## wp17226

## Source details

**Canonical URL:** [wp17226](https://www.imf.org/-/media/files/publications/wp/2017/wp17226.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2017/wp17226.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2017/wp17226.pdf.json)

---

### Introduction: growth and risks in the asset management industry
- Worldwide Fund Industry: Total Net Assets grew from 2006 to 2016 (Figure 1). Sources: ICI, IIFA, EFAMA.
- Holdings and concentration:
  - Significant footprint in U.S. corporate bonds (Figure 2). Source: Flow of Funds, IMF staff.
  - Growth in EM bond funds and HY bond funds (Figures 3 and 4). Source: EPFR.
- Liquidity mismatch risk:
  - Most investment funds (UCITS in the EU, mutual funds in the U.S.) offer daily liquidity while investing in less liquid asset classes (e.g., HY or EM debt).
  - Notable incidents: collapse of the U.S. HY Third Avenue Credit Focused Fund in December 2015; suspensions of redemptions among U.K. real estate funds after the June 2016 Brexit referendum.
- Motivation:
  - FSB recommendations urge liquidity stress testing and enhanced liquidity risk monitoring (FSB (2016), FSB (2017)).
  - Paper objective: provide a framework for liquidity stress testing usable by regulators and market participants; draws on IMF FSAPs for the U.S., Ireland, Sweden, Luxembourg.

### Liquidity stress testing for funds — overview and methods
- Objective:
  - Assess resilience of investment funds to severe, but plausible, redemption shocks at individual and industry levels.
- Core components:
  - Redemption shock calibration (historical distribution, adverse scenario, ad hoc, event study).
  - Measurement of liquidity buffers (Time to Liquidation, HQLA, cash and short-term debt).
  - Liquidation method modeling (waterfall, prorata).
  - Resilience metric: Redemption Coverage Ratio (RCR).
- Treatment of Liquidity Management Tools (LMTs):
  - LMTs (redemption fees, gates, borrowing facilities) exist but are excluded from stress tests due to uncertainty about use/impact.

### Calibration of the redemption shock
- Definition:
  - Redemption shock = net outflows in percent of Total Net Assets (TNA).
- Flow estimation when direct flows unavailable:
  - Flow_t = TNA_t − (1 + R_t) × TNA_{t−1}
- Main calibration approaches:
  - Historical approach: fund-specific 1st percentile of net flows (worst 1 percent net flows).
    - Used in FSAPs for the U.S., Sweden, Luxembourg.
    - Aggregation choice matters: fund-style aggregation typically yields lower shocks than individual-fund calibration.
  - Macroeconomic approach: econometric model linking net flows to macrofinancial variables from an adverse scenario; allows aggregation across funds.
  - Ad hoc thresholds and event studies are alternatives (examples: 5, 10, 20 percent).
- Case studies and sample statistics:
  - Top 50 HY and EM bond funds in the U.S. and Europe: sample of 200 funds covers around $850 billion in assets (Table 1).
  - Luxembourg FSAP sample: 191 funds covering €656 billion in assets.
  - Luxembourg historical-based redemption shocks (averages, percent of TNA):
    - EM: 18%
    - HY: 19%
    - Mixed funds: 9%
    - Other bond funds: 18%
    - Short Term CNAV: 19%
    - Short Term VNAV: 23%
    - Other MMFs: 18%
  - Luxembourg macroeconomic-projected redemption shocks (percent of TNA):
    - EM: 9%
    - HY: 11%
    - Mixed funds: -11%* (net inflows under adverse scenario)
    - Other bond funds: 6%
    - *Under the adverse scenario, mixed funds would experience net inflows.
- Macrofinancial adverse scenario inputs (selected initial values and projected changes):
  - 3M-Euribor: -0.33 / -0.13 / 0.20
  - EA 10-Year: 11.32 / 0.33
  - Term spread: 1.331 / 1.45 / 0.13
  - Eurostoxx 50: 100 / 94 / 6%
  - VIX: 14.82 / 23.44 / 58%
  - EM spreads (bps): 478 / 858 / 380
  - HY Spreads (bps): 498 / 1076 / 578
  - Sources: Thomson Reuters Datastream and IMF staff calculations.

### Measurement of liquidity buffers
- Two broad methods:
  - Time to Liquidation (TTL): security-level TTL or aggregated TTL using dealer inventories or turnover.
    - Security-level TTL requires models and trading data (outsourced providers: Bloomberg LQA, MSCI LiquidMetrics).
    - Aggregated TTL used in U.S. (selling pressure vs dealer inventories) and Sweden (selling pressure vs turnover).
  - Tiered approach: High-Quality Liquid Assets (HQLA) inspired by Basel III LCR.
    - Liquidity weights by asset class and rating (examples):
      - Cash and AAA–AA- sovereign/corporate: 100% / 85% / 85% depending on bucket.
      - A+ to A-: 85% / 50% / 50%
      - BB+ to BBB-: 50% / 50% / 0%
      - Below BBB-: 0% / 0% / 0% / 50% (equities bucket at 50%)
    - Liquidity index = sum over securities of (liquidity weight_k × share_k in percent of TNA).
- Cash and short-term debt securities approach:
  - Focus on cash and debt securities with residual maturity < 1 year.
  - Considerations:
    - Cash often operational; some cash serves operational needs (issuances, settlements, margin), not purely liquidity buffers.
    - Empirical evidence: managers may hoard cash prior to anticipated redemptions; less liquid funds hoard more.
- Case-study HQLA outcomes (sample of 200 U.S. and Europe funds, aggregated data):
  - HQLA averages (in percent of TNA):
    - U.S. EM Bond Funds: 12%
    - Europe EM Bond Funds: 6%
    - U.S. HY Bond Funds: 40%
    - Europe HY Bond Funds: 58%
  - Note: liquidity buffers higher for EM bond funds than HY funds by HQLA measure due to sovereign exposure composition.
- Luxembourg sample liquidity buffers (percent of TNA):
  - Cash and short-term debt / HQLA averages:
    - EM: 16% / 42%
    - HY: 12% / 12%
    - Mixed funds: 31% / 55%
    - Other bond funds: 20% / 56%
  - For MMFs, HQLA measure not used.

### Liquidation approaches (how funds would raise cash)
- Waterfall approach:
  - Managers sell most liquid assets first (IG sovereigns, cash) before less liquid securities.
  - Evidence mixed: some funds sold liquid assets first during GFC; others sold less liquid assets first.
- Prorata approach:
  - Managers sell across the portfolio in proportion to holdings to preserve portfolio structure and investment policy.
- Practice:
  - Funds may use a mix of waterfall and prorata in stress.
  - Recent FSAPs applied both approaches (U.S., Sweden, Luxembourg).
- Methods used in recent FSAPs (summaries):
  - U.S. (IMF (2015c)): Redemption shock 1% (historical distribution at fund style, monthly); TTL aggregated; prorata and waterfall; resilience measured via selling pressure vs dealer inventories.
  - Sweden (IMF (2016b)): 1% historical (fund style, quarterly); TTL aggregated; prorata and waterfall; resilience measured via selling pressure vs turnover.
  - Luxembourg (IMF (2017a)): 1% and model (individual funds and fund style, monthly); tiered approach (HQLA and short-term assets) using aggregated and security-level data; prorata and waterfall; resilience measured via Redemption Coverage Ratio (RCR).
  - Ireland (IMF (2016a)): Ad hoc thresholds (5/10/20%) at individual fund level, daily; TTL security-level; prorata; resilience measured via Time to Liquidation.

### Measuring resilience: Redemption Coverage Ratio (RCR) and liquidity shortfall
- RCR definition:
  - RCR = Liquid assets / Net outflows
  - If RCR > 1: fund has sufficient liquid assets to meet redemption shock.
  - If RCR < 1: fund must sell less liquid assets, risking discounts and potential market contagion.
- Liquidity shortfall (percent of TNA) for RCR < 1:
  - Liquidity shortfall = Net outflows − Liquid assets
- Case-study results (top 50 HY and EM bond funds U.S. and Europe, historical approach):
  - Average redemption shock (% TNA) and HQLA:
    - U.S. EM Bond Funds: 14% redemption shock; HQLA 12%
    - Europe EM Bond Funds: 15% redemption shock; HQLA 6%
    - U.S. HY Bond Funds: 9% redemption shock; HQLA 40%
    - Europe HY Bond Funds: 13% redemption shock; HQLA 58%
  - Many HY funds have liquidity shortfalls; some shortfalls exceed 10% of TNA.
- Luxembourg FSAP stress test outcomes:
  - Historical approach:
    - MMFs: enough liquid assets to cope with severe redemptions.
    - EM and HY funds: more difficulties (RCR below 1).
    - Under HQLA, median RCR for HY funds is 50%.
    - Some HY funds' liquidity shortfalls exceed the UCITS borrowing limit of 10% of TNA, implying potential use of LMTs, fire sales, or suspension of redemptions.
  - Historical approach summary (average redemption shock % TNA; cash and ST debt; HQLA):
    - Short Term CNAV: 19% / 0% ^
    - Short Term VNAV: 23% / 0% ^
    - Other MMFs: 18% / 0% ^
    - EM: 18% / 71% / 2%
    - HY: 19% / 78% / 75%
    - Mixed funds: 9% / 28% / 5%
    - Other bond funds: 18% / 52% / 8%
    - ^ The HQLA measure is not used for MMFs.
  - Forward-looking (macroeconomic) approach shows milder stress results due to generally lower projected redemption shocks.
- Other country case studies:
  - Ireland (security-level TTL, prorata): most sampled funds could liquidate within a limited number of days; EM funds faster to sell than HY funds.
  - U.S. (IMF (2015c)): selling pressure from mutual funds (MBS, Munis, Corporate bonds) under tail-event redemption shocks would be three to four times larger than dealer inventories.
  - Sweden (IMF (2016b)): domestic corporate bond-investing funds face greater exposure to redemption risk; selling pressures exceed market turnover.

### Bank–fund interlinkages
- Channels from funds to banks:
  - Direct: banks’ holdings of fund shares, funds’ deposits in banks, banks as counterparties to funds’ derivatives and securities financing, bank loans to funds.
  - Indirect: funds holding bank debt or equity; common exposures (e.g., sovereigns).
- Observations:
  - Banks’ asset-side exposures to funds are typically limited (low holdings of fund shares; limited loans).
  - Deposits from funds into banks can be sizeable, particularly for depositary banks in financial centers.
  - Derivatives exposures can be significant but data granularity is limited.
- Euro area exposures (as of June 2016):
  - Investment funds' assets to euro area banks (€ billion):
    - Cash: 345
    - Debt securities: 183
    - Equities: 56
    - Total assets: 584
  - Investment funds' liabilities to euro area banks (€ billion):
    - Loans: 64
    - Fund shares: 267
    - Total liabilities: 331
  - Notes: euro area bank debt securities held by funds account for around 4 percent of euro area bank debt outstanding; €56 billion of euro area bank equities held by funds is 14 percent of euro area bank equities.
- Luxembourg case (as of June 2016):
  - Exposures of Luxembourg funds to banks (billions of euros):
    - Deposits & loans: 175 (of which deposits to Luxembourg banks: 101)
    - Debt securities: 155 (of which to Luxembourg banks: 8)
    - Stocks: 2 (of which to Luxembourg banks: 0)
    - Derivatives: 126 (of which to Luxembourg banks: 1)
    - Investment funds' liabilities total: 330 (with limited exposures of banks holding fund shares).
  - Depositary banks: funds account for around 40 percent of depositary banks' liabilities in 2015.
- Deposit withdrawal stress testing and results (Luxembourg forward-looking approach):
  - Assumptions: funds may withdraw deposits to meet redemptions; identity of banks where funds park cash often assumed to be depositary bank.
  - Deposit outflows estimates (including mixed funds vs without mixed funds; amounts in billions of euros):
    - Total deposits: 24.7 (including mixed funds) / 21.6 (without mixed funds)
    - Deposit outflows (waterfall): 3.0 / 3.3
    - Deposit outflows (waterfall, in % of fund deposits): 12% / 15%
    - Deposit outflows (prorata): 6.8 / 7.2
    - Deposit outflows (prorata, in % of fund deposits): 28% / 33%
    - Deposit outflows (average of waterfall and prorata): 4.9 / 5.3
    - Deposit outflows (average of waterfall and prorata in % of funds deposits): 20% / 24%
  - Interpretation: prorata produces larger deposit outflows because cash is used proportionally; waterfall can yield lower deposit outflows when HQLA alone covers redemptions.

### Conclusion and policy implications
- The paper provides a flexible framework for liquidity stress testing of investment funds adaptable to data availability and methodological assumptions.
- Key needs and future work:
  - Integrate fund-driven fire sales into system-wide stress testing to estimate spillovers on markets and other agents.
  - Model joint liquidity and market shocks, account for market absorbing capacity, and include fire-sale dynamics.
  - Address data gaps, particularly granular bank–fund linkage data, to enable more robust analysis.
  - Explore deeper integration of bank–fund interlinkages for system-wide stress testing as recommended by FSB (2017).

### Appendix I. Estimating HQLA Measures Using Aggregated Data — Procedure and Findings
- Procedure:
  - Securities grouped by buckets depending on the asset type (equity, debt, cash).
  - Debt securities grouped by issuer type (sovereign, corporates) and ratings.
  - Aggregated information on asset type, issuer type and ratings typically available from commercial data providers but the split is typically done by issuer types and ratings but not both at the same time.
  - Proxy assumption: most investment grade issuers are sovereigns and the non-sovereign part is then attributed to corporates; this proxy tends to overestimate fund liquidity since liquidity weights on IG sovereign bonds are higher than on corporates.
  - Given that the table used for liquidity weights requires more granular data than available, averages are used.
- Example portfolio and HQLA computations (all shares in percent of TNA unless otherwise stated):
  - Table A.1 example: Sovereign 40% / Corporates 55% / Cash 5% / Total 100%
  - Table A.2 example by rating: Investment grade 60% — 60%x95%=57% ; High Yield 40% — 40%*95%=38% ; Total 100% — 95%
  - Table A.3 liquidity weights (averages used for aggregate data):
    - Sovereign AAA to AA-: Initial weights 100% — Weights used for aggregate data 78%
    - Corporate AAA to AA-: Initial weights 85% — Weights used for aggregate data 62%
    - Sovereign A+ to A-: Initial weights 85% — Weights used for aggregate data 50%
    - Corporate BBB+ to BBB-: Initial weights 50% — Weights used for aggregate data 50%
    - Below BBB-: Initial weights 0% — Weights used for aggregate data 0%
    - Cash: 100%
  - Table A.4 computation (proxy assuming most IG issuers are sovereigns):
    - Sovereign IG — 40% — 78% — 31%
    - Sovereign HY — 0% — 0% — 0%
    - Corporate IG — 17% — 62% — 10.5%
    - Corporate HY — 38% — 0% — 0%
    - Cash — 5% — 100% — 5%
    - Total — 100% —  — 46.5%
  - Table A.5 alternative computation (proxy assuming most IG issuers are corporates):
    - Sovereign IG — 2% — 78% — 1.5%
    - Sovereign HY — 38% — 0% — 0%
    - Corporate IG — 55% — 62% — 34.1%
    - Corporate HY — 0% — 0% — 0%
    - Cash — 5% — 100% — 5%
    - Total — 100% —  — 40.6%
- Aggregate vs. Security-Level HQLA (empirical assessment):
  - Sample: 132 funds domiciled in Luxembourg, included in the liquidity stress test for the FSAP.
  - Findings:
    - The aggregate measure performs relatively well, as most observations are close to the black line (which assumes that both measures are equal).
    - The average difference between the two measures is 1.1 percentage point.
    - A simple regression of HQLA (using aggregated data) on security-level HQLA gives a R² of 64 percent.
    - Both measures are very close for HY funds; Chart A.2 regression shows y = 1.04x and R² = 0.93.

*Source: wp17226 - REFERENCES (IMF staff paper).*

### References .............................................................................................................

### wp17226 - References

### Introduction: growth and risks in the asset management industry
- Worldwide Fund Industry: Total Net Assets grew from 2006 to 2016 (Figure 1). Sources: ICI, IIFA, EFAMA.
- Holdings and concentration:
  - Significant footprint in U.S. corporate bonds (Figure 2). Source: Flow of Funds, IMF staff.
  - Growth in EM bond funds and HY bond funds (Figures 3 and 4). Source: EPFR.
- Liquidity mismatch risk:
  - Most investment funds (UCITS in the EU, mutual funds in the U.S.) offer daily liquidity while investing in less liquid asset classes (e.g., HY or EM debt).
  - Notable incidents: collapse of the U.S. HY Third Avenue Credit Focused Fund in December 2015; suspensions of redemptions among U.K. real estate funds after the June 2016 Brexit referendum.
- Motivation:
  - FSB recommendations urge liquidity stress testing and enhanced liquidity risk monitoring (FSB (2016), FSB (2017)).
  - Paper objective: provide a framework for liquidity stress testing usable by regulators and market participants; draws on IMF FSAPs for the U.S., Ireland, Sweden, Luxembourg.

### Liquidity stress testing for funds — overview and methods
- Objective:
  - Assess resilience of investment funds to severe, but plausible, redemption shocks at individual and industry levels.
- Core components (Figure 5):
  - Redemption shock calibration (historical distribution, adverse scenario, ad hoc, event study).
  - Measurement of liquidity buffers (Time to Liquidation, HQLA, cash and short-term debt).
  - Liquidation method modeling (waterfall, prorata).
  - Resilience metric: Redemption Coverage Ratio (RCR).
- Treatment of Liquidity Management Tools (LMTs):
  - LMTs (redemption fees, gates, borrowing facilities) exist but are excluded from stress tests due to uncertainty about use/impact.

### Calibration of the redemption shock
- Definition:
  - Redemption shock = net outflows in percent of Total Net Assets (TNA).
- Flow estimation when direct flows unavailable:
  - Flow_t = TNA_t − (1 + R_t) × TNA_{t−1}
- Main calibration approaches:
  - Historical approach: fund-specific 1st percentile of net flows (worst 1 percent net flows).
    - Used in FSAPs for the U.S., Sweden, Luxembourg.
    - Aggregation choice matters: fund-style aggregation typically yields lower shocks than individual-fund calibration.
  - Macroeconomic approach: econometric model linking net flows to macrofinancial variables from an adverse scenario; allows aggregation across funds.
  - Ad hoc thresholds and event studies are alternatives (examples: 5, 10, 20 percent).
- Case studies and sample statistics:
  - Top 50 HY and EM bond funds in the U.S. and Europe: sample of 200 funds covers around $850 billion in assets (Table 1).
  - Luxembourg FSAP sample: 191 funds covering €656 billion in assets.
  - Luxembourg historical-based redemption shocks (averages, percent of TNA):
    - EM: 18%
    - HY: 19%
    - Mixed funds: 9%
    - Other bond funds: 18%
    - Short Term CNAV: 19%
    - Short Term VNAV: 23%
    - Other MMFs: 18%
  - Luxembourg macroeconomic-projected redemption shocks (percent of TNA):
    - EM: 9%
    - HY: 11%
    - Mixed funds: -11%* (net inflows under adverse scenario)
    - Other bond funds: 6%
    - *Under the adverse scenario, mixed funds would experience net inflows.

- Macrofinancial adverse scenario inputs (selected initial values and projected changes):
  - Initial value / Projection / Change:
    - 3M-Euribor: -0.33 / -0.13 / 0.20
    - EA 10-Year: 11.32 / 0.33
    - Term spread: 1.331 / 1.45 / 0.13
    - Eurostoxx 50: 100 / 94 / 6%
    - VIX: 14.82 / 23.44 / 58%
    - EM spreads (bps): 478 / 858 / 380
    - HY Spreads (bps): 498 / 1076 / 578
  - Sources: Thomson Reuters Datastream and IMF staff calculations.

### Measurement of liquidity buffers
- Two broad methods:
  - Time to Liquidation (TTL): security-level TTL or aggregated TTL using dealer inventories or turnover.
    - Security-level TTL requires models and trading data (outsourced providers: Bloomberg LQA, MSCI LiquidMetrics).
    - Aggregated TTL used in U.S. (selling pressure vs dealer inventories) and Sweden (selling pressure vs turnover).
  - Tiered approach: High-Quality Liquid Assets (HQLA) inspired by Basel III LCR.
    - Liquidity weights by asset class and rating (examples from Table 6):
      - Cash and AAA–AA- sovereign/corporate: 100% / 85% / 85% depending on bucket.
      - A+ to A-: 85% / 50% / 50%
      - BB+ to BBB-: 50% / 50% / 0%
      - Below BBB-: 0% / 0% / 0% / 50% (equities bucket at 50%)
    - Liquidity index = sum over securities of (liquidity weight_k × share_k in percent of TNA).
- Cash and short-term debt securities approach:
  - Focus on cash and debt securities with residual maturity < 1 year.
  - Considerations:
    - Cash often operational; some cash serves operational needs (issuances, settlements, margin), not purely liquidity buffers.
    - Empirical evidence: managers may hoard cash prior to anticipated redemptions; less liquid funds hoard more.
- Case-study HQLA outcomes (sample of 200 U.S. and Europe funds, aggregated data):
  - HQLA averages (in percent of TNA):
    - U.S. EM Bond Funds: 12%
    - Europe EM Bond Funds: 6%
    - U.S. HY Bond Funds: 40%
    - Europe HY Bond Funds: 58%
  - Note: liquidity buffers higher for EM bond funds than HY funds by HQLA measure due to sovereign exposure composition.
- Luxembourg sample liquidity buffers (percent of TNA):
  - Cash and short-term debt / HQLA averages:
    - EM: 16% / 42%
    - HY: 12% / 12%
    - Mixed funds: 31% / 55%
    - Other bond funds: 20% / 56%
  - For MMFs, HQLA measure not used.

### Liquidation approaches (how funds would raise cash)
- Waterfall approach:
  - Managers sell most liquid assets first (IG sovereigns, cash) before less liquid securities.
  - Evidence mixed: some funds sold liquid assets first during GFC; others sold less liquid assets first.
- Prorata approach:
  - Managers sell across the portfolio in proportion to holdings to preserve portfolio structure and investment policy.
- Practice:
  - Funds may use a mix of waterfall and prorata in stress.
  - Recent FSAPs applied both approaches (U.S., Sweden, Luxembourg).
- Methods used in recent FSAPs (Table 9 summary):
  - U.S. (IMF (2015c)): Redemption shock 1% (historical distribution at fund style, monthly); TTL aggregated; prorata and waterfall; resilience measured via selling pressure vs dealer inventories.
  - Sweden (IMF (2016b)): 1% historical (fund style, quarterly); TTL aggregated; prorata and waterfall; resilience measured via selling pressure vs turnover.
  - Luxembourg (IMF (2017a)): 1% and model (individual funds and fund style, monthly); tiered approach (HQLA and short-term assets) using aggregated and security-level data; prorata and waterfall; resilience measured via Redemption Coverage Ratio (RCR).
  - Ireland (IMF (2016a)): Ad hoc thresholds (5/10/20%) at individual fund level, daily; TTL security-level; prorata; resilience measured via Time to Liquidation.

### Measuring resilience: Redemption Coverage Ratio (RCR) and liquidity shortfall
- RCR definition:
  - RCR = Liquid assets / Net outflows
  - If RCR > 1: fund has sufficient liquid assets to meet redemption shock.
  - If RCR < 1: fund must sell less liquid assets, risking discounts and potential market contagion.
- Liquidity shortfall (percent of TNA) for RCR < 1:
  - Liquidity shortfall = Net outflows − Liquid assets
- Case-study results (top 50 HY and EM bond funds U.S. and Europe, historical approach):
  - Average redemption shock (% TNA) and percent of funds unable to cover (HQLA):
    - U.S. EM Bond Funds: 14% redemption shock; HQLA 12% (percentage of funds unable to cover: shown in Table 10 graphics).
    - Europe EM Bond Funds: 15% redemption shock; HQLA 6%.
    - U.S. HY Bond Funds: 9% redemption shock; HQLA 40%.
    - Europe HY Bond Funds: 13% redemption shock; HQLA 58%.
  - Many HY funds have liquidity shortfalls; some shortfalls exceed 10% of TNA.
- Luxembourg FSAP stress test outcomes:
  - Historical approach (average redemption shock and median RCR implications):
    - MMFs: enough liquid assets to cope with severe redemptions.
    - EM and HY funds: more difficulties (RCR below 1).
    - Under HQLA, median RCR for HY funds is 50%.
    - Some HY funds' liquidity shortfalls exceed the UCITS borrowing limit of 10% of TNA, implying potential use of LMTs, fire sales, or suspension of redemptions.
  - Historical approach summary (average redemption shock % TNA; cash and ST debt; HQLA):
    - Short Term CNAV: 19% / 0% ^
    - Short Term VNAV: 23% / 0% ^
    - Other MMFs: 18% / 0% ^
    - EM: 18% / 71% / 2%
    - HY: 19% / 78% / 75%
    - Mixed funds: 9% / 28% / 5%
    - Other bond funds: 18% / 52% / 8%
    - ^ The HQLA measure is not used for MMFs.
  - Forward-looking (macroeconomic) approach shows milder stress results due to generally lower projected redemption shocks.
- Other country case studies:
  - Ireland (security-level TTL, prorata): most sampled funds could liquidate within a limited number of days; EM funds faster to sell than HY funds.
  - U.S. (IMF (2015c)): selling pressure from mutual funds (MBS, Munis, Corporate bonds) under tail-event redemption shocks would be three to four times larger than dealer inventories.
  - Sweden (IMF (2016b)): domestic corporate bond-investing funds face greater exposure to redemption risk; selling pressures exceed market turnover.

### Bank–fund interlinkages
- Channels from funds to banks:
  - Direct: banks’ holdings of fund shares, funds’ deposits in banks, banks as counterparties to funds’ derivatives and securities financing, bank loans to funds.
  - Indirect: funds holding bank debt or equity; common exposures (e.g., sovereigns).
- Observations:
  - Banks’ asset-side exposures to funds are typically limited (low holdings of fund shares; limited loans).
  - Deposits from funds into banks can be sizeable, particularly for depositary banks in financial centers.
  - Derivatives exposures can be significant but data granularity is limited.
- Euro area exposures (as of June 2016, Table 13):
  - Investment funds' assets to euro area banks (€ billion):
    - Cash: 345
    - Debt securities: 183
    - Equities: 56
    - Total assets: 584
  - Investment funds' liabilities to euro area banks (€ billion):
    - Loans: 64
    - Fund shares: 267
    - Total liabilities: 331
  - Notes: euro area bank debt securities held by funds account for around 4 percent of euro area bank debt outstanding; €56 billion of euro area bank equities held by funds is 14 percent of euro area bank equities.
- Luxembourg case (as of June 2016, Table 14 and Figure 12):
  - Exposures of Luxembourg funds to banks (billions of euros):
    - Deposits & loans: 175 (of which deposits to Luxembourg banks: 101)
    - Debt securities: 155 (of which to Luxembourg banks: 8)
    - Stocks: 2 (of which to Luxembourg banks: 0)
    - Derivatives: 126 (of which to Luxembourg banks: 1)
    - Investment funds' liabilities total: 330 (with limited exposures of banks holding fund shares).
  - Depositary banks: funds account for around 40 percent of depositary banks' liabilities in 2015 (Figure 12).
- Deposit withdrawal stress testing and results (Luxembourg forward-looking approach):
  - Assumptions: funds may withdraw deposits to meet redemptions; identity of banks where funds park cash often assumed to be depositary bank.
  - Deposit outflows estimates (including mixed funds vs without mixed funds; amounts in billions of euros):
    - Total deposits: 24.7 (including mixed funds) / 21.6 (without mixed funds)
    - Deposit outflows (waterfall): 3.0 / 3.3
    - Deposit outflows (waterfall, in % of fund deposits): 12% / 15%
    - Deposit outflows (prorata): 6.8 / 7.2
    - Deposit outflows (prorata, in % of fund deposits): 28% / 33%
    - Deposit outflows (average of waterfall and prorata): 4.9 / 5.3
    - Deposit outflows (average of waterfall and prorata in % of funds deposits): 20% / 24%
  - Interpretation: prorata produces larger deposit outflows because cash is used proportionally; waterfall can yield lower deposit outflows when HQLA alone covers redemptions.

### Conclusion and policy implications
- Paper provides a flexible framework for liquidity stress testing of investment funds adaptable to data availability and methodological assumptions.
- Key needs and future work:
  - Integrate fund-driven fire sales into system-wide stress testing to estimate spillovers on markets and other agents.
  - Model joint liquidity and market shocks, account for market absorbing capacity, and include fire-sale dynamics.
  - Address data gaps, particularly granular bank–fund linkage data, to enable more robust analysis.
  - Explore deeper integration of bank–fund interlinkages for system-wide stress testing as recommended by FSB (2017).

*Source: wp17226 - References (IMF staff paper).*

### REFERENCES

### REFERENCES

### References list
- Association of the Luxembourg Fund Industry, 2013, “Guidelines for UCITS Liquidity Risk Management”, March.  
- Autorité des Marchés Financiers, 2016, “Public Consultation: Guide to the use of stress tests as part of risk management within asset management companies”, August. 
- Cetorelli, N., F. Duarte and T. Eisenbach, 2016, “Are Asset Managers Vulnerable to Fire Sales?", Liberty Street Economics, Federal Reserve Bank of New York, February 18. 
- Coval, J. and E. Stafford, 2007, “Asset fire sales (and purchases) in equity markets”, Journal of Financial Economics, Vol. 86, 479–512. 
- European Central Bank, 2015, “Systemic risk, contagion and financial networks”, Financial Stability Review, November 2015. 
- European Securities and Markets Authority, 2015, “Measuring the Shadow banking system - a focused approach”, Trends, Risks and Vulnerabilities, No.2, 2015. 
- European Systemic Risk Board, 2016, “Risk Dashboard”, Issue 15, March 2016, Frankfurt. 
- Financial Stability Board, 2016, “Proposed Policy Recommendations to Address Structural Vulnerabilities from Asset Management Activities”, Consultative Document, 22 June 2016, Basel. 
- –––––, 2017, “Policy Recommendations to Address Structural Vulnerabilities from Asset Management Activities”, 12 January 2017, Basel. 
- Greenwood, R., A. Landier, and D. Thesmar, 2015, “Vulnerable Banks”, Journal of Financial Economics, Vol. 115 (3), 417–485. 
- Hau, H. and S. Lai, 2016, “The Role of Equity Funds in the Financial Crisis Propagation”, Review of Finance, forthcoming. 
- International Monetary Fund, 2014a, “Making the Transition from Liquidity to Growth-Driven Markets”, Global Financial Stability Report, Chapter 1, April 2014, Washington D.C. 
- –––––, 2014b, “Improving the Balance Between Financial and Economic Risk Taking”, Global Financial Stability Report, Chapter 1, October 2014, Washington D.C. 
- –––––, 2015a, “Enhancing Policy Traction and Reducing Risks”, Global Financial Stability Report, Chapter 1, April 2015, Washington D.C. 
- –––––, 2015b, “The Asset Management Industry and Financial Stability”, Global Financial Stability Report, Chapter 3, April 2015, Washington D.C. 
- –––––, 2015c, “Stress Testing —Technical Note”, United States Financial Sector Assessment Program, IMF Country Report No. 15/173, July 2015, Washington D.C. 
- –––––, 2016a, “Technical Note —Asset Management and Financial Stability”, Ireland Financial Sector Assessment Program, IMF Country Report No. 16/312, September 2016, Washington D.C. 
- –––––, 2016b, “Sweden: Financial System Stability Assessment”, Sweden Financial Sector Assessment Program, IMF Country Report No. 16/355, November 2016, Washington D.C. 
- –––––, 2017a, “Luxembourg: Financial System Stability Assessment”, Luxembourg Financial Sector Assessment Program, IMF Country Report No. 17/122, May 2017, Washington D.C. 
- –––––, 2017b, “Technical Note —Risk Analysis”, Luxembourg Financial Sector Assessment Program, IMF Country Report No. 17/261, August 2017, Washington D.C. 
- Malik, S. and P. Lindner, 2017, “On Swing Pricing and Systemic Risk Mitigation”, IMF Working Paper, forthcoming, Washington D.C. 
- Manconi, A., M. Massa and A. Asuda, 2012, “The role of institutional investors in propagating the crisis of 2007–2008”, Journal of Financial Economics, Vol. 104(3), 491–518, June. 
- Morris, S., I. Shim and H. Shin, 2017, “Redemption risk and cash hoarding by asset managers”, BIS Working Paper No. 608, January 2017. 
- Office of Financial Research, 2013, “Asset Management and Financial Stability”, September, Washington D.C. 
- Scholes, M., 2000, “Crisis and Risk Management”, American Economic Review, Vol. 90(2), 17–21.  

### Appendix I. Estimating HQLA Measures Using Aggregated Data — Procedure and Findings
- Procedure:
  - Securities are grouped by buckets depending on the asset type (equity, debt, cash).
  - Debt securities are grouped by issuer type (sovereign, corporates) and ratings.
  - Aggregated information on asset type, issuer type and ratings are typically available from commercial data providers but the split is typically done by issuer types and ratings but not both at the same time.
  - As a proxy, it is assumed that most investment grade issuers are sovereigns and the non-sovereign part is then attributed to corporates. This proxy tends to overestimate the liquidity of the fund since liquidity weights on IG sovereign bonds are higher than on corporates.
  - Given that the table used for liquidity weights requires more granular data than available, averages are used.

- Table A.1: Example of Portfolio Composition by Issuer Type
  - Portfolio by issuers — Share (in percent of TNA)
    - Sovereign 40%
    - Corporates 55%
    - Cash 5%
    - Total 100%

- Table A.2: Example of Portfolio Composition by Rating
  - Portfolio by rating — Share (in percent of securities) — Share in percent of TNA
    - Investment grade 60% — 60%x95%=57%
    - High Yield 40% — 40%*95%=38%
    - Total 100% — 95%

- Table A.3: Liquidity Weights (averages used for aggregate data)
  - Portfolio by issuer / Initial weights / Weights used for aggregate data
    - Sovereign AAA to AA- — Initial weights 100% — Weights used for aggregate data 78%
    - Corporate AAA to AA- — Initial weights 85% — Weights used for aggregate data 62%
    - Sovereign A+ to A- — Initial weights 85% — Weights used for aggregate data 50%
    - Corporate BBB+ to BBB- — Initial weights 50% — Weights used for aggregate data 50%
    - Below BBB- — Initial weights 0% — Weights used for aggregate data 0%
    - Cash — 100%

- Table A.4: Computation of HQLA indicator (proxy assuming most IG issuers are sovereigns)
  - Portfolio by issuer — Share (in percent of TNA) [A] — Liquidity weight [B] — Liquidity index [A]x[B]
    - Sovereign IG — 40% — 78% — 31%
    - Sovereign HY — 0% — 0% — 0%
    - Corporate IG — 17% — 62% — 10.5%
    - Corporate HY — 38% — 0% — 0%
    - Cash — 5% — 100% — 5%
    - Total 100% —  — 46.5%

- Table A.5: Alternative Computation of HQLA Indicator (proxy assuming most IG issuers are corporates)
  - Portfolio by issuer — Share (in percent of TNA) [A] — Liquidity weight [B] — Liquidity index [A]x[B]
    - Sovereign IG — 2% — 78% — 1.5%
    - Sovereign HY — 38% — 0% — 0%
    - Corporate IG — 55% — 62% — 34.1%
    - Corporate HY — 0% — 0% — 0%
    - Cash — 5% — 100% — 5%
    - Total 100% —  — 40.6%

- Aggregate vs. Security-Level HQLA (empirical assessment)
  - Sample: 132 funds domiciled in Luxembourg, included in the liquidity stress test for the FSAP.
  - Findings:
    - The aggregate measure performs relatively well, as most observations are close to the black line (which assumes that both measures are equal).
    - The average difference between the two measures is 1.1 percentage point.
    - A simple regression of HQLA (using aggregated data) on security-level HQLA gives a R² of 64 percent.
    - Both measures are very close for HY funds; Chart A.2 regression shows y = 1.04x and R² = 0.93.

*Source: wp17226 - REFERENCES (IMF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17226.pdf_