## CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS

## Source details

**Canonical URL:** [CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS](https://www.imf.org/-/media/files/publications/gfsr/2022/october/english/ch3.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/gfsr/2022/october/english/ch3.pdf.md)
- [Structured JSON version](/-/media/files/publications/gfsr/2022/october/english/ch3.pdf.json)

---

### Chapter at a glance — scale, trends, and current risk context
- Total net assets of open-end investment funds (OEFs) have quadrupled since 2008, reaching $41 trillion in the first quarter of 2022 and accounting for approximately one-fifth of the assets of the nonbank financial sector.
- Most OEFs are domiciled in advanced economies and invest in equities issued in advanced economies, while the share investing in relatively less liquid assets (such as corporate bonds or emerging market bonds and equities) has been rising rapidly.
- Recent stress signals and flows:
  - The March 2020 market turmoil saw historic outflows from relatively less liquid OEFs and a “dash for cash,” prompting major central bank interventions including purchases of corporate bonds and ETFs.
  - Emerging market economies experienced large and abrupt outflows of about $78 billion at the onset of the pandemic in March 2020, followed by sustained inflows.
  - Since the beginning of 2022, outflows from emerging market equity and bond funds totaled $69 billion.
- The resilience of the OEF sector is being tested as central banks normalize policy amid persistent inflationary pressures and tightening financial conditions observed since the beginning of 2022.

### Sample, measurement, and conceptual framework
- Sample and period:
  - 17,000 open-end investment funds domiciled in 43 countries holding more than 450,000 bond and equity securities.
  - Sample period: fourth quarter of 2013 to second quarter of 2022.
- Key measurement constructs:
  - Fund-level illiquidity measured as the value-weighted average of bid-ask spreads of securities held by the fund.
  - Asset-level vulnerability: weighted-average liquidity of funds holding the asset (weights = fund share of asset ownership).
  - Stress dummy: Chicago Board Options Exchange Volatility Index (VIX) above its 90th sample percentile.
- Liquidity-mismatch mechanism and run risk:
  - OEFs offering daily redemptions while holding illiquid assets create an asset-liability “liquidity” mismatch and first-mover advantage that can trigger runs.
  - Redemption-driven liquidation loop: redemptions → forced asset sales → adverse price impact → further redemptions, potentially spreading stress across markets and countries.
  - Central bank interventions can restore liquidity but may create moral hazard and underpricing of risk.

### Empirical observations: fragility, amplification, and stress episodes
- Asset-level vulnerability and volatility:
  - A one standard deviation increase in the vulnerability measure of an average bond increases its return volatility by 23 percent relative to the median return volatility of the bond.
  - Comparable study: Jiang and others (2022) find a one standard deviation increase in vulnerability of US corporate bonds is associated with a 16 percent higher return volatility.
  - A one standard deviation increase in the vulnerability measure is associated with about a 20 percent increase in bond return volatility (relative to median volatility) when VIX or monetary policy uncertainty is high (75th percentile) relative to when they are low (25th percentile).
  - For emerging market corporate bonds held by funds domiciled in advanced economies, a one standard deviation increase in vulnerability is associated with a 23 percent increase in their return volatility relative to their median volatility.
  - A one standard deviation increase in the fund-level vulnerability measure has a 3 percent to 5 percent larger effect on return volatility (relative to the median) for securities exposed to sell-herding compared with those that are not exposed.
- Stress episodes and selling pressure:
  - March 2020: fixed-income securities held by more illiquid funds experienced sharper price declines than those held by more liquid funds; bond return volatility and liquidity deteriorated dramatically.
  - The estimated coefficient for high-yield corporate bonds’ selling pressure is equal to 46 percent.
  - A significant decline in fund liquidity comparable to that observed during the March 2020 market turmoil can increase bond return volatility by more than 20 percent.
  - Bonds are generally held by more illiquid funds and hence are on average more vulnerable than equities; corporate high-yield and emerging market bonds are particularly vulnerable.
- Pecking order and liquidation behavior:
  - Funds tend to follow a pecking order of liquidation (horizontal slicing): sell relatively more liquid assets within portfolios first during stress.
  - Less liquid funds tend to face larger outflows, particularly when the VIX Index is high.
- Cross-border spillovers:
  - Vulnerabilities from advanced-economy funds spill over to emerging market asset prices, with stronger effects for emerging market economies.
  - Increases in asset-level vulnerabilities for less liquid assets (bonds) are associated with a significant tightening of domestic financial conditions in the following period; no similar effect is evident for equity securities.
  - Increased holdings of domestic assets by nonresident advanced-economy illiquid funds are associated with significant tightening in recipient countries’ domestic financial conditions in the period that follows.

### Liquidity buffers, cash holdings, and behavior under stress
- Observed cash-buffer patterns (percent of fund’s net assets):
  - Equity funds: cash buffers range from 0.5 percent to 4 percent.
  - Bond funds: cash buffers range from 1 percent to 9 percent.
  - Funds holding relatively illiquid securities on average hold larger cash buffers.
- Role and limits of cash buffers:
  - Liquidity buffers can provide flexibility to time asset sales but do not eliminate the first-mover advantage.
  - In normal times, funds facing outflows deplete cash buffers to pay investors; in periods of severe market stress (VIX above 90th percentile), funds appear to preserve portfolio liquidity and rely less on cash buffers.

### Swing pricing, antidilution levies, and other liquidity management tools
- Swing pricing mechanism and evidence:
  - Swing pricing adjusts the fund price via a "swing factor" so that transaction costs associated with redemptions are borne by redeeming investors; antidilution levies serve a similar purpose.
  - Adoption of swing pricing reduces the adverse impact of fund vulnerabilities on the volatility of bond returns by about one-third (percent of median volatility).
  - Panel regressions classify funds domiciled in Luxembourg or the United Kingdom as swing pricing funds.
- Calibration challenges and limits:
  - Optimal swing-factor estimates in the literature range from 0 to 9 percent, with the higher end applying in periods of stress and for funds whose investors react strongly to poor performance.
  - Many funds are constrained by maximum swing-factor caps set in prospectuses, typically substantially below 9 percent; caps often omit indirect costs such as price impact.
  - In extreme stress when market liquidity is very poor, swing factors or antidilution levies may be very large or difficult to calibrate; redemption suspensions or gates may be an alternative.
- Availability and use of other tools:
  - Tools that limit redemption ability—redemption suspensions, redemption fees, redemption gates, in-kind redemptions—are widely available across jurisdictions but are typically used only in extreme stress and carry stigma concerns.
  - Mandatory minimum liquidity buffer requirements are among the least-used tools across jurisdictions.
  - Calibration of liquidity buffers, redemption terms, and other tools is essential to limit first-mover advantages and reduce systemic spillovers.

### Exchange-traded funds (ETFs): differences, mispricing, and vulnerabilities
- Structural differences vs OEFs:
  - ETFs trade continuously in secondary markets, do not guarantee end-of-day NAV redemption, and do not suffer the same first-mover advantage that gives rise to run risk in OEFs.
  - ETF prices are tied to NAV through an arbitrage mechanism executed by authorized participants.
- Empirical evidence on ETF behavior during stress:
  - During March 2020 stress: discounts on bond ETFs increased dramatically, reaching more than 5 percent across all bond ETFs; discounts reached up to 27 percent for high-yield bond ETFs and up to 13 percent for investment-grade bond ETFs.
  - Bonds held by ETFs experience a smaller increase in volatility during periods of stress than comparable bonds held by OEFs.
  - ETF mispricing (difference between ETF closing price and fund NAV divided by fund NAV) can be interpreted as a market-implied swing factor that an OEF with a similar portfolio would require.
- ETF vulnerabilities and potential amplification:
  - Intraday liquidity provision and arbitrage can attract liquidity traders and increase nonfundamental volatility.
  - When market liquidity deteriorates and broker-dealer balance sheets are constrained, the gap between NAV and ETF share price can increase (ETF mispricing).

### Policy implications, recommendations, and supervisory priorities
- General orientation:
  - Ex ante measures that reduce investor run risk are generally preferable to ex post measures that limit redemptions once runs have started.
  - Policy design should aim to lower the probability of central bank interventions by strengthening fund resilience ex ante.
- Recommended actions for fund and market resilience:
  - Encourage more widespread adoption and appropriate calibration of price-based liquidity management tools (swing pricing, antidilution levies) so transaction costs are passed to redeeming investors and first-mover incentives are reduced.
  - Consider requiring funds to eliminate prospectus caps on swing factors and calibrate swing factors to fully reflect the price impact of fund asset sales.
  - Improve transparency and data:
    - Encourage disclosure of swing pricing practices and calibration methodologies.
    - Improve availability of aggregate fund flow data in real time to assist appropriate swing-factor setting during stress.
    - Collect additional data on funds’ liquidity risks and leverage, including synthetic leverage via derivatives.
  - Strengthen supervisory monitoring of liquidity risk management practices and consider mandating liquidity management tools and enhanced disclosure where voluntary adoption is insufficient due to competitive pressure or stigma.
  - Consider linking frequency of redemptions to portfolio liquidity for funds holding very illiquid assets or when price-based tools are operationally infeasible; offer early redemption in exchange for a calibration-based redemption fee where appropriate.
  - Bolster market liquidity and resilience by encouraging central clearing and greater transparency in bond trading.
  - Recipient countries facing volatile capital flows from international funds should emphasize continued deepening of domestic markets; use appropriate debt management tools; deploy macroeconomic, prudential, capital flow management, and foreign exchange intervention tools in line with the IMF’s Institutional View.
  - Given cross-border spillovers and potential for recurring central bank interventions, consider broader regulation of investment funds if liquidity management practices remain inadequate, and pursue international regulatory coordination to ensure consistent deployment of liquidity management practices.

*International Monetary Fund | Chapter 3 authors: Andrea Deghi, Zhi Ken Gan, Pierre Guérin, Anna-Theresa Helmke, Tara Iyer, Junghwan Mok, Xinyi Su, and Felix Suntheim (lead); guidance by Fabio Natalucci, Mahvash Qureshi, and Mario Catalán; Itay Goldstein served as expert advisor.*

### Chapter 3 at a Glance

### Chapter 3 at a Glance

### Rapid growth and systemic relevance of open-end investment funds (OEFs)
- Total net assets of open-end investment funds have quadrupled since 2008, reaching $41 trillion in the first quarter of 2022 and accounting for approximately one-fifth of the assets of the nonbank financial sector.
- Most OEFs are domiciled in advanced economies and invest in equities issued in advanced economies, but the share investing in relatively less liquid assets (such as corporate bonds or emerging market bonds and equities) has been rising rapidly.
- OEFs are an important component of nonbank financial intermediation and reflect a shift in financial intermediation from banks to nonbank financial institutions.

### Mechanisms of asset price fragility and amplification
- OEFs that offer daily redemptions while holding illiquid assets create a liquidity mismatch between liabilities and asset holdings; this raises the likelihood of investor runs and asset fire sales.
- Runs and forced selling can generate downward pressure on asset prices, inducing additional redemptions and amplifying initial shocks through a vicious cycle between investor runs and asset market volatility.
- Herding behavior by funds can further intensify price pressures by causing investors to mimic others’ trading behavior.
- Leverage is noted as another potential exacerbating factor (analysis outside scope due to data limitations).

### Empirical observations and stress episodes
- The March 2020 market turmoil illustrated OEF vulnerabilities: OEFs invested in relatively less liquid assets experienced historic outflows and a “dash for cash,” contributing to market dislocations and liquidity problems that major central banks addressed with large policy measures, including purchases of corporate bonds and ETFs.
- A significant decline in fund liquidity comparable to that observed during the March 2020 market turmoil can increase bond return volatility by more than 20 percent.
- Investments by advanced economy OEFs in emerging markets have grown significantly; a comparable decline in liquidity of advanced economy bond funds can increase the return volatility of emerging market corporate bonds by more than 20 percent.
- Emerging market economies experienced large and abrupt outflows of about $78 billion at the onset of the pandemic in March 2020, followed by sustained inflows. Since the beginning of 2022, outflows from emerging market equity and bond funds totaled $69 billion.

### Current risk context and transmission channels
- The resilience of the OEF sector may be tested as central banks normalize policy amid persistent inflationary pressures and tightening financial conditions observed since the beginning of 2022.
- Recent months have seen large outflows from OEFs, especially from high-yield corporate bond funds and emerging market equity and bond funds, in tandem with monetary policy tightening by major central banks.
- A disorderly tightening of global financial conditions could trigger significant redemptions from OEFs holding illiquid assets, amplify stress in asset markets, and transmit adverse cross-border spillovers to recipient economies via volatile capital flows.

### Policy recommendations and liquidity-management tools
- Policymakers should ensure adequate liquidity management tools are used by OEFs; effective implementation of these tools is currently lacking.
- Price-based tools that aim to limit run incentives and reduce vulnerability, such as swing pricing and antidilution levies, can be potentially effective in mitigating asset price fragilities associated with less liquid OEFs.
  - Swing pricing is routinely used in some jurisdictions; to strengthen effectiveness, policymakers should provide guidance on implementation, ensure that swing factors fully reflect the price impact of trades, and encourage disclosure of swing pricing practices and calibration methodologies.
- Additional tools include limiting the frequency of redemptions by linking redemption frequency to the liquidity of funds’ portfolios to address the liquidity mismatch directly.
- Other liability-side tools include in-kind redemptions, redemption suspensions or gates, side pockets, redemption fees, and antidilution levies; price-based measures ensure trading costs are borne by exiting investors and do not restrict daily liquidity per se.
- Tighter monitoring of funds’ liquidity risk management practices by supervisors and regulators should be considered.
- Recipient economies should adopt policy responses to mitigate systemic risks from volatile capital flows sourced from OEFs, including continued deepening of domestic markets; the use of macroeconomic, prudential, and capital flow management measures; and foreign exchange intervention in line with the IMF’s Institutional View.

*International Monetary Fund | Chapter 3 authors: Andrea Deghi, Zhi Ken Gan, Pierre Guérin, Anna-Theresa Helmke, Tara Iyer, Junghwan Mok, Xinyi Su, and Felix Suntheim (lead); guidance by Fabio Natalucci, Mahvash Qureshi, and Mario Catalán; Itay Goldstein served as expert advisor.*

### 2. Cumulative Cross-Border Equity and Bond Fund Flows into EMs

### 2. Cumulative Cross-Border Equity and Bond Fund Flows into EMs

### Overview and scope
- Sample: 17,000 open-end investment funds (OEFs) domiciled in 43 countries holding more than 450,000 bond and equity securities.
- Sample period: fourth quarter of 2013 to second quarter of 2022.
- Focus: quantify OEF vulnerabilities (mainly portfolio illiquidity), assess contribution to asset price fragility (volatility of equity and bond returns), examine cross-border spillovers to emerging market (EM) asset prices, and evaluate liquidity risk management tools and policy options.
- Key observation: investment funds lack access to central bank liquidity facilities and deposit insurance and are subject to less intensive prudential oversight than banks, creating potential systemic vulnerabilities during severe market stress.

### Conceptual framework: liquidity mismatch and run risk
- OEFs offering daily redemptions while holding illiquid assets create an asset-liability “liquidity” mismatch that can trigger runs on funds.
- First-mover advantage: investors can redeem at current net asset value without internalizing full transaction costs; remaining investors bear those costs.
- Mechanism of amplification:
  - Redemptions (outflows) → forced asset sales if cash/cash-like buffers are insufficient.
  - Asset sales → adverse price impact, particularly for less liquid assets.
  - Lower asset prices → reduced fund performance and further redemptions, potentially spreading to other financial and nonfinancial entities and tightening financial conditions.
- Central bank interventions to restore liquidity (for example, purchases of risky assets) can be warranted but may create moral hazard and underpricing of risk.

### Stylized facts on fund and asset vulnerabilities
- Illiquidity metric: fund-level illiquidity measured as the value-weighted average of bid-ask spreads of securities held by the fund.
- Cross-asset patterns:
  - Bond funds are generally much more illiquid than equity funds.
  - Among bond funds, those holding corporate high-yield bonds and emerging market bonds are the most illiquid; sovereign bond funds are the most liquid.
- Time variation:
  - Fund-portfolio liquidity deteriorated dramatically in March 2020 and worsened again in the first half of 2022.
  - For emerging market bond funds, portfolio illiquidity in 2022 reached levels similar to March 2020.
- Asset-level vulnerability measure: weighted-average liquidity of funds holding the asset (weights = fund share of asset ownership).
  - Bonds are generally held by more illiquid funds and hence are on average more vulnerable than equities.
  - Corporate high-yield and emerging market bonds are particularly vulnerable.
  - Vulnerability of these assets rose dramatically during the COVID-19 crisis and increased again in 2022, in some cases near March 2020 levels.
- Price performance during stress:
  - March 2020: fixed-income securities held by more illiquid funds experienced sharper price declines (lower returns) than those held by more liquid funds.
  - 2022: a similar pattern of underperformance for vulnerable bonds occurred in the first half of 2022 amid monetary policy tightening and the war in Ukraine.
  - For equities, no meaningful difference in returns between assets held by more versus less vulnerable funds, consistent with greater liquidity in equity markets.

### Empirical findings on fragility and amplification
- Contribution to volatility:
  - After controlling for bond characteristics (including liquidity, rating, and maturity), illiquidity of OEFs contributes to higher volatility of bond returns.
  - A one standard deviation increase in the vulnerability measure of an average bond increases its return volatility by 23 percent relative to the median return volatility of the bond.
  - Comparable result: Jiang and others (2022) find a one standard deviation increase in vulnerability of US corporate bonds is associated with a 16 percent higher return volatility.
- Sensitivity to market stress:
  - Two stress measures considered: (1) VIX Index (financial market uncertainty) and (2) US monetary policy uncertainty (textual newspaper-based measure).
  - The adverse impact of asset-level vulnerability on bond return volatility is more pronounced when financial or monetary policy uncertainty is elevated.
  - Quantitatively: a one standard deviation increase in the vulnerability measure is associated with about a 20 percent increase in bond return volatility (relative to median volatility) when VIX or monetary policy uncertainty is high (75th percentile) relative to when they are low (25th percentile).
- Spillovers and pecking order:
  - In periods of high macro-financial uncertainty, return volatility of more liquid assets such as sovereign bonds also increases, consistent with funds following a “pecking order” when liquidating assets and potentially selling even relatively liquid instruments under stress.

### Policy implications and mitigation options
- Need for a policy and regulatory framework that:
  - Addresses vulnerabilities associated with OEFs stemming from liquidity mismatches and illiquid asset holdings.
  - Mitigates potential risks to financial stability while minimizing reliance on central bank intervention.
- Liquidity risk management and tools:
  - Fund-level liquidity risk management tools (discussed further in the chapter) can reduce vulnerabilities and the likelihood of asset fire sales.
  - Appropriate calibration of liquidity buffers, redemption terms, and other tools is essential to limit first-mover advantages and reduce systemic spillovers.
- Trade-offs:
  - While central bank asset purchases can stanch liquidity spirals during severe stress, such interventions may create moral hazard and systematic underpricing of risk; policy design should aim to lower the probability of interventions by strengthening fund resilience ex ante.

*International Monetary Fund. Global Financial Stability Report: Navigating the High-Inflation Environment, Chapter 3 excerpt.*

### CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS

### CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS

### Fund vulnerabilities amplify asset return volatility
- Funds with liquidity mismatches increase fragility in asset markets, especially fixed-income markets.  
- A one standard deviation increase in the fund-level vulnerability measure has a 3 percent to 5 percent larger effect on return volatility (relative to the median) for securities exposed to sell-herding compared with those that are not exposed.  
- For emerging market corporate bonds held by funds domiciled in advanced economies, a one standard deviation increase in the vulnerability measure is associated with a 23 percent increase in their return volatility relative to their median volatility.  
- The impact of vulnerability on bond return volatility is magnified during periods of high macro-financial uncertainty: a one standard deviation increase in vulnerability is associated with a 14 percent higher impact on bond return volatility in high-VIX periods compared with low-VIX periods.

### Herding, market stress, and cross-border amplification
- Herding (tendency of funds to trade in the same direction) amplifies the adverse effect of fund vulnerabilities on asset-price volatility.  
- The fragility generated by fund vulnerabilities is larger for securities with higher levels of sell-herding.  
- Vulnerabilities from funds domiciled in advanced economies spill over to emerging market asset prices, especially corporate bond markets.  
- Spillovers from advanced-economy fund vulnerabilities to emerging-market financial conditions are present for the full sample of countries but are much stronger for emerging market economies.

### Transmission channels from OEFs to price fragility
- Redemption-driven liquidation loop: investor redemptions force funds to liquidate portfolios, creating selling pressure that reduces security prices and prompts further redemptions.  
- Empirical findings confirming this mechanism:
  - Less liquid funds tend to face larger outflows, particularly when the VIX Index is high.  
  - Bonds with higher vulnerability (held by less liquid funds) are more likely to be liquidated when funds experience large outflows, with effects particularly pronounced for high-yield bonds.  
  - Funds tend to follow a pecking order of liquidation (horizontal slicing): during market stress (e.g., COVID-19 turmoil), funds sell relatively more liquid assets within their portfolios first.  
  - Selling-pressure estimates show substantial impacts on bond prices during stress—the estimated coefficient for high-yield corporate bonds’ selling pressure is equal to 46 percent.  
  - Event-study evidence: interacting a 10 percent fund outflow with a fund-security liquidation rank affects the likelihood of liquidation; liquidation-adjusted outflows materially affect abnormal returns during the COVID-19 market turmoil.

### Macro-financial spillovers and financial conditions
- Average domestic financial conditions are correlated with average asset-level vulnerability: financial conditions tighten with increases in asset holdings by less liquid OEFs.  
- Country-level panel regressions show that increases in asset-level vulnerabilities for less liquid assets (bonds) are associated with a significant tightening of domestic financial conditions in the following period; no similar effect is evident for equity securities.  
- The tightening effect of fund vulnerabilities on financial conditions is amplified when financial conditions are already tight.  
- Increased holdings of domestic assets by nonresident advanced-economy illiquid funds are associated with significant tightening in recipient countries’ domestic financial conditions in the period that follows.

### Liquidity management tools: availability and role
- Liquidity management tools can potentially reduce OEF vulnerabilities and mitigate amplification of asset-price fragility.  
- Tools that limit investors’ ability to redeem during severe outflows—redemption suspensions, redemption fees, redemption gates, in-kind redemptions—are the most widely available across jurisdictions but are typically used only in extreme stress and carry stigma concerns.  
- Antidilution levies and swing pricing can reduce vulnerabilities ex ante by passing transaction and liquidation costs to exiting investors, lowering run incentives; however, they are available in a limited number of jurisdictions and their utilization remains limited.  
- Mandatory minimum liquidity buffer requirements are among the least-used tools across jurisdictions.  
- Swing pricing is commonly used in Europe but has not been implemented by funds in the United States despite regulatory approval in 2018; a key impediment in the United States is that funds may not know net flow size before the price is determined, precluding application of a net-flow–based swing factor.

*International Monetary Fund | October 2022 — CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS*

### CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS

### CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS

### Liquidity buffers and cash holdings
- Liquidity buffers could provide funds with additional flexibility to time asset sales when facing outflows, but they:
  - Do not eliminate the first-mover advantage.
  - Can adversely impact long-term fund performance by constraining funds’ capacity to provide exposure to particular investment themes or asset classes.
- Observed cash-buffer patterns:
  - Cash and cash equivalents, 2013:Q4–2022:Q1: median and interquartile ranges reported across fund types.
  - Cash buffers vary widely within and across funds:
    - Equity funds: range from 0.5 percent to 4 percent (percent of fund’s net assets).
    - Bond funds: range from 1 percent to 9 percent (percent of fund’s net assets).
  - Funds holding relatively illiquid securities—as measured by their bid-ask spread—on average hold larger cash buffers (percent of total net assets).
- Behavior under normal vs stressed conditions:
  - Regression analysis uses a stress dummy equal to 1 when the Chicago Board Options Exchange Volatility Index is above its 90th sample percentile.
  - In normal times, funds facing outflows deplete cash buffers to pay investors.
  - In periods of severe market stress, funds appear to preserve portfolio liquidity and rely less on cash buffers to manage redemptions.

### Swing pricing: mechanism and empirical effectiveness
- Mechanism:
  - Swing pricing is an ex ante, price-based tool that imposes transaction costs associated with redemptions on redeeming investors by adjusting the fund price via a "swing factor".
  - Antidilution levies can have a similar effect by imposing a fee on redeeming investors.
- Empirical findings:
  - Adoption of swing pricing reduces the adverse impact of fund vulnerabilities on the volatility of bond returns by about one-third (percent of median volatility).
  - The mitigating effect is not sufficient to fully offset the increase in return volatility induced by illiquid funds’ bond holdings.
  - Panel regressions define swing pricing exposure by classifying funds domiciled in Luxembourg or the United Kingdom as swing pricing funds.
- Calibration challenges and limits:
  - Optimal swing-factor estimates in the literature range from 0 to 9 percent, with the higher end applying in periods of stress and for funds whose investors react strongly to poor performance.
  - Many funds are constrained by maximum swing-factor caps set in prospectuses, typically substantially below 9 percent.
  - Caps are often set based on direct trading costs (commissions and bid-ask spreads) and may not account for indirect costs such as price impact of asset sales.
  - In periods of extreme stress when market liquidity is very poor, swing factors or antidilution levies may be very large or difficult to calibrate; redemption suspensions or gates may be an alternative.

### Asset price fragility linked to OEFs and cross-border spillovers
- Core findings:
  - OEFs holding illiquid assets that offer daily redemptions are a key driver of asset price fragility.
  - Less liquid markets, such as corporate bonds, are most affected: return volatility increases significantly—especially in times of market stress—when these assets are held by more illiquid funds.
  - Fund vulnerabilities can have significant cross-border spillover effects and increase asset price volatility in emerging market economies.
  - Systemic implications include tightening domestic financial conditions and reinforcing the cycle of redemptions, fund asset sales, and price impact.
- Data and measurement notes:
  - Cash and cash equivalents include cash held in bank accounts, certificates of deposit, currency, money market holdings, and other high-quality fixed-income securities with a maturity of less than 92 days.
  - Panel 3 regression decomposes net inflows (positive net fund flows), net outflows (negative net fund flows made positive in the regression), interaction terms with stress dummy, and reports statistical significance at 10 percent or lower.

### Exchange-traded funds (ETFs) comparison
- Structural differences:
  - ETFs trade continuously in secondary markets and do not guarantee redemption at end-of-day net asset value; investors bear their own transaction costs.
  - ETFs are not subject to the same first-mover advantage that gives rise to run risk in OEFs.
- Empirical observations:
  - Bonds held by ETFs experience a smaller increase in volatility during periods of stress than comparable bonds held by OEFs.
  - ETF prices are tied to NAV through an arbitrage mechanism executed by authorized participants; when market liquidity deteriorates and broker-dealer balance sheets are constrained, the gap between NAV and ETF share price can increase (ETF mispricing, percent NAV).

### Policy conclusions and recommendations
- General conclusion:
  - The share of global financial assets held by OEFs has grown dramatically, and liquidity mismatch between asset holdings and liabilities in some OEFs creates run risk and systemic vulnerabilities.
- Preferred policy orientation:
  - Ex ante measures that reduce investor run risk are generally preferable to ex post measures that limit redemptions once runs have started.
- Specific recommendations:
  - Encourage more widespread adoption and appropriate calibration of price-based liquidity management tools (swing pricing, antidilution levies) so transaction costs are passed to redeeming investors and first-mover incentives are reduced.
  - Consider requiring funds to eliminate prospectus caps on swing factors and calibrate swing factors to fully reflect the price impact of fund asset sales.
  - Improve transparency and data:
    - Encourage disclosure of swing pricing practices and calibration methodologies.
    - Improve availability of aggregate fund flow data in real time to assist appropriate swing-factor setting during stress.
    - Collect additional data on funds’ liquidity risks and leverage, including synthetic leverage via derivatives.
  - Strengthen supervisory monitoring of liquidity risk management practices and consider mandating liquidity management tools and enhanced disclosure where voluntary adoption is insufficient due to competitive pressure or stigma.
  - Consider linking frequency of redemptions to portfolio liquidity for funds holding very illiquid assets or when price-based tools are operationally infeasible; offer early redemption in exchange for a calibration-based redemption fee where appropriate.
  - Recipient countries facing volatile capital flows from international funds should:
    - Emphasize continued deepening of domestic markets.
    - Use appropriate debt management tools.
    - Deploy macroeconomic, prudential, capital flow management, and foreign exchange intervention tools in line with the IMF’s Institutional View.
  - Bolster market liquidity and resilience by encouraging central clearing and greater transparency in bond trading.
  - Given cross-border spillovers and potential for recurring central bank interventions, consider broader regulation of investment funds if liquidity management practices remain inadequate, and pursue international regulatory coordination to ensure consistent deployment of liquidity management practices.

*Source: CHAPTER 3 ASSET PRICE FRAGILITY IN TIMES OF STRESS: ThE ROLE OF OPEN-END INvESTMENT FUNDS (Global Financial Stability Report: Navigating the High-Inflation Environment, October 2022).*

### 1. Total Net Assets and Flows of

### Box 3.1. Exchange-Traded Funds Generate Less Asset Price Fragility but May Also Be Vulnerable

### Evidence on net assets, flows, and mispricing
- Panel 1: Total Net Assets and Flows of Bond ETFs and OEFs (Trillions of US dollars; percent of total lagged net assets).
- Panel 3: Difference between ETF Price and Fund Net Asset Value in Percent of the Fund Net Asset Value for Bond ETFs (Percent).
  - ETF mispricing is calculated at daily frequency as the difference between the ETF closing price and the fund NAV divided by the fund NAV.
  - Sample: bond ETFs domiciled in the United States or Luxembourg.
  - During the March 2020 stress episode:
    - Discounts on ETFs increased dramatically, reaching more than 5 percent across all bond ETFs.
    - Discounts reached up to 27 percent for high-yield bond ETFs.
    - Discounts reached up to 13 percent for investment-grade bond ETFs.
- Data sources: FactSet; Morningstar; Refinitiv; and IMF staff calculations.

### Regression evidence on ownership and volatility
- Panel 2: Effect of OEF Ownership and ETF Ownership on Bond Return Volatility (Percent of median volatility).
  - Regression specification:
    - Dependent variable: weekly asset price volatilities.
    - Regressors include asset ownership variables and an interaction term between asset ownership and a stress dummy equal to 1 when the Chicago Board Options Exchange Volatility (VIX) Index is above its 90th sample percentile and zero otherwise.
    - “Mutual fund owned” refers to the total amount of an asset held by OEFs, but not by ETFs, as a percentage of its market capitalization.
    - Regression includes asset and issuer fixed effects.
    - Standard errors are clustered at the asset and quarter levels.
    - Solid bars indicate statistical significance at 10 percent or lower.

### Mechanisms and interpretation
- ETF features that reduce asset price fragility:
  - Intraday liquidity provision by ETFs permits trading without imposing immediate redemption pressure on underlying assets.
  - Creation and redemption arbitrage by authorized participants helps align ETF prices with underlying asset values and can absorb flows without direct sales of underlying assets by funds.
  - The ETF discount relative to NAV can be interpreted as a market-implied swing factor that an OEF with a similar portfolio structure and investor base would require.
- ETF vulnerabilities that can increase fragility:
  - The intraday liquidity provision makes ETFs attractive to liquidity traders with short-term horizons, facilitating transmission of nonfundamental shocks from liquidity traders to securities markets.
  - Arbitrage activities and intraday trading can increase nonfundamental volatility in asset markets (Ben-David, Franzoni, and Moussawi 2018).
  - ETFs can amplify the sensitivity of cross-border capital flows to global financial conditions (Converse, Levy-Yeyati, and Williams 2020).
  - Leveraged and inverse ETFs that rely on derivatives and short sales can introduce additional volatility because of the need to rebalance leveraged positions at the end of the trading day.

### Implications for mutual funds and liquidity management
- ETF price discounts during stressed liquidity conditions provide a revealed measure of the swing (transaction-cost) factor that mutual funds would need when applying swing pricing.
- The large ETF discounts observed in March 2020 (more than 5 percent on average; up to 27 percent for high-yield; up to 13 percent for investment-grade) indicate substantial liquidity stress and the potential size of swing factors needed for OEFs in similar circumstances.

*Sources: FactSet; Morningstar; Refinitiv; and IMF staff calculations.*

---


_Source: https://www.imf.org/-/media/files/publications/gfsr/2022/october/english/ch3.pdf_