, DC.
- Jiang, Erica Xuewei, Gregor Matvos, Tomasz Piskorski, and Amit Seru. 2023. “Monetary Tightening and US Bank Fragility in 2023: Mark-to-Market Losses and Uninsured Depositor Runs?” NBER Working Paper 31048, National Bureau of Economic Research, Cambridge, MA.
- Jordà, Òscar. 2005. “Estimation and Inference of Impulse Responses by Local Projections.” American Economic Review 95 (1): 161–82.
- Khandelwal, Padamja, Ezequiel Cabezon, Sanan Mirzayev, and Rayah Al-Farah. 2022. “Macroprudential Policies to Enhance Financial Stability in the Caucasus and Central Asia.” IMF Departmental Paper 2022/006, International Monetary Fund, Washington, DC.
- Teodoru, Iulia Ruxandra, and Klakow Akepanidtaworn. 2022. “Managing Financial Sector Risks from the COVID-19 Crisis in the Caucasus and Central Asia.” IMF Departmental Paper 22/005, International Monetary Fund, Washington, DC.

### Notes from references and data sources
- See Adams and others (2022) for more details on policy proposals related to state-owned banks.
- The regional and country recommendations are based partially on data from the IMF’s Monetary Operations and Instruments database.
- See Teodoru and Akepanidtaworn (2022) for a more detailed discussion of dollarization and exchange rate risks in the Caucasus and Central Asia.
- Because of limitations in the availability of recent banking data, the analysis covers Georgia and Kazakhstan.
- The magnitude of the increase (sovereign spreads) is consistent with the external shock scenario in the 2021 Financial Sector Assessment Program for Georgia (IMF 2021).

---

### Box 3.1. Banking Sector Stress Test for the Caucasus and Central Asia

### Context and added assumptions to the baseline stress test
- The Caucasus and Central Asia (CCA) faces financial stability challenges including relatively high levels of dollarization, associated exchange rate risks, and strong recent inflows from Russia.
- Additional assumptions added to the baseline banking sector stress test:
  - Inflows from Russia normalize, and profitability returns to its prepandemic average amid a normalization of net foreign currency gains.
  - An adverse external shock triggers a rise by 150 percent in sovereign spreads.
  - Exchange rates depreciate by 30 percent, stressing unhedged corporate and household borrowers.

### Key quantitative results (losses and capital impacts)
- Liquidity stress scenario:
  - Losses of between 4 and 7 percent of regulatory capital.
- Corporate stress with large currency depreciation:
  - Losses rise to 7.3 percent of regulatory capital.
- Combined liquidity and corporate stress:
  - Losses as large as 15.3 percent of regulatory capital.
- Undercapitalization and aggregate capital ratio effects:
  - 1.8 percent of banks (weighted by assets) would become undercapitalized.
  - Capital ratios would decline by 2.6 percentage points in aggregate, from 17.4 percent to 14.8 percent.
- Corporate stress scenario (aggregate Tier 1 impacts):
  - Aggregate Tier 1 capital ratio would drop by 1.2 percentage points.
- Profitability shock without currency depreciation:
  - Without currency depreciation, the decline would be negligible even if profitability returns to prepandemic averages (a 44 percent decline from current levels).

### Vulnerabilities and drivers
- Vulnerabilities related to dollarization and unhedged foreign exchange borrowers are key drivers of the losses.
- CCA banks are resilient against small shocks due to high profitability and cash buffers, but their main vulnerabilities stem from:
  - An adverse shock that triggers a surge in sovereign spreads.
  - Sizable currency depreciations.
- Foreign exchange liquidity is a potential vulnerability that is not part of the stress test because the bank-level data used do not include a breakdown of liquid assets between local currency and foreign currency liquid assets; thus it is not possible to evaluate the extent to which banks have sufficient foreign exchange liquidity, especially in US dollars.
- Teodoru and Akepanidtaworn (2022) highlight that the simultaneous realization of foreign exchange credit and liquidity risks would have compounding effects on the banking sectors in the Caucasus and Central Asia, with the largest and state-owned banks being most vulnerable.

### Scope and data
- Countries included in the stress test analysis: Georgia and Kazakhstan.
- Data sources: Fitch Connect; IMF, Financial Soundness Indicators database; and IMF staff calculations.
- Panel definitions:
  - Panel 1 reports estimated losses—in excess of banks’ net income—across five scenarios. Losses are scaled as a share of Tier 1 regulatory capital.
  - The corporate (no depreciation) scenario assumes a growth shock only to corporations. The full corporate scenario adds a further rise in nonperforming loans because of borrowers’ unhedged foreign exchange exposures.
  - Panel 2 reports current baseline Tier 1 capital ratio for Caucasus and Central Asia banks and counterfactual capital ratios across scenarios.
- Abbreviations: bps = basis points; RWA = risk-weighted assets.

*Prepared by Thomas Kroen; sources as listed in the chapter references.*

---


_Source: https://www.imf.org/-/media/files/publications/reo/mcd-cca/2023/october/english/ch3.pdf_