## Executive summary (wpiea2024026-print-pdf)

## Source details

**Canonical URL:** [Executive summary (wpiea2024026-print-pdf)](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024026-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024026-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024026-print-pdf.pdf.json)

---

### Liquidity risk in investment funds: context and motivation
- Market turmoil in March 2020 raised market volatility and margin calls dramatically; euro area investment funds sold securities worth almost EUR 300 billion in the first quarter of 2020.
- OTC derivatives reforms introduced daily exchange of variation and initial margins, reducing counterparty credit risk but increasing liquidity (funding) risk because counterparties must hold high-quality liquid collateral to meet margin calls.
- Variation margin emphasized because:
  - Variation margin collateralization of funds’ portfolios currently exceeds that by initial margin.
  - Variation margin tends to be more procyclical and volatile than initial margin.

### Data sources and sample
- Key data:
  - Transaction-by-transaction EMIR data (reference date December 20, 2018).
  - Refinitiv Lipper data on funds’ liquidity buffers (cash and security-by-security holdings).
  - ECB’s Centralised Securities Database (CSDB) and market data sources (ANNA DSB, Bloomberg, Datastream).
- Notional values and processing:
  - After cleaning/filtering: euro area funds’ derivatives notional ≈ EUR 17 trillion in December 2018; 31 percent centrally cleared.
  - EMIR data processing effects:
    - Total before cleaning: EUR 48,216 trillion notional; 20,190,406 rows.
    - After cleaning/filtering for euro area investment funds: EUR 17 trillion notional; 1,893,893 rows.
- Coverage and heterogeneity:
  - Out of almost 60,000 euro area investment funds, around 35 percent use derivatives.
  - Derivatives use by NAV:
    - Over 80 percent of funds with NAV over EUR 5 billion use derivatives.
    - Less than 5 percent of funds with NAV below EUR 1 million use derivatives.
  - By strategy:
    - Bond, hedge, and mixed funds: derivatives use over 45 percent.
    - Equity funds: 33 percent.
    - Real estate funds: 8 percent.

### Methodology overview
- Construct pricing functions for the 10 most prevalent derivative types across interest rate, currency, and equity derivatives.
- Calibrate pricing functions using EMIR transaction-level data and external market data to derive potential variation margin calls under market stress scenarios and netting assumptions.
- Compare simulated potential variation margin calls at individual fund level with actual liquidity buffers from Refinitiv Lipper to identify funds with liquidity shortfalls.
- Aggregate individual fund results to compute sector-level liquidity needs and potential for spillovers.

### Step 1 — Identify funds’ derivatives and liquidity positions (summary of implementation and liquidity buffer stats)
- Data elements from EMIR: LEIs of counterparties, collateral portfolio identifiers, attributes enabling contract repricing.
- Derivatives asset classes considered (covering 89 percent of notional): interest derivatives, equity derivatives, currency derivatives.
- Liquidity positions (Lipper Refinitiv) defined as:
  - Cash (deposits).
  - High-quality government debt securities (euro area sovereigns and AAA-rated sovereigns), following Basel HQLA-L1 definition.
- Key statistics (Table 3) for funds with asset holdings available in Lipper data:
  - Sample counts:
    - EA funds: 58,544
    - Funds with asset holdings in Lipper data: 6,634
  - For funds with asset holdings:
    - Average cash and equivalents: EUR 11.9 million
    - Average cash and HQLA-L1: EUR 71.6 million
    - 25th Percentile: cash 0.0; cash and HQLA-L1 1.4
    - Median: cash 1.0; cash and HQLA-L1 7.2
    - 75th percentile: cash 5.8; cash and HQLA-L1 37.8
    - Total: cash 78,859.0; cash and HQLA-L1 475,179.0
  - Of which also reporting derivatives in EMIR data:
    - Count: 4,305
    - Average cash and equivalents: EUR 15.6 million
    - Average cash and HQLA-L1: EUR 91.7 million
    - 25th Percentile: cash 0.0; cash and HQLA-L1 2.5
    - Median: cash 1.4; cash and HQLA-L1 12.3
    - 75th percentile: cash 8.3; cash and HQLA-L1 53.6
    - Total: cash 67,266.0; cash and HQLA-L1 394,760.0
  - After filtering for applicability of methodology:
    - Sample left with 3,500 funds, with NAV of around EUR 1.7 trillion.
    - EMIR-indicated holding of derivatives portfolios for around 14,000 funds (≈68 percent of funds using derivatives).
- Rationale:
  - Cash is emphasized because over three quarters of peak margin calls in Feb–Apr 2020 were paid from cash deposits; cash is fungible and accepted by CCPs for variation margin.
  - High-quality government bonds proxied for repo funding capacity (government bonds account for 85 percent of euro-denominated repo transactions).
  - MMFs considered insufficiently liquid as a funding source given strains and outflows in March 2020.

### Elasticities of variation margin to independent shocks (Table 7)
- Elasticities (variation margin call, EUR billion) for marginal changes in underlying variables:
  - Interest rates: 1bp parallel downward shift → 0.065
  - FX/currency: 1% USD depreciation vis-à-vis EUR → 0.750
  - Equities: 1% decline → 0.630
- Interpretation:
  - 1 basis point parallel downward shift in yield curves → margin call EUR 0.065 billion in the small sample (symmetrical for upward shift).
  - 1 percent USD depreciation vis-à-vis EUR → margin call EUR 0.75 billion in the small sample.
  - 1 percent decline in equities → margin call EUR 0.63 billion in the small sample.
- Caveat: Elasticities approximate small shocks; less reliable for larger shocks and combinations because of nonlinearities and netting.

### Sensitivity and scenario results (stylized simulations and rescaled scenarios)
- Stylized simple shocks (small sample, one-day and prolonged two-week):
  - Interest-rate-only shocks:
    - 25 bps parallel downward shift → margin call EUR 1.7 billion; shortfall EUR 0.5 billion.
    - 75 bps parallel downward shift → margin call EUR 5.3 billion; shortfall EUR 0.8 billion.
  - Adding FX and equity shocks substantially increases shortfalls; adding equity shocks to interest rate shocks in prolonged turmoil increased shortfalls from EUR 0.8 billion to EUR 9.4 billion.
  - Combined shocks across all three markets:
    - Benchmark one-day scenarios: largest shortfalls reach EUR 4 billion and EUR 4.5 billion.
    - Benchmark prolonged scenarios: largest shortfalls reach EUR 8.7 billion and EUR 9.4 billion.
  - Note: Equity-derivative shortfalls can be induced even when shocks apply only to interest rates due to repricing via discount curves.
- Scenario analysis rescaled to full sample (focus on combined shocks with interest rate declines):
  - Scenario 1a (one-day, interest rate declines):
    - Share of funds with derivatives exposures lacking sufficient cash buffers: 33 percent.
    - Share higher for bond funds: 35 percent; "other" funds: 40 percent.
    - Estimated cash shortfalls for sample of ~3,500 funds: EUR 4.5 billion.
    - Rescaled to full sample of 14,000 funds:
      - Simulated aggregate variation margin: EUR 46 billion.
      - Estimated overall cash shortfall: EUR 31 billion.
      - Composition of variation margin calls: equity derivatives 53 percent; interest rate derivatives 26 percent; currency derivatives 21 percent.
  - Scenario 2a (prolonged turmoil, interest rate declines):
    - Share of funds with insufficient liquidity buffers: 13 percent.
    - Equity funds particularly affected: 25 percent with insufficient buffer.
    - Estimated liquidity shortfall for ~3,500-fund sample: EUR 9.4 billion.
    - Rescaled to full sample:
      - Simulated aggregate variation margin: EUR 129 billion.
      - Estimated broader liquidity shortfall: around EUR 76 billion.
- Aggregate headline estimates:
  - Around EUR 30 billion additional liquidity needs for an extreme one-day market shock (aggregate variation margin simulated around EUR 45 billion).
  - Around EUR 70 billion additional liquidity needs under prolonged market turmoil (aggregate variation margin simulated around EUR 130 billion).
  - Up to EUR 140 billion in a scenario of prolonged market turmoil (aggregate margin calls), of which only half would be covered by funds’ (highly) liquid assets (cash and government bonds eligible for the Basel Liquidity Coverage Ratio).
- Fund-level vulnerability:
  - Between 13 percent and 33 percent of euro area funds with sizable derivatives exposures may not have sufficient liquidity buffers to meet variation margin calls under adverse market scenarios.
- Historical comparators:
  - Daily variation margin calls on funds during March 2020 likely reached tens of billions of euro; Rousová 2020b estimated fund margin calls to peak at around EUR 40 billion on March 16, 2020.
  - In March 2020, about 53 percent of variation margin calls faced by euro area investment funds originated from equity derivatives, even though equity derivatives account for around 11 percent of funds’ derivative portfolios by notional value.

### Robustness analysis: netting and liquidity buffer assumptions (Table 9)
- Benchmark one-day market move scenario (negative interest rate shock, small sample of 3,523 funds):
  - Netting = Partial; Liquidity buffer = cash only → Variation margin EUR 6.61 billion; Shortfall EUR 4.50 billion; Share of funds with shortfall 33 percent.
  - Netting = Full; Liquidity buffer = cash only → Variation margin EUR 6.27 billion; Shortfall EUR 4.41 billion; Share of funds with shortfall 28 percent.
  - Netting = Partial; Liquidity buffer = cash + HQLA-L1 → Variation margin EUR 6.61 billion; Shortfall EUR 2.93 billion; Share of funds with shortfall 16 percent.
  - Netting = Full; Liquidity buffer = cash + HQLA-L1 → Variation margin EUR 6.27 billion; Shortfall EUR 2.90 billion; Share of funds with shortfall 13 percent.
- Findings:
  - Netting assumption alters simulated variation margin by around 5 percent only.
  - Liquidity buffer assumption is substantial: including high-quality government bonds reduces estimated shortfall by around one-third and halves the number of funds with shortfalls (but share remains close to 15 percent).
  - Reason: Broader liquidity buffer is on average around six times larger than the narrow cash buffer for the sample.

### Heterogeneity, amplification channels, and historical examples
- Results vary significantly by shock type and fund category; some categories show particularly severe shortages.
- Funds with liquidity shortages would likely:
  - Redeem MMF shares,
  - Procyclically sell assets (fire sales),
  - Draw on credit lines,
  potentially amplifying market dynamics and causing spillovers.
- Historical amplification/spillover episodes:
  - Archegos failure (March 2021) and LDI funds’ heavy gilt sales (September 2022) highlighted vulnerabilities and market disruption risks, especially in high interest rate environments.

### Contributions relative to existing literature and limitations
- Novelty:
  - First comprehensive analysis (to the authors’ knowledge) of liquidity risk from margin calls faced by euro area investment funds using EMIR and Lipper Refinitiv granular datasets.
  - Simulations for 10 contract types across interest rate, currency, and equity derivatives.
- Limitations:
  - Focus on variation margin only (not initial margin).
  - Assumes static portfolios (no rebalancing after a shock).
  - Funds may not be able to quickly monetize valuation gains; some assets may be encumbered or less liquid.
  - Margin calls could combine with redemption requests and initial margin calls, aggravating liquidity needs.
  - Further research suggested (e.g., include sharp yield increase scenarios).

### Policy implications and recommendations
- Need for enhanced macroprudential tools to address liquidity (funding) risk in the fund sector, emphasizing prevention and containment of vulnerability buildup before risks materialize.
- Potential measures include:
  - Regulatory requirements to strengthen funds’ ability under stress to meet potential funding needs, including variation margin calls.
  - Aligning funds’ liquidity risk with liquidity of funds’ assets and redemption policies.
  - Limits on synthetic leverage to reduce funds’ exposure to liquidity risk arising from derivatives exposures.
- Expected outcomes:
  - Such tools would increase sector resilience to future financial turbulence and reduce the need for ex post interventions.
- Further use:
  - Methodology could inform system-wide and fund-level liquidity stress testing exercises that incorporate margin-call liquidity risk and inform potential regulatory requirements for robust liquidity stress tests at the individual fund level.

*Source: Executive summary and Sections 3.1 and 4.1 of wpiea2024026-print-pdf*

### Executive summary ......................................................................................................

### wpiea2024026-print-pdf - Executive summary

### Executive summary
- Entry point of the document, listed on page 5.

### 1 Introduction
- Section 1 titled "Introduction".

### 2 Data
- Section 2 titled "Data".

### 3 Methodology
- Section 3 titled "Methodology".
- Subsections (as listed):
  - 3.1 Step 1: Identify funds’ derivatives and liquidity positions
  - 3.2 Step 2: Define market shocks
  - 3.3 Step 3: Reprice derivatives
  - 3.4 Step 4: Estimate margin calls
  - 3.5 Step 5: Measure liquidity shortfall
  - 3.6 Step 6: Rescale liquidity shortfalls to the full sample

### 4 Results
- Section 4 titled "Results".
- Subsections (as listed):
  - 4.1 Sensitivity analysis: Simple and combined shocks
  - 4.2 Scenario analysis: Combined shocks rescaled to the full sample
  - 4.3 Robustness analysis: Netting and liquidity buffer assumptions

### 5 Conclusions
- Section 5 titled "Conclusions".

### References
- References section titled "References".

*Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024026-print-pdf.pdf*

### Executive summary

### wpiea2024026-print-pdf - Executive summary

### Liquidity risk in investment funds: context and motivation
- Market turmoil in March 2020 raised market volatility and margin calls dramatically; euro area investment funds sold securities worth almost EUR 300 billion in the first quarter of 2020.
- OTC derivatives reforms introduced the daily exchange of variation and initial margins for the vast majority of derivatives exposures, reducing counterparty credit risk but increasing liquidity (funding) risk because counterparties must hold (or have ready access to) high-quality liquid collateral to meet margin calls.
- Variation margin is the primary focus because:
  - Variation margin collateralization of funds’ portfolios currently exceeds that by initial margin.
  - Variation margin tends to be more procyclical and volatile than initial margin.

### Data sources and sample
- Key data:
  - Transaction-by-transaction EMIR data (trade repository data), cleaned and processed with reference date December 20, 2018.
  - Refinitiv Lipper data on individual funds’ liquidity buffers (cash and security-by-security holdings).
  - ECB’s Centralised Securities Database (CSDB) and market data sources (ANNA DSB, Bloomberg, Datastream) for underlying product characteristics.
- After cleaning and filtering, euro area investment funds’ derivatives notional value amounted to around EUR 17 trillion in December 2018, of which 31 percent was centrally cleared.
- EMIR data processing effects (selected steps):
  - Total before cleaning: EUR 48,216 trillion notional; 20,190,406 rows.
  - After dropping matured, terminated, inconsistent, outliers, currency adjustments and filtering for euro area investment funds: EUR 17 trillion notional; 1,893,893 rows.
- Coverage: out of almost 60,000 euro area investment funds, around 35 percent use derivatives.
  - Derivatives use increases with fund size: over 80 percent of funds with NAV over EUR 5 billion use derivatives; less than 5 percent of funds with NAV below EUR 1 million do so.
  - By strategy: bond, hedge, and mixed funds have derivatives use over 45 percent; equity funds 33 percent; real estate funds 8 percent.

### Methodology overview
- Construct pricing functions for the 10 most prevalent types of derivatives held by funds across three asset classes: interest rate, currency, and equity derivatives.
- Use EMIR transaction-level data and external market data to calibrate pricing functions and derive potential variation margin calls under a range of market stress scenarios and netting assumptions.
- Compare simulated potential variation margin calls at the individual fund level with actual liquidity buffers from Refinitiv Lipper to identify funds with liquidity shortfalls.
- Aggregate individual fund results to compute sector-level liquidity needs and potential for spillovers.

### Key quantitative findings and scenarios
- Estimated liquidity needs under stress:
  - Around EUR 30 billion for an extreme one-day market shock.
  - Around EUR 70 billion under a prolonged market turmoil.
  - Up to EUR 140 billion in a scenario of prolonged market turmoil (aggregate margin calls), of which only half would be covered by funds’ (highly) liquid assets (cash and government bonds eligible for the Basel Liquidity Coverage Ratio).
- Fund-level vulnerability:
  - Between 13 percent and 33 percent of euro area funds with sizable derivatives exposures may not have sufficient liquidity buffers to meet variation margin calls under adverse market scenarios.
- Historical comparators:
  - Daily variation margin calls on funds during March 2020 likely reached tens of billions of euro; Rousová 2020b estimated fund margin calls to peak at around EUR 40 billion on March 16, 2020.
- Portfolio composition and shock relevance:
  - Interest rate and currency derivatives dominate notional values, but equity derivatives can account for a large share of variation margin: in March 2020, about 53 percent of variation margin calls faced by euro area investment funds originated from equity derivatives, even though equity derivatives account for around 11 percent of funds’ derivative portfolios by notional value.

### Heterogeneity and amplification channels
- Results vary significantly by shock type and fund category; some fund categories show particularly severe liquidity shortages.
- Funds with liquidity shortages would likely:
  - Redeem money market fund (MMF) shares,
  - Procyclically sell assets (fire sales),
  - Draw on credit lines,
  potentially amplifying market dynamics and causing spillovers to other markets and sectors.
- Historical episodes illustrating amplification and spillovers:
  - Archegos failure (March 2021) and LDI funds’ heavy gilt sales (September 2022) exposed vulnerabilities and market disruption risks, especially in high interest rate environments.

### Contributions relative to existing literature and limitations
- Novelty:
  - First comprehensive analysis (to the authors’ knowledge) of liquidity risk from margin calls faced by euro area investment funds using EMIR and Lipper Refinitiv granular datasets.
  - Simulations for 10 contract types across interest rate, currency, and equity derivatives.
- Limitations:
  - Focus on variation margin only (not initial margin). Reasons: dominant collateralization by variation margin, greater procyclicality of variation margin, and complexity of modeling initial margin.

### Policy implications and recommendations
- The results call for enhanced macroprudential tools to address liquidity (funding) risk in the fund sector, emphasizing prevention and containment of vulnerability buildup before risks materialize.
- Potential measures include:
  - Regulatory requirements to strengthen funds’ ability under stress to meet potential funding needs, including variation margin calls.
  - Aligning funds’ liquidity risk with the liquidity of funds’ assets and redemption policies.
  - Limits on synthetic leverage to reduce funds’ exposure to liquidity risk arising from derivatives exposures.
- Expected outcomes:
  - Such tools would increase sector resilience to future financial turbulence and reduce the need for ex post interventions.

*Source: Executive summary of wpiea2024026-print-pdf*

### 3.1 Step 1: Identify funds’ derivatives and liquidity positions

### 3.1 Step 1: Identify funds’ derivatives and liquidity positions

### Data sources and scope
- Granular data used: EMIR and Refinitiv Lipper (presented in Section 2).
- From EMIR data retrieved:
  - Legal entity identifiers (LEIs) of funds’ counterparties.
  - Collateral portfolio identifiers.
  - Several other attributes enabling repricing of contracts in funds’ portfolios.
- Derivatives asset classes considered (covering 89 percent of the notional value of euro area funds’ derivatives portfolios):
  - Interest derivatives.
  - Equity derivatives.
  - Currency derivatives.
- Liquidity positions estimated using Lipper Refinitiv as funds’ holdings of:
  - Cash (deposits).
  - High-quality government debt securities, defined as:
    - Those issued by euro area sovereigns, and
    - AAA-rated debt securities issued by other sovereigns,
    - Following the Basel definition of level 1 high-quality liquid assets (HQLA-L1) developed for banks.6

### Key statistics on liquidity buffers (Table 3)
- Sample counts:
  - EA funds: 58,544
  - Of which having asset holdings available in Lipper data: 6,634
- For funds with asset holdings available in Lipper data:
  - Average cash and equivalents: EUR 11.9 million
  - Average cash and HQLA-L1: EUR 71.6 million
  - 25th Percentile: cash 0.0; cash and HQLA-L1 1.4
  - Median: cash 1.0; cash and HQLA-L1 7.2
  - 75th percentile: cash 5.8; cash and HQLA-L1 37.8
  - Total: cash 78,859.0; cash and HQLA-L1 475,179.0
- Of which also reporting derivatives in EMIR data:
  - Count: 4,305
  - Average cash and equivalents: EUR 15.6 million
  - Average cash and HQLA-L1: EUR 91.7 million
  - 25th Percentile: cash 0.0; cash and HQLA-L1 2.5
  - Median: cash 1.4; cash and HQLA-L1 12.3
  - 75th percentile: cash 8.3; cash and HQLA-L1 53.6
  - Total: cash 67,266.0; cash and HQLA-L1 394,760.0
- After filtering for applicability of methodology:
  - Sample left with 3,500 funds, with NAV of around EUR 1.7 trillion.
  - EMIR-indicated holding of a derivatives portfolio for around 14,000 funds (representing around 68 percent of funds using derivatives).

### Rationale for liquidity buffer composition
- Cash is emphasized because:
  - Figure 3 (BCBS, CPMI, IOSCO (2022)) indicates over three quarters of peak margin calls in February to April 2020 paid by “clients” came from available cash deposits.
  - Cash is fungible and can be transferred between counterparties quickly.
  - Cash is the only collateral for variation margin accepted by central clearing counterparties (CCPs).7
- High-quality government bonds used as proxy for repo funding capacity because:
  - Government bonds account for 85 percent of all transactions in the euro-denominated repo market (ECB 2021).8
  - Government bonds dominated collateral in euro repo markets and can be used for collateral transformation in prolonged stress.
  - Repo markets with high-quality collateral endured March 2020 market turmoil (ICMA 2020).
- Money market funds (MMFs) considered an insufficiently liquid funding source due to:
  - Significant outflows and strains that MMFs experienced during March 2020 (Boucinha et al. 2020).
  - Evidence that nonbanks’ need to pay variation margin contributed to strains faced by euro area MMFs (Ghio et al. 2022).

### Practical implementation notes for Step 1
- Collateral portfolio identifiers reported in EMIR enable grouping of contracts into portfolios (netting sets).
- Unique collateral portfolio code constructed by concatenating the LEIs of the two counterparties with the reported collateral portfolio codes; if missing, replaced by reported variation margin value.
- For liquidity buffer estimation:
  - Cash holdings from Lipper Refinitiv are used directly.
  - HQLA-L1 holdings identified as defined above and included where relevant.
- Rescaling and sample representativeness:
  - Because liquidity position information is available for only around 4,300 funds and the derivates-pricing-capable sample includes around 14,000 funds, rescaling is applied to extrapolate shortfalls from the small sample to the larger sample.
  - Rescaling assumption: the ratio of the liquidity shortfall to the size of the variation margin call is the same in the two samples.14

*Source: wpiea2024026-print-pdf - 3.1 Step 1: Identify funds’ derivatives and liquidity positions*

### 4.1 Sensitivity analysis: Simple and combined shocks

### 4.1 Sensitivity analysis: Simple and combined shocks

### Elasticities of variation margin to independent shocks
- Method: Elasticities calculated as changes in the value of the contracts held by funds in the sample for marginal changes in underlying variables (PV01 for interest rate products and Delta for equities).
- Table 7: Elasticity of margin calls to independent shocks (variation margin call, EUR billion)
  - Interest rates: 1bp parallel downward shift → 0.065
  - FX/currency: 1% USD depreciation vis-à-vis EUR → 0.750
  - Equities: 1% decline → 0.630
- Interpretation:
  - A 1 basis point parallel downward shift in yield curves would yield a margin call of EUR 0.065 billion in the small sample; a 1 basis point upward shift would be symmetrical (funds receive variation margin payments).
  - A 1 percent depreciation in the US dollar vis-à-vis the euro → margin call of EUR 0.75 billion in the small sample.
  - A 1 percent decline in all equity indexes/stocks underlying the derivatives → margin call of EUR 0.63 billion in the small sample.
- Caveat: Elasticities provide reasonable approximation for small shocks but are less reliable for larger shocks and shock combinations due to nonlinearities and portfolio netting.

### Results for simple and combined market shocks (stylized simulations)
- Approach: Simulations for combinations of stylized shocks (see Table 8) for one-day and prolonged (two weeks) durations.
- Interest-rate-only shocks:
  - Parallel downward shift of 25 bps → margin call EUR 1.7 billion; shortfall EUR 0.5 billion.
  - Parallel downward shift of 75 bps → margin call EUR 5.3 billion; shortfall EUR 0.8 billion.
  - Shortfalls larger in the prolonged market stress scenario despite broader liquidity buffer.
- Role of FX and equities:
  - Adding FX and equity shocks to interest rate shocks substantially increases liquidity shortfalls.
  - Adding equity shocks to interest rate shocks under the prolonged turmoil scenario increases shortfalls from EUR 0.8 billion to EUR 9.4 billion (more than 10-fold).
- Combined shocks across all three markets:
  - Benchmark one-day market move scenarios: largest shortfalls reach EUR 4 billion and EUR 4.5 billion.
  - Benchmark prolonged market turmoil scenarios: largest shortfalls reach EUR 8.7 billion and EUR 9.4 billion.
- Note: Some margin shortfalls from equity derivatives can be induced even when shocks apply only to interest rates due to repricing via discount curves.

### Scenario analysis: Combined shocks rescaled to the full sample
- Focus: One-day combined shocks with interest rate declines (scenario 1a) and prolonged market turmoil with interest rate declines (scenario 2a).
- Scenario 1a (one-day, interest rate declines):
  - Estimated share of funds with derivatives exposures lacking sufficient cash buffers: 33 percent.
  - Share higher for bond funds: 35 percent; "other" funds: 40 percent.
  - Estimated cash shortfalls for sample of ~3,500 funds: EUR 4.5 billion.
  - Rescaled to full sample of 14,000 funds:
    - Simulated aggregate variation margin: EUR 46 billion.
    - Estimated overall cash shortfall: EUR 31 billion.
    - Composition of variation margin calls: equity derivatives 53 percent, interest rate derivatives 26 percent, currency derivatives 21 percent.
- Scenario 2a (prolonged turmoil, interest rate declines):
  - Estimated share of funds with insufficient liquidity buffers: 13 percent.
  - Equity funds particularly affected: 25 percent with insufficient buffer.
  - Estimated liquidity shortfall for ~3,500-fund sample: EUR 9.4 billion.
  - Rescaled to full sample:
    - Simulated aggregate variation margin: EUR 129 billion.
    - Estimated broader liquidity shortfall: around EUR 76 billion.

### Robustness analysis: Netting and liquidity buffer assumptions
- Table 9: Benchmark one-day market move scenario with negative interest rate shock (scenario 1a), small sample of 3,523 euro-area funds (liquidity buffers available).
  - Netting = Partial; Liquidity buffer = cash only → Variation margin EUR 6.61 billion; Shortfall EUR 4.50 billion; Share of funds with shortfall 33 percent.
  - Netting = Full; Liquidity buffer = cash only → Variation margin EUR 6.27 billion; Shortfall EUR 4.41 billion; Share of funds with shortfall 28 percent.
  - Netting = Partial; Liquidity buffer = cash + HQLA-L1 → Variation margin EUR 6.61 billion; Shortfall EUR 2.93 billion; Share of funds with shortfall 16 percent.
  - Netting = Full; Liquidity buffer = cash + HQLA-L1 → Variation margin EUR 6.27 billion; Shortfall EUR 2.90 billion; Share of funds with shortfall 13 percent.
- Findings:
  - Netting assumption alters simulated variation margin by around 5 percent only (minor role).
  - Liquidity buffer assumption is substantial: using a broader liquidity buffer including high-quality government bonds reduces estimated shortfall by around one-third and halves the number of funds with shortfalls (but the share remains substantial, close to 15 percent).
  - Reason: Broader liquidity buffer is on average around six times larger than the narrow cash buffer for the sample (Table 3 reference).

### Conclusions and policy implications
- Main quantitative summary:
  - Additional liquidity needs estimated around EUR 30 billion for an extreme one-day market shock (aggregate variation margin simulated around EUR 45 billion).
  - Additional liquidity needs estimated around EUR 70 billion under prolonged market turmoil (aggregate variation margin simulated around EUR 130 billion).
- Contextual evidence:
  - Results comparable with evidence from prior episodes: mid-March 2020 daily variation margin calls on funds likely reached tens of billions of euro; Autumn 2022 UK pension funds raised GBP 40 bn in cash for LDI margin/collateral calls.
  - Large derivative exposures of euro area funds: around EUR 17 trillion of notional value.
- Limitations noted:
  - Assumes static portfolios (no rebalancing after a shock).
  - Funds may not be able to quickly monetize valuation gains due to illiquidity; actual cash/liquidity buffers may differ (some assets encumbered or less liquid).
  - Margin calls could combine with redemption requests and initial margin calls, aggravating liquidity needs.
  - Further research suggested to expand simulation tool (e.g., include sharp yield increase scenarios).
- Policy recommendations:
  - Need for enhanced prudential tools and a macroprudential approach to address liquidity risk in the fund sector to contain systemic implications.
  - Methodology could inform development of system-wide and fund-level liquidity stress testing exercises that incorporate margin-call liquidity risk.
  - Results could inform potential regulatory requirements for robust liquidity stress tests at the individual fund level and strengthen liquidity risk management frameworks.
  - Consideration of limits on synthetic leverage to reduce funds’ exposure to liquidity risk arising from derivatives.

*Source: Authors’ calculations, as presented in "4.1 Sensitivity analysis: Simple and combined shocks" from the provided IMF content unit.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024026-print-pdf.pdf_