## Chapter 2 — Risk and Resilience in the Global Foreign Exchange Market

## Source details

**Canonical URL:** [Chapter 2 — Risk and Resilience in the Global Foreign Exchange Market](https://www.imf.org/-/media/files/publications/gfsr/2025/october/english/ch2.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/gfsr/2025/october/english/ch2.pdf.md)
- [Structured JSON version](/-/media/files/publications/gfsr/2025/october/english/ch2.pdf.json)

---

### Overview and key findings
- The global foreign exchange (FX) market has an average daily turnover exceeding $9.6 trillion.
- Cross-border transactions account for about two-thirds of global FX market turnover; the US dollar is the dominant trading currency.
- Structural shifts—greater involvement of nonbank financial institutions (NBFIs), growing trade in derivatives (notably FX swaps), and greater electronification—have improved competition and efficiency while increasing vulnerabilities.
- Increased macrofinancial uncertainty can:
  - Significantly raise funding costs,
  - Impair liquidity,
  - Amplify excess exchange rate return volatility.
- Effects of shocks are more pronounced for:
  - Emerging markets,
  - Currencies with high NBFI participation,
  - Currencies with concentrated dealer networks,
  - Currencies with elevated hedging activity.
- FX market stress can spill over to other asset classes, tighten financial conditions, and pose risks to macrofinancial stability—especially in countries with significant currency mismatches and fiscal vulnerabilities.
- Outages in critical payment systems and settlement failures materially impair market liquidity, increase excess exchange rate return and its volatility, and raise the cost of FX transactions.
- Investor strategies diverge under elevated uncertainty; following US tariff announcements in early April 2025, some investors reduced US dollar holdings while others maintained exposures.

### Structural changes and vulnerabilities
- Market structure changes
  - Trading shifted from concentration among large dealers in the 1990s toward greater NBFI participation.
  - Decline in the share of spot trading and notable growth in derivatives, especially FX swaps used for funding and hedging.
  - Expansion of execution methods and trading platforms with increased electronification.
- Benefits and risks of NBFI involvement
  - Benefits: increased diversity, potential for greater liquidity, lower transaction costs, improved price discovery and risk sharing.
  - Risks: many NBFIs face less regulatory oversight, may lack access to central bank facilities, employ leverage, short-term arbitrage, and high-frequency trading strategies that can amplify market swings and shift inventory risk onto market-making dealers.
  - Liquidity mismatches in NBFIs (for example, mutual funds funding longer-term or less liquid assets with short-term liabilities) can heighten systemic risk during volatility.
- Dealer concentration and currency mismatches
  - Nearly half of global FX turnover is intermediated by a small group of dominant dealers (mostly large, regulated banks), exposing markets if these institutions scale back activity under stress.
  - Persistent currency mismatches drive sustained demand for short-term FX derivatives, increasing rollover and funding risks when conditions tighten.

### Risks, channels, and empirical patterns
- How shocks affect FX markets
  - Macrofinancial uncertainty is captured using three indicators: VIX, MOVE, and the economic policy uncertainty (EPU) index.
  - A surge in financial uncertainty can dampen investor risk appetite and prompt a flight to quality toward US dollar assets.
  - Portfolio rebalancing toward dollar assets:
    - Increases demand for dollar-denominated assets and unwinds positions in other currencies.
    - Leads institutions outside the United States to hedge dollar exposures using FX swap contracts (buy dollars spot, sell dollars forward).
    - Forces dealer banks to expand balance sheets and often borrow dollars, which can be costly when regulatory constraints limit dealers’ ability to supply liquidity.
  - When dealers face balance sheet constraints, the cost of FX swaps rises and cross-currency bases widen, tightening FX market conditions and limiting portfolio rebalancing.
- Amplification mechanisms
  - NBFI trading strategies, leverage, margin calls, and forced deleveraging can amplify volatility and liquidity strains.
  - Opacity in over-the-counter derivatives markets complicates risk monitoring and can obscure systemic risk buildup.
  - Algorithmic and high-frequency traders often withdraw under stress, reducing liquidity and amplifying volatility.
- Measured responses to uncertainty shocks
  - Uncertainty shocks (unexpected index changes exceeding two standard deviations) can raise weekly spot trading growth by up to 24 percentage points in episodes like the 2020 COVID-19 turmoil.
  - After VIX or MOVE spikes, weekly growth rates in trading volumes of nonresident NBFIs rise by about 40 percentage points on average; dealer and nondealer banks’ weekly growth rates rise by about 15 percentage points.
  - VIX shocks raise FX swap activity by about 5 percentage points among banks and about 10 percentage points among NBFIs.
  - Uncertainty shocks widen the three-month basis by up to 13 basis points over a week (one-standard-deviation increase in global macrofinancial uncertainty indicators).
  - Weekly excess exchange rate return volatility increases by about 5–10 basis points in response to shocks to the VIX, EPU, and MOVE indices.
  - Bid-ask spreads widen by 1–3 basis points in response to those shocks.
  - Effects persist for up to three months, peaking about four weeks after the shock.
- Differential impacts
  - Including emerging market currencies yields stronger and more persistent effects across FX market condition measures.
  - Cross-currency bases and bid-ask spreads widen, on average, more than twice the amounts estimated for advanced economies when emerging markets are included.
- Amplifiers of transmission
  - Banks’ FX mismatches (measured by the cross-currency funding ratio, CCFR), increased NBFI hedging activity, higher dealer concentration (HHI), and lower dealer capital ratios amplify CIP deviations and excess exchange rate return volatility.
- Stabilizing backstops
  - Federal Reserve US dollar liquidity swap lines and related facilities ease dollar funding stress, limit CIP deviations, and stabilize FX swap markets.
  - Newly activated swap lines reduced CIP deviations by up to 30 basis points in the 2020 episode and significantly lowered excess exchange rate return volatility.
  - Economies with stronger international reserve buffers—about one standard deviation above the average—experience notably smaller CIP deviations and lower excess exchange rate return volatility following shocks.

### Operational risk and market infrastructure vulnerabilities
- Settlement and operational risks
  - Settlement risk (one party delivers currency but does not receive the countervalue) is acute in cross-border transactions due to time zone differences and operational delays.
  - Most emerging market currencies remain outside PvP frameworks, leaving a substantial portion of FX transactions exposed to settlement risk.
  - Operational disruptions (technical failures, cyberattacks, power outages) can impair FX market functioning, generate liquidity strains and volatility, and increase failed settlements.
  - Recent examples cited include the 2018 Fedwire cyber incident and TARGET2 outages in 2000 and 2025.
- PvP systems and effects
  - About 25 percent of the deliverable turnover of currencies is without risk mitigation mechanisms.
  - Natural experiment—Hungary’s accession to CLS on November 16, 2015:
    - Average daily excess exchange rate returns declined by about 11 basis points in the one-month window before vs after CLS participation.
    - Volatility declined by 3 basis points in the same window.
    - Average excess returns over the month before CLS accession were about 28 basis points.
  - Panel analysis (January 2000 to May 2025) covering 26 currencies:
    - CLS participation is associated with a significant decline of 34 basis points in excess FX returns, on average.
    - CLS participation is associated with a 3 basis point decline in volatility, on average.
  - Interpretation: PvP systems like CLS reduce settlement uncertainty and counterparty exposure, lowering settlement risk premiums and volatility.
- Interdealer platform outages
  - Outage effects on affected currencies:
    - Spot and swap market bid-ask spreads increased during outages.
    - Average decline in spot market volumes on outage days: $2.8 billion across affected currencies (all dollar volumes adjusted for inflation to December 2024 US dollars).
    - Estimated relation: a $1 billion decrease in trading volume is associated with a 0.3 basis point widening of spot market bid-ask spreads.
    - Trading volume required to move daily FX returns by one standard deviation (about 3.4 percent) declined from $19.2 billion to $18.6 billion.
    - Dispersion of transaction prices across counterparties increased by about 0.2 standard deviations (about 11 percent).
  - Aggregate spillovers:
    - Transaction costs rose in spot and swap markets even for currencies traded on venues that remained operational.
    - Average bid-ask spreads against the US dollar widened from 3 to 4 basis points in the spot market and from 4 to 5 basis points in the swap market during outages.
  - Interpretation: Even relatively short-lived outages of interdealer trading platforms can materially impair FX market liquidity and raise transaction costs, with both direct effects and spillovers.

### Evolving market structure and key statistics
- Market expansion and composition
  - Average daily trading volumes have increased fivefold since the late 1990s.
  - Growth largely driven by FX swap and spot transactions; FX swap activity has risen notably reflecting increased NBFI participation.
  - A majority of FX swaps are of short duration, typically with tenors up to three months.
  - The US dollar remains the dominant trading currency in spot and swap markets.
  - About one-fourth of transactions involve the euro against the US dollar.
  - One-fifth of transactions involve the Japanese yen.
  - The euro–US dollar share of total transactions has declined from about one-third in 2015; the share of other currencies relative to the dollar has increased.
  - The dollar’s relative importance appears to have remained stable through May 2025, with no major shift after the US tariff announcements in early April 2025.
- Counterparty structure and interconnectedness
  - A large share of FX transactions takes place between banks.
  - Among NBFIs, investment funds dominate FX trading.
  - Banks remain central to the FX ecosystem and form the core of the global FX network; banks in major economies, especially the United States, are highly interconnected and concentrated in intermediation of major currency pairs.
  - Bank participation varies by currency pair: euro–US dollar trades are intermediated by US banks and banks in France, Germany, and the United Kingdom; yen–US dollar trades are dominated by Japanese and US banks.
- NBFI participation and hedging pressure
  - Median share of NBFIs in FX trading activity has averaged about 8 percent over the past decade.
  - Some currency pairs, such as euro–US dollar, have had NBFI participation exceeding 15 percent in recent years.
  - Hedging pressure (net short swap position of NBFIs relative to total outstanding market swap position) has generally been on an increasing trend and is positively correlated across major currencies.

### Financial spillovers and macroeconomic implications
- Cross-currency basis widening and financial conditions
  - A one-standard-deviation widening (about 25 basis points) of cross-currency bases reduces longer-term sovereign bond yields by about 25 basis points, with effects lasting up to three months.
  - A one-standard-deviation widening of cross-currency bases tightens financial conditions by 0.4 to 0.7 standard deviations over the following year.
  - In economies with high FX mismatches, a one-standard-deviation shock can tighten financial conditions by as much as two standard deviations; in economies with low FX mismatches, the effect is negligible.
- Channels to real economy and policy transmission
  - Elevated FX volatility and hedging costs raise uncertainty and the cost of managing currency exposure, potentially affecting yields and risk premiums in sovereign bond and equity markets.
  - Higher funding costs can erode intermediation capacity, tighten financial conditions, and amplify systemic stress, creating adverse macrofinancial feedback loops.
  - Excess exchange rate return volatility can complicate monetary policy transmission and undermine investor confidence, prompting portfolio rebalancing away from riskier assets.

### Policy recommendations and resilience measures
- Enhance FX market surveillance
  - Implement systemic risk monitoring, stress testing, and scenario analysis to capture liquidity shocks and spillovers.
  - Enhance FX liquidity stress tests and systemwide stress tests to incorporate scenarios with heightened volatility, wider bid-ask spreads, and cross-currency bases.
  - Monitor and mitigate rollover and liquidity risks from short-tenor FX swap positions.
  - Use scenario analysis that includes severe and persistent technical failures, cyberattacks, physical disasters, and defaults by major FX dealers.
  - Close data gaps by improving reporting and data sharing on bilateral exposures, settlement practices, intraday trading, and counterparty concentrations, especially for transactions outside centralized infrastructures like PvP systems.
- Ensure robust liquidity and capital buffers
  - Require institutions with dominant/systemic FX roles to maintain adequate hedges and capital and liquidity buffers.
  - Strengthen access to intraday central bank liquidity and credit facilities, including for NBFIs, while limiting moral hazard.
  - Economies reliant on external financing should maintain sufficient international reserve buffers.
  - Strengthen and expand central bank swap line networks; use the IMF’s lending toolkit as part of the global financial safety net for countries facing FX liquidity pressures.
  - Consider FX intervention and macroprudential and capital flow management measures calibrated to country-specific conditions when external shocks cause undesirable macroeconomic fluctuations in the presence of significant FX mismatches.
- Strengthen operational resilience and reduce settlement risk
  - Strengthen operational resilience of financial market infrastructures per the Principles for Financial Market Infrastructures (BIS-CPSS-IOSCO 2012): identify operational risk sources, implement robust systems, comprehensive business continuity planning, cyber resilience frameworks, and regular contingency testing.
  - Coordinate responses among central banks and oversight agencies given cross-jurisdictional interconnectedness.
  - Reduce FX settlement risks through wider adoption of PvP arrangements; interim dealer-bank arrangements include “pre-settlement netting” and “on-us” settlement.
  - Implement strong anti-money laundering/combating the financing of terrorism measures to reduce settlement uncertainty.
  - Explore policy initiatives leveraging digital technologies—linking faster payment systems or developing cross-border central bank digital currency—while recognizing potential risks if not properly designed and regulated.
- Encourage migration toward well-designed financial platforms
  - Better-designed interoperable financial platforms can lower transaction costs and volatility, reduce counterparty and settlement risks, and mitigate dealer constraints in over-the-counter markets.

### Data and empirical approach
- Core dataset: unique CLS Group data covering FX spot and swap transactions with daily and weekly information from January 1, 2015, to May 31, 2025, on FX flows and pricing for 18 major currencies, disaggregated by four institutional sectors (banks, investment funds, other ...).
- Macrofinancial uncertainty measures used: VIX, MOVE, EPU.
- FX market condition measures evaluated: (1) cross-currency basis (through CIP deviation), (2) annualized excess spot-return volatility, and (3) quoted bid-ask spreads.
- Baseline panel: 11 major US dollar pairs with disaggregated CLS data; alternative sample includes 16 emerging market currencies.
- Empirical specifications: panel regressions with time and currency effects; confidence intervals at 90 percent obtained using Driscoll-Kraay standard errors with the number of lags equal to 4/T.

### Key quantitative figures and illustrative points
- Average daily turnover: exceeding $9.6 trillion.
- Cross-border share of global FX turnover: about two-thirds.
- Median NBFI share in FX trading activity over the past decade: about 8 percent.
- NBFI participation in some pairs (euro–US dollar): exceeding 15 percent in recent years.
- Uncertainty shock impact on weekly spot trading growth in COVID-19–like episodes: up to 24 percentage points.
- Weekly growth in trading volumes after VIX/MOVE spikes:
  - Nonresident NBFIs: about 40 percentage points.
  - Dealer and nondealer banks: about 15 percentage points.
- VIX shock effect on three-month basis: up to 13 basis points over a week (one-standard-deviation increase).
- Weekly excess exchange rate return volatility increase after shocks: about 5–10 basis points.
- Bid-ask spread widening after shocks: 1–3 basis points.
- Persistence: effects can last up to three months, peaking about four weeks after shock.
- Standard deviations used for normalization:
  - CIP deviations: about 40 basis points,
  - Excess exchange rate return volatility: 0.3 percentage point,
  - Bid-ask spreads (normalized by mid-rate): 0.06 percent.
- PvP and settlement statistics:
  - About 25 percent of deliverable turnover without risk mitigation mechanisms.
  - CLS establishment: 2002.
  - Hungary joined CLS: November 16, 2015.
  - Hungary CLS accession effects (one-month before vs after): excess returns down by about 11 basis points; volatility down by 3 basis points; pre-CLS average excess returns about 28 basis points.
  - Panel-average CLS effect (Jan 2000–May 2025): excess FX returns down by 34 basis points; volatility down by 3 basis points.
- Historical operational-loss incidents:
  - Bankhaus Herstatt failure: June 26, 1974.
  - 2008 unilateral loss example: €300 million (equivalent to $426 million at the time) transferred to Lehman Brothers without receiving corresponding payment.
  - March 2020 Barclays loss: $129 million.
- Outage-related market metrics:
  - Average decline in spot market volumes on outage days (affected currencies): $2.8 billion (December 2024 US dollars).
  - Volume–spread relationship: $1 billion decrease in trading volume → 0.3 basis point widening of bid-ask spreads.
  - Trading volume to move daily returns by one SD (~3.4 percent): fell from $19.2 billion to $18.6 billion.
  - Price dispersion increase during outages: ~0.2 standard deviations (~11 percent).
  - Spot bid-ask spreads widened from 3 to 4 basis points on average; swap bid-ask spreads widened from 4 to 5 basis points on average during outages.
- April 2, 2025 US tariff announcement episode:
  - Broad US dollar index depreciation on impact: about 2 percent.
  - Spot dollar purchases by nonresident investors rose by about $265 billion on a net basis between January 1 and April 1, 2025.
  - Cumulative net spot purchases remained broadly stable as of end-May 2025.
  - Cross-country sectoral patterns: Canada shifted from net buyer to net seller after mid-April 2025; major euro area countries increased net spot dollar sales after April 2, 2025.
  - Non-US nonbank financial institutions were active buyers of US dollars around the episode; non-US nonbank investors’ FX swap activity against the US dollar increased notably after April 2, 2025.

*Source: Chapter 2, "Risk and Resilience in the Global Foreign Exchange Market," Global Financial Stability Report, October 2025.*

### Chapter 2 at a Glance

### Chapter 2 at a Glance

### Overview and key findings
- The global foreign exchange (FX) market is a cornerstone of the international monetary and financial system with an average daily turnover exceeding $9.6 trillion.
- Cross-border transactions account for about two-thirds of global FX market turnover, with the US dollar being the dominant trading currency.
- Structural shifts—including increased involvement of nonbank financial institutions (NBFIs), growing trade in derivatives (notably FX swaps), and greater electronification—have enhanced competition and efficiency but also increased vulnerabilities.
- Increased macrofinancial uncertainty can strain FX market conditions by significantly raising funding costs, impairing liquidity, and amplifying excess exchange rate return volatility.
- The effect of shocks is more pronounced for emerging markets and for currencies with high NBFI participation, concentrated dealer networks, and elevated hedging activity.
- FX market stress can spill over to other asset classes, tightening financial conditions and posing risks to macrofinancial stability—especially in countries with significant currency mismatches and fiscal vulnerabilities.
- Outages in critical payment systems and risk of settlement failure significantly impair market liquidity and increase excess exchange rate return and its volatility, raising the cost of FX transactions.
- Investor strategies are evolving amid elevated uncertainty: following the US tariff announcements in early April 2025, investors in some countries reduced their US dollar holdings, whereas others maintained exposures, highlighting diverging cross-country responses.

### Structural changes and vulnerabilities
- Market structure changes
  - Shift from trading concentrated among large dealers in the 1990s toward greater NBFI participation.
  - Decline in share of spot trading and notable growth in derivatives, especially FX swaps used for funding and hedging.
  - Expansion of execution methods and trading platforms with increased electronification.
- Benefits and risks of NBFI involvement
  - Benefits: increased diversity, potential for greater liquidity, lower transaction costs, improved price discovery and risk sharing.
  - Risks: many NBFIs face less regulatory oversight, may lack access to central bank facilities, employ leverage, short-term arbitrage, and high-frequency trading strategies that can amplify market swings and shift inventory risk onto market-making dealers.
  - Liquidity mismatches in NBFIs (for example, mutual funds funding longer-term or less liquid assets with short-term liabilities) can heighten systemic risk during volatility.
- Dealer concentration and currency mismatches
  - Nearly half of global FX turnover is intermediated by a small group of dominant dealers (mostly large, regulated banks), exposing markets if these institutions scale back activity under stress.
  - Persistent currency mismatches drive sustained demand for short-term FX derivatives, increasing rollover and funding risks when conditions tighten.

### Risks, channels, and empirical patterns
- How shocks affect FX markets
  - Macroeconomic uncertainty shifts investor risk sentiment and interest rate expectations, triggering rapid portfolio adjustments, liquidity strains, and volatility.
  - Historical episodes of elevated global macrofinancial uncertainty show that FX funding and market liquidity pressures—reflected in wider cross-currency bases (CIP deviation), bid-ask spreads, and excess exchange rate return volatility—tend to rise with uncertainty.
- Amplification mechanisms
  - NBFI trading strategies, leverage, margin calls, and forced deleveraging can amplify volatility and liquidity strains.
  - Opacity in over-the-counter derivatives markets complicates risk monitoring and can obscure systemic risk buildup.
  - Algorithmic and high-frequency traders often withdraw under stress, reducing liquidity and amplifying volatility.
- Operational and settlement risks
  - Settlement risk (one party delivers currency but does not receive the countervalue) is acute in cross-border transactions due to time zone differences and operational delays.
  - Most emerging market currencies remain outside payment-versus-payment (PvP) frameworks, leaving a substantial portion of FX transactions exposed to settlement risk.
  - Operational disruptions to FX market infrastructure (technical failures, cyberattacks, power outages) can impair FX market functioning, generate liquidity strains and volatility, and increase failed settlements.
  - Recent examples cited include the 2018 Fedwire cyber incident and TARGET2 outages in 2000 and 2025.
- Spillovers to other asset classes
  - Elevated FX volatility and hedging costs (wider cross-currency bases) raise uncertainty and the cost of managing currency exposure, potentially affecting yields and risk premiums in sovereign bond and equity markets.
  - Higher funding costs can erode intermediation capacity, tighten financial conditions, and amplify systemic stress, creating adverse macrofinancial feedback loops.

### Key policy recommendations
- Enhance FX market surveillance through systemic risk monitoring, stress testing, and scenario analysis to capture liquidity shocks and spillovers.
- Close critical FX data gaps by improving reporting and transparency, especially regarding NBFIs and bilateral exposures outside centralized infrastructures.
- Ensure robust liquidity and capital buffers, backed by effective safeguards, such as access to central bank liquidity with proper oversight, sufficient international reserves, and expanded central bank swap lines.
- Strengthen operational resilience of financial market infrastructures and financial institutions through cyber risk frameworks, contingency planning, and coordinated oversight.
- Reduce settlement risks and market inefficiencies in over-the-counter FX markets by encouraging payment-versus-payment adoption and exploring digital innovations to develop interoperable financial platforms.

### Data and empirical approach
- The chapter uses a unique data set covering FX spot and swap transactions from CLS Group with daily and weekly information from January 1, 2015, to May 31, 2025, on FX flows and pricing for 18 major currencies, disaggregated by four institutional sectors (banks, investment funds, other ...).

*Source: Chapter 2, "Risk and Resilience in the Global Foreign Exchange Market," Global Financial Stability Report, October 2025.*

### CHAPTER 2 RISk ANd RESILIENCE IN ThE GLOBAL FOREIGN ExChANGE MARkET

### CHAPTER 2 RISk ANd RESILIENCE IN ThE GLOBAL FOREIGN ExChANGE MARkET

### Macrofinancial shocks and FX market transmission: conceptual framework
- The global FX market is decentralized, mostly over-the-counter, and enables continuous trading across time zones among dealers (banks and NBFIs that act as market makers), nondealer banks and NBFIs, nonfinancial firms, central banks, and retail investors.
- Macrofinancial uncertainty is captured using three indicators: the Chicago Board Options Exchange Volatility Index (VIX), the Merrill Lynch Option Volatility Estimate (MOVE) index, and the economic policy uncertainty (EPU) index of Baker, Bloom, and Davis (2016).
- A surge in financial uncertainty (often captured by the VIX) can dampen investor risk appetite and prompt a flight to quality, reallocating portfolios toward safer assets, such as those denominated in US dollars.
- Portfolio rebalancing toward dollar assets:
  - Increases demand for dollar-denominated assets and unwinds positions in other currencies.
  - Leads institutions outside the United States to hedge increased dollar exposures using FX swap contracts (buy dollars spot, sell dollars forward).
  - Forces dealer banks to expand balance sheets and often borrow dollars, which can be costly when regulatory constraints limit dealers’ ability to supply liquidity.
- When dealers face balance sheet constraints, the cost of FX swaps rises and cross-currency bases between the dollar and other currencies widen, tightening FX market conditions and placing a limit on portfolio rebalancing.

### Amplification channels and feedback loops
- FX swaps serve multiple roles: hedging, speculation (carry trades), and short-term dollar funding for longer-term positions (maturity transformation).
- Widening US dollar–foreign currency basis:
  - Signals rising costs and reduced availability of dollar funding, particularly for institutions outside the United States.
  - Can force unhedged exposures or asset sales, increasing market volatility.
  - Creates a feedback loop: higher volatility → greater flight to safety → increased demand for safe assets and FX swaps → further widening of cross-currency basis → deeper funding stress.
- Wider bid-ask spreads and reduced liquidity:
  - Rising uncertainty raises the cost of holding foreign currencies for dealers (higher capital needs), passed to customers as wider bid-ask spreads.
  - Wider spreads discourage participation, reduce liquidity, and make prices more sensitive to orders, reinforcing volatility.
- Structural fragilities that amplify shocks:
  - Dealer concentration in market making increases the likelihood that regulatory constraints during price declines will lead to funding and liquidity stress.
  - Greater NBFI share and leverage, smaller liquidity buffers, and larger FX mismatches increase reliance on FX swaps and procyclical behavior during stress.
- Transmission to financial system and real economy:
  - Elevated FX funding costs and wider cross-currency bases can tighten capital constraints (for example, leverage ratio requirements), prompting deleveraging, asset sales, and contraction of credit supply.
  - Higher hedging costs reduce hedge ratios; institutions may self-insure by shifting into safer, more liquid assets, increasing demand for sovereign bonds and potentially lowering yields in countries with stronger fiscal fundamentals.
  - Excess exchange rate return volatility can complicate monetary policy transmission and undermine investor confidence, prompting portfolio rebalancing away from riskier assets.

### Operational risk and market infrastructure vulnerabilities
- Operational disruptions (trading platform outages, messaging system failures, payment and settlement interruptions) can impair FX market functioning, increase illiquidity, and raise counterparty risk.
- The decentralized structure and substitutability across platforms have so far limited systemic impact from single-platform outages, but simultaneous outages (cyber incidents, power outages) across multiple platforms could trigger systemic stress by cutting off access to liquidity and risk management tools.
- Prolonged disruptions to payment systems and settlement infrastructures (for example, CLS; TARGET2; Fedwire; and the Clearing House Automated Payment System) are inherently disruptive and require robust safeguards and backup arrangements.

### The evolving structure of the global FX market: key trends and statistics
- Market expansion:
  - Average daily trading volumes have increased fivefold since the late 1990s.
  - Growth has been driven largely by FX swap and spot transactions, with FX swap activity rising notably in recent years reflecting increased NBFI participation.
- Tenor composition:
  - A majority of FX swaps are of short duration, typically with tenors up to three months.
- Currency composition and shares:
  - The US dollar remains the dominant trading currency in spot and swap markets.
  - About one-fourth of transactions involve the euro against the US dollar.
  - One-fifth of transactions involve the Japanese yen.
  - The euro–US dollar share of total transactions has declined from about one-third in 2015, while the share of other currencies relative to the dollar has increased.
  - The dollar’s relative importance appears to have remained stable through May 2025, with no major shift after the US tariff announcements in early April 2025.
- Counterparty structure and interconnectedness:
  - A large share of FX transactions takes place between banks.
  - Among NBFIs, investment funds dominate FX trading.
  - Banks remain central to the FX ecosystem and form the core of the global FX network; banks in major economies, especially the United States, are highly interconnected and concentrated in intermediation of major currency pairs.
  - Bank participation varies by currency pair: euro–US dollar trades are intermediated by US banks and banks in France, Germany, and the United Kingdom; yen–US dollar trades are dominated by Japanese and US banks.
- Data sources noted: BIS OTC Derivatives Statistics; CLSMarketData; IMF staff calculations.

### Policy implications and resilience measures
- Mitigants to transmission and amplification:
  - Require financial institutions to hold adequate foreign currency liquidity buffers and stable dollar funding, such as customer deposits, to reduce reliance on short-term wholesale dollar funding.
  - Central bank interventions, including dollar liquidity swap lines, have been important historically for breaking cycles generated by shock-induced portfolio rebalancing and restoring market functioning.
  - Strengthen safeguards and backup arrangements for payment and settlement infrastructures to contain systemic risks from prolonged operational disruptions.
- Address structural vulnerabilities:
  - Monitor dealer concentration and the growing role of NBFIs in FX markets to limit procyclical behavior and ensure adequate buffers and risk management.
  - Enhance transparency and monitoring of cross-currency bases and hedging pressures to anticipate funding stress and its real-economy spillovers.

*Source: CHAPTER 2 RISk ANd RESILIENCE IN ThE GLOBAL FOREIGN ExChANGE MARkET (IMF PDF chapter).*

### 3. Network of Spot Yen–US Dollar Sectoral Trading Activity, 2024

### 3. Network of Spot Yen–US Dollar Sectoral Trading Activity, 2024

### Dealer concentration across currency pairs and instruments
- Dealer bank concentration varies across currency pairs and instruments (Figure 2.5, panel 3).
- Dealer concentration is measured using the Herfindahl-Hirschman Index; the index scale ranges from 0 to 10,000. The horizontal lines refer to the Bank for International Settlements’ benchmarks for concentration in FX markets (Figure 2.5, panel 4).
- Certain currency pairs show higher degrees of concentration, evidenced by the upper range of the interquartile distribution in Figure 2.5, panel 4.
- The degree of dealer concentration is also high in swap markets with longer tenors, indicating that certain segments remain reliant on a concentrated group of dealers and that this reliance can amplify systemic risk if key dealers face financial or operational disruptions.

### NBFI participation and hedging pressure
- The median share of NBFIs in FX trading activity has averaged about 8 percent over the past decade (Figure 2.5, panel 5).
- Some currency pairs, such as euro–US dollar, have had NBFI participation exceeding 15 percent in recent years (Figure 2.5, panel 5).
- Hedging pressure is defined as the net short swap position of NBFIs in specific currencies with respect to the US dollar relative to the total outstanding market swap position (Bräuer and Hau 2023). This measure is shown as a three-month rolling average (Figure 2.5, panel 6).
- Hedging of currency exposures among NBFIs has generally been on an increasing trend (Figure 2.5, panel 6).
- Hedging pressure appears to be positively correlated across major currencies, implying a synchronized need for dollar hedging that can strain liquidity and amplify volatility in stress episodes.
- Hedging pressure from NBFIs across countries tends to be strongly correlated with these institutions’ net bond investment positions with respect to the United States (see Online Annex 2.3; BIS 2025b).

### Macrofinancial uncertainty and FX trading dynamics — direction and magnitude
- Multiple uncertainty measures are considered: VIX (financial market volatility), MOVE (monetary policy uncertainty via bond market volatility), and EPU (economic policy uncertainty).
- Uncertainty shocks are defined as unexpected changes in the values of those indices exceeding two standard deviations (Figure 2.7, panel 1; Online Annex 2.4).
- An increase in uncertainty tends to raise nonresident demand for US dollars. Uncertainty shocks of the magnitude observed during episodes like the 2020 COVID-19 turmoil can raise weekly spot trading growth by up to 24 percentage points (Figure 2.7, panel 1).
- NBFIs, particularly investment funds, respond more strongly to global uncertainty shocks than banks or nonfinancial firms:
  - After a spike in the VIX or MOVE index, weekly growth rates in trading volumes of nonresident NBFIs rise by about 40 percentage points, on average (Figure 2.7, panels 2 and 3).
  - Dealer and nondealer banks’ weekly growth rates rise by about 15 percentage points in the same episodes (Figure 2.7, panels 2 and 3).
- Nonresident NBFIs typically increase purchases of US dollars in spot and swap markets when the VIX or US EPU index spikes (Figure 2.6, panels 1 and 2). Net spot purchases of other safe haven currencies (euro, Swiss franc, yen) also react strongly to these shocks (Online Annex Figure 2.3.5).
- In the April 2025 episode triggered by US tariff announcements, nonresident NBFIs increased purchases of safe haven assets; demand for US dollar swaps by non-US NBFIs rose sharply as they shifted to hedge previously unhedged exposures. Overall net spot purchases of US dollars by non-US banks and non-US NBFIs were relatively subdued compared with 2020 COVID-19 levels (Box 2.1; Online Annex Figure 2.7.1).

### FX swap market responses and hedging behavior
- NBFIs account for a growing share of short-term US dollar funding and hedging in the FX swap market.
- Heightened uncertainty raises FX swap activity in longer maturities, indicating stronger hedging rather than interdealer activity (Figure 2.8, panel 1).
- VIX shocks raise FX swap activity by about 5 percentage points among banks and nearly twice as much among NBFIs (about 10 percentage points), consistent with spot market patterns (Figure 2.8, panel 2).
- The response to MOVE shocks is similar but somewhat weaker, potentially because term-spread differential controls absorb part of the MOVE effect in regression models.
- The response is especially pronounced for longer-dated transactions (those of more than seven days), which are more indicative of hedging than short-term trades (Du, Tepper, and Verdelhan 2018; Bräuer and Hau 2023).
- FX exposures and currency mismatches materially drive hedging behavior:
  - Countries whose banks have larger dollar funding gaps tend to exhibit stronger responses to uncertainty shocks (Figure 2.8, panel 3).
  - Net international investment positions in the dollar help explain cross-country differences in hedging behavior; net hedging activity is proportional to a country’s net investment position in the corresponding currency (Gabaix and Maggiori 2015; Devereux and Yu 2020; Liao and Zhang 2025) (Figure 2.8, panel 4).
- The banking sector’s US dollar FX mismatch is measured as the ratio of dollar-denominated assets minus dollar-denominated liabilities, normalized by dollar-denominated assets. The dollar net investment position is the difference between foreign holdings of US long-term debt and US holdings of foreign long-term debt, scaled by total outstanding holdings of long-term debt.

### Uncertainty shocks, market stress indicators, and sample coverage
- The analysis evaluates effects of uncertainty shocks on three FX market condition measures: (1) cross-currency basis (through CIP deviation), (2) annualized excess spot-return volatility, and (3) quoted bid-ask spreads (Online Annex 2.5).
- The baseline panel uses 11 major US dollar pairs with disaggregated CLS data; an alternative sample includes 16 emerging market currencies to assess impacts on emerging market FX (coverage varies across regressions; Online Annex 2.5).

*Italic: International Monetary Fund, GLOBAL FINANCIAL STABILITY REPORT: ShIFTING GROuNd BENEATh ThE CALM, October 2025 — Chapter 2 excerpt.*

### 1. Change in US Dollar Swap Inows after an Uncertainty Shock

### 1. Change in US Dollar Swap Inflows after an Uncertainty Shock

### Effect of uncertainty shocks on FX market conditions
- A one-standard-deviation increase in global macrofinancial uncertainty indicators (such as the VIX or the EPU index) widens the three-month basis by up to 13 basis points over a week.
- The effect of changes in the MOVE index is not statistically significant once the effect of term-spread differentials is controlled for.
- Large uncertainty shocks—those exceeding twice the standard deviation—produce disproportionately larger effects, indicating a nonlinear response.
- Weekly excess exchange rate return volatility increases by about 5–10 basis points in response to shocks to the VIX, EPU, and MOVE indices.
- Bid-ask spreads widen by 1–3 basis points in response to those shocks, equivalent to about half a standard deviation, on average.
- These effects persist for up to three months, peaking about four weeks after the shock.

### Differential impact on emerging market currencies
- When emerging market currencies are included, effects tend to be somewhat stronger and more persistent across all measures of FX market conditions.
- Cross-currency bases and bid-ask spreads widen, on average, more than twice the amounts estimated for advanced economies.
- Estimated effects on excess exchange rate return volatility also increase notably for emerging market currencies.

### Quantitative baselines and normalization statistics
- The standard deviation used for normalization is about:
  - 40 basis points for CIP deviations,
  - 0.3 percentage point for excess exchange rate return volatility,
  - 0.06 percent for bid-ask spreads (normalized by the mid-rate).

### Market fragilities that amplify transmission
- Banks’ FX mismatches (measured by the cross-currency funding ratio, CCFR) amplify the effect of uncertainty on CIP deviations and on excess exchange rate return volatility.
- Increased NBFI hedging activity (net hedging activity of NBFIs measured as the difference between aggregate short and long FX swap or forward positions, scaled by the global average of outstanding US dollar contracts) tightens synthetic dollar funding conditions and amplifies CIP deviations.
- Dealer concentration (Herfindahl-Hirschman Index, HHI) and a larger share of NBFIs in a currency’s trading amplify excess exchange rate return volatility by raising transaction costs, reducing market depth, and increasing order-flow imbalances.
- Dealer balance sheet constraints matter: stronger dealer capital ratios mitigate the effects of uncertainty shocks by improving capacity to intermediate FX swaps and absorb risk; lower capital ratios worsen FX market conditions.

### Policy backstops and stabilizing factors
- The Federal Reserve’s US dollar liquidity swap lines and related facilities ease dollar funding stress, limit CIP deviations, and help stabilize FX swap markets.
- Newly activated swap lines reduced CIP deviations by up to 30 basis points, nearly offsetting the entire impact of the initial VIX shock in the 2020 episode, and significantly lowered excess exchange rate return volatility.
- International reserves are stabilizing: economies with stronger reserve buffers—about one standard deviation above the average—experience notably smaller CIP deviations and lower excess exchange rate return volatility following macrofinancial uncertainty shocks.

### Spillovers to other asset classes and financial conditions
- A widening of cross-currency bases triggers a flight-to-quality, compressing local currency sovereign bond yields and reducing stock prices.
- A one-standard-deviation widening (about 25 basis points) reduces longer-term sovereign bond yields by about 25 basis points, with effects lasting up to three months.
- A one-standard-deviation widening of cross-currency bases tightens financial conditions by 0.4 to 0.7 standard deviations over the following year.
- The transmission to broader financial conditions is amplified by vulnerabilities such as currency mismatches on financial institutions’ balance sheets or elevated public debt.
- In economies with high FX mismatches, a one-standard-deviation shock can tighten financial conditions by as much as two standard deviations; in economies with low FX mismatches, the effect is negligible.

*Source: Chapter 2 excerpt — GLOBAL FINANCIAL STABILITY REPORT: SHIFTING GROUND BENEATH THE CALM (International Monetary Fund | October 2025).*

### 2. Effect of Cross-Currency Bases on Financial

### 2. Effect of Cross-Currency Bases on Financial Conditions with FX Mismatches

### Financial spillovers from cross-currency basis widening
- Panel regressions show that cross-currency bases of local currencies against the US dollar affect financial conditions; estimates instrument cross-currency bases with idiosyncratic demand shocks to dollar funding in the FX swap market for tenors: less than 7 days, between 7 and 35 days, and more than 35 days.
- The cross-currency basis and the idiosyncratic demand shocks are standardized for each currency and tenor.
- Regression specifications include time and currency effects; confidence intervals shown are 90 percent, obtained using Driscoll-Kraay standard errors with the number of lags equal to 4/T, where T denotes the number of time periods in the sample.
- Currencies analyzed: the euro, the Japanese yen, the British pound, the Swiss franc, the Canadian dollar, the Australian dollar, the New Zealand dollar, the Swedish krona, and the Norwegian krone, trading against the US dollar.

### Amplification by FX mismatches and fiscal vulnerabilities
- Cross-currency basis widening tightens financial conditions "particularly when currency mismatches are large" (panel 2 interaction with FX-mismatch dummy = 1 when FX mismatches are above the sample median).
- The effect of cross-currency basis widening on five-year sovereign bond yields is greater for economies with high public debt relative to GDP (panel 3 interaction with debt-to-GDP dummy = 1 when the debt-to-GDP ratio is above the sample median), consistent with flight-to-quality favoring fiscally sound economies.
- Similar results for sovereign bond yields are obtained when fiscal vulnerability is proxied by sovereign credit default swap spreads.

### Macroeconomic implications
- FX market stress can transmit into tighter financial conditions, exacerbating downside tail risks to real GDP growth and threatening macrofinancial stability (references to Adrian, Boyarchenko, and Giannone 2019; October 2024 Global Financial Stability Report).
- Structural vulnerabilities—high dealer concentration, growing role of NBFIs, intensified FX hedging and funding pressures—and operational disruptions in FX market infrastructure can amplify stress and impair liquidity, spilling over into debt and equity markets.

### Key empirical and illustrative figures and observations
- Shaded areas in figures represent 90 percent confidence intervals.
- CLS’s netting process typically reduces funding requirements by approximately 96 percent (CLS Group 2025).
- Box example: On April 2, 2025, the United States announced increased tariff rates on imports; the announcement triggered a spike in financial uncertainty measures (Chicago Board Options Exchange Volatility Index) and a depreciation of the broad US dollar index by about 2 percent on impact.
- Spot dollar purchases by nonresident investors rose by about $265 billion on a net basis between January 1 and April 1, 2025.
- Cumulative net spot purchases remained broadly stable as of end-May 2025.
- Cross-country and sectoral patterns around the April tariff announcement:
  - Canada was a net buyer of spot dollars from November 2024 through mid-April 2025 but shifted to net selling thereafter.
  - Major euro area countries increased net spot dollar sales after April 2, 2025.
  - Non-US nonbank financial institutions have been active buyers of US dollars in the spot market around the episode, in contrast with the COVID-19 turmoil in March 2020 when banks dominated FX trading.
- Non-US nonbank investors’ FX swap activity against the US dollar increased notably after the April 2 tariff announcement; hedging demand (selling US dollar forward contracts) has been stronger and more persistent versus the COVID-19 episode.
- Cumulative swap flows presented do not account for maturity and refinancing activities or mark-to-market valuation changes; swap flow signs indicate sectoral net buying/selling patterns where a positive bar in panel 4 indicates the institutional sector is a net buyer of US dollars in the forward leg and a net seller in the near leg; panels exclude US dollar swap flows with tenors less than 35 days in panel 4.

### Policy conclusions and recommendations
- Strengthen surveillance to monitor systemic risk arising from FX market stress:
  - Implement a more structured surveillance approach to capture FX market vulnerabilities and cross-border spillovers.
  - Enhance FX liquidity stress tests and systemwide stress tests to incorporate scenarios with heightened volatility, wider bid-ask spreads, and cross-currency bases.
  - Monitor and mitigate rollover and liquidity risks from short-tenor FX swap positions.
  - Use scenario analysis that includes severe and persistent technical failures, cyberattacks, physical disasters, and defaults by major FX dealers.
  - Close data gaps by improving reporting and data sharing on bilateral exposures, settlement practices, intraday trading, and counterparty concentrations, especially for transactions outside centralized infrastructures like PvP systems.
- Ensure adequate capital and liquidity buffers at financial institutions, supported by robust crisis management:
  - Require institutions with dominant/systemic FX roles to maintain adequate hedges and capital and liquidity buffers.
  - Strengthen access to intraday central bank liquidity and credit facilities, including for NBFIs, while limiting moral hazard.
  - Economies reliant on external financing should maintain sufficient international reserve buffers.
  - Strengthen and expand central bank swap line networks; use the IMF’s lending toolkit as part of the global financial safety net for countries facing FX liquidity pressures.
  - Consider FX intervention and macroprudential and capital flow management measures calibrated to country-specific conditions when external shocks cause undesirable macroeconomic fluctuations in the presence of significant FX mismatches.
- Manage operational and settlement risk:
  - Strengthen operational resilience of financial market infrastructures per the Principles for Financial Market Infrastructures (BIS-CPSS-IOSCO 2012): identify operational risk sources, implement robust systems, comprehensive business continuity planning, cyber resilience frameworks, and regular contingency testing.
  - Coordinate responses among central banks and oversight agencies given cross-jurisdictional interconnectedness.
  - Reduce FX settlement risks through wider adoption of PvP arrangements; interim dealer-bank arrangements include “pre-settlement netting” and “on-us” settlement.
  - Implement strong anti-money laundering/combating the financing of terrorism measures to reduce settlement uncertainty.
  - Explore policy initiatives leveraging digital technologies—linking faster payment systems or developing cross-border central bank digital currency—while recognizing potential risks if not properly designed and regulated.
- Encourage migration toward well-designed financial platforms to lower transaction costs and volatility, reduce counterparty and settlement risks, and mitigate dealer constraints in over-the-counter markets.

*International Monetary Fund, Chapter 2, “Risk and Resilience in the Global Foreign Exchange Market” (October 2025).*

### Box 2.1 (continued)

### Box 2.1 (continued)

### FX settlement risk: definition and current relevance
- FX settlement risk (Herstatt risk) arises when one party delivers the currency it sold but fails to receive the currency it bought, creating liquidity pressures, credit losses, and potential systemic disruptions.
- Settlement risk remains a concern for emerging market and developing economies that often:
  - lack access to simultaneous settlement mechanisms like payment-versus-payment (PvP) systems;
  - rely more on correspondent banking relationships that introduce additional counterparty exposure and operational complexity; and
  - operate within payment and legal frameworks that may not align with global FX settlement arrangements.
- About 25 percent of the deliverable turnover of currencies is without risk mitigation mechanisms (Glowka and Nilsson 2022).

### Key historical episodes and mitigation frameworks
- Bankhaus Herstatt failure on June 26, 1974, when Deutsche mark payments were received but corresponding US dollar leg was not delivered; event prompted establishment of the Basel Committee on Banking Supervision.
- 2008: Kreditanstalt für Wiederaufbau transferred €300 million (equivalent to $426 million at the time) to Lehman Brothers on the morning of Lehman’s bankruptcy filing but did not receive the corresponding payment, resulting in a unilateral loss.
- March 2020: Barclays suffered a $129 million FX loss when UAE Exchange failed to deliver amid COVID-19-related market stress.
- Major mitigation advances and standards:
  - Establishment of CLS in 2002 as a multicurrency PvP system to ensure simultaneous settlement of both legs of FX transactions.
  - Standard-setting guidance by the Basel Committee on Banking Supervision, the Committee on Payments and Market Infrastructures–IOSCO, and the Global Foreign Exchange Committee.
  - Publication of the FX Global Code in 2017, setting principles for market conduct, risk management, and settlement practices.
  - Wider adoption of real-time gross settlement systems and improved legal frameworks to support netting and cross-border enforceability.

### PvP systems and effects on currency risk premiums
- Mechanisms to mitigate FX settlement risk:
  - Presettlement netting: reduces amounts exchanged by offsetting obligations.
  - Simultaneous settlement mechanisms: PvP or on-us settlement. PvP ensures each currency leg settles only if the other does; on-us settlement processes both legs in the same institution (protection assured only if settlement is simultaneous or within preauthorized credit lines—“on-us with loss protection”).
- Empirical approaches used:
  - Natural experiment: Hungary’s accession to CLS on November 16, 2015, using a difference-in-difference analysis against the Czech koruna and Polish zloty as controls.
  - Panel regression: January 2000 to May 2025 covering 26 currencies (including 16 settled through CLS and 4 with other PvP arrangements—Brazil, India, Malaysia, Thailand).

Findings
- Hungary (forint) accession to CLS:
  - Average daily excess exchange rate returns declined by about 11 basis points (bps) in the one-month window before vs after CLS participation.
  - Volatility declined by 3 bps in the same window.
  - Average excess returns over the month before CLS accession were about 28 bps, suggesting CLS participation eliminated this excess return.
- Panel analysis (Jan 2000–May 2025):
  - CLS participation is associated with a significant decline of 34 bps in excess FX returns, on average.
  - CLS participation is associated with a 3 bps decline in volatility, on average.
- Interpretation: PvP systems like CLS reduce settlement uncertainty and counterparty exposure, lowering settlement risk premiums and volatility.

### Operational disruptions in FX markets: case study of interdealer platform outages
- Context:
  - Core FX interdealer liquidity and price discovery remain concentrated in a few venues, notably Electronic Broking Services (EBS) and London Stock Exchange Group’s FX Matching.
  - EBS experienced an outage in 2023; FX Matching experienced disruption in 2015. Both outages occurred when London and New York sessions overlapped (period of high liquidity).
- Analysis approach:
  - Compare currencies primarily traded on the affected platform versus those not; analyze spot and forward bid-ask spreads, realized illiquidity, price dispersion, and traded volumes over the outage days and 90 days before and after.
  - Bid-ask spreads sampled at 30-minute intervals; other measures constructed at daily frequency. Sample currencies: euro, Japanese yen, British pound, Swiss franc, Canadian dollar, Australian dollar, New Zealand dollar, Swedish krona, Norwegian krone, all trading against the US dollar.
- Effects on affected currencies:
  - Spot and swap market bid-ask spreads increased during outages.
  - Average decline in spot market volumes on outage days: $2.8 billion across affected currencies (all dollar volumes adjusted for inflation to December 2024 US dollars).
  - Estimated relation: a $1 billion decrease in trading volume is associated with a 0.3 basis point widening of spot market bid-ask spreads (instrumented by outages).
  - Price impact: trading volume required to move daily FX returns by one standard deviation (about 3.4 percent) declined from $19.2 billion to $18.6 billion.
  - Dispersion of transaction prices across counterparties increased by about 0.2 standard deviations (about 11 percent), indicating some counterparties accepted less favorable terms.
- Aggregate (spillover) effects across all currencies:
  - Transaction costs rose in spot and swap markets even for currencies mainly traded on venues that remained operational, suggesting liquidity migration and congestion.
  - Average bid-ask spreads against the US dollar widened from 3 to 4 basis points in the spot market and from 4 to 5 basis points in the swap market during outages.
- Interpretation: Even relatively short-lived outages of interdealer trading platforms can materially impair FX market liquidity and raise transaction costs, with both direct effects on affected currencies and spillovers to other venues.

### Key quantitative points
- 25 percent: share of deliverable currency turnover without risk mitigation mechanisms.
- June 26, 1974: Bankhaus Herstatt failure date.
- 2002: establishment of CLS.
- November 16, 2015: Hungary’s forint joined CLS.
- 2000–25 / January 2000 to May 2025: panel sample period for PvP analysis.
- Hungary CLS effect: excess returns down by about 11 basis points; volatility down by 3 basis points.
- Pre-CLS average excess returns for forint: about 28 basis points.
- Panel-average CLS effect: excess FX returns down by 34 basis points; volatility down by 3 basis points.
- 2008 incident: €300 million (equivalent to $426 million at the time) transferred to Lehman Brothers without receiving corresponding payment.
- March 2020 incident: Barclays lost $129 million.
- Outage-related volume decline (affected currencies): $2.8 billion (average, outage days; December 2024 US dollars).
- Volume–spread relationship estimate: $1 billion decrease in trading volume → 0.3 basis point widening of bid-ask spreads.
- Trading volume to move daily returns by one SD (~3.4 percent): fell from $19.2 billion to $18.6 billion.
- Price dispersion increase during outages: ~0.2 standard deviations (~11 percent).
- Spot bid-ask spreads widened from 3 to 4 basis points on average; swap bid-ask spreads widened from 4 to 5 basis points on average during outages.

*International Monetary Fund | October 2025*

### CHAPTER 2 RISk ANd RESILIENCE IN ThE GLOBAL FOREIGN ExChANGE MARkET

### CHAPTER 2 RISk ANd RESILIENCE IN ThE GLOBAL FOREIGN ExChANGE MARkET

### Selected references cited in the chapter
- Du, Wenxin, and Jesse Schreger. 2022. “CIP Deviations, the Dollar, and Frictions in International Capital Markets.” In Handbook of International Economics, vol. 6: 147–97.
- Du, Wenxin, Alexander Tepper, and Adrien Verdelhan. 2018. “Deviations from Covered Interest Rate Parity.” Journal of Finance 73 (3): 915–57.
- Duffie, Darrell. 2023. “Resilience Redux in the US Treasury Market.” Paper presented at the Federal Reserve Bank of Kansas City’s Jackson Hole Symposium, Jackson Hole, Wyoming, August 25.
- Eguren-Martin, Fernando, Mattias O. Busch, and Dennis Reinhardt. 2024. “Global Banks and Synthetic Funding: The Benefits of Foreign Relatives.” Journal of Money, Credit, and Banking 56 (1): 115–52.
- Financial Stability Board (FSB). 2022. “US Dollar Funding and Emerging Market Economy Vulnerabilities.” Report to the Group of Twenty, FSB, Basel, Switzerland. https://www.fsb.org/2022/04/us-dollar-funding-and-emerging-market-economy-vulnerabilities.
- Financial Stability Board (FSB). 2025. “Leverage in Nonbank Financial Intermediation: Final Report.” Policy document, FSB, Basel, Switzerland. https://www.fsb.org/2025/07/leverage-in-nonbank-financial-intermediation-final-report.
- Gabaix, Xavier, and Matteo Maggiori. 2015. “International Liquidity and Exchange Rate Dynamics.” Quarterly Journal of Economics 130 (3): 1369–420.
- Glowka, Marc, and Thomas Nilsson. 2022. “FX Settlement Risk: An Unsettled Issue.” BIS Quarterly Review (December): 75–81.
- Greenwood, Robin, Samuel Hanson, Jeremy C. Stein, and Adi Sunderam. 2023. “A Quantity-Driven Theory of Term Premiums and Exchange Rates.” Quarterly Journal of Economics 138 (4): 2327–89.
- Grothe, Magdalena, Peter McQuade, Martino Ricci, and Luca Tondo. 2025. “Recent Patterns in Global Risk Behaviour in Financial Markets.” Column, voxEU, Center for Economic Policy Research, August 12, 2025.
- He, Zhiguo, Bryan Kelly, and Asaf Manela. 2017. “Intermediary Asset Pricing: New Evidence from Many Asset Classes.” Journal of Financial Economics 126 (1): 1–35.
- Hofmann, Boris, Ilhyock Shim, and Hyun Song Shin. 2020. “Bond Risk Premiums and the Exchange Rate.” Journal of Money, Credit and Banking 52 (S2): 497–520.
- Hossain, Ashrafee T., Abdullah-Al Masum, and Samir Saadi. 2024. “The Impact of Geopolitical Risks on Foreign Exchange Markets: Evidence from the Russia–Ukraine War.” Finance Research Letters 59 (January): 104750.
- Huang, Wenqian, Angelo Ranaldo, Andreas Schrimpf, and Fabricius Somogyi. 2025. “Constrained Liquidity Provision in Currency Markets.” Journal of Financial Economics 167: 104028.
- Hui, Hon Chung. 2021. “The Long-Run Effects of Geopolitical Risk on Foreign Exchange Markets: Evidence from Some ASEAN Countries.” International Journal of Emerging Markets 17 (6): 1543–64.
- International Monetary Fund (IMF). 2016. “Guidance Note on the Assessment of Reserve Adequacy and Related Considerations.” IMF Policy Paper, Washington, DC, June 3, 2016. https://www.imf.org/en/Publications/Policy-Papers/Issues/2016/12/31/Guidance-Note-on-the-Assessment-of-Reserve-Adequacy-and-Related-Considerations-PP5046.
- International Monetary Fund (IMF). 2023. “Integrated Policy Framework—Principles for the Use of Foreign Exchange Intervention.” IMF Policy Paper 23/061, Washington, DC.
- International Monetary Fund (IMF). 2024a. “Central Bank Digital Currency: Progress and Further Considerations.” IMF Policy Paper 24/052, Washington, DC.
- International Monetary Fund (IMF). 2024b. “G-20 Note on Financial Platforms: What Are They and What Are Their Macro-financial Implications?” IMF, Washington DC.
- Jiang, Zhengyang, Arvind Krishnamurthy, Hanno N. Lustig, Robert Richmond, and Chenzi Xu. 2025. “Dollar Upheaval: This Time Is Different.” Working Paper, Northwestern University.
- Khiaonarong, Tanai, Harry Leinonen, and Ryan Rizaldy. 2021. “Operational Resilience in Digital Payments: Experiences and Issues.” IMF Working Paper 21/288, International Monetary Fund, Washington, DC.
- King, Michael R., Carol Osler, and Dagfinn Rime. 2012. “Foreign Exchange Market Structure, Players, and Evolution.” In Handbook of Exchange Rates, edited by Jessica James, Ian W. Marsh, and Lucio Sarno, 3–44. Hoboken, NJ: Wiley.
- Kloks, Peteris, Edouard Mattille, and Angelo Ranaldo. 2023. “Foreign Exchange Swap Liquidity.” Research Paper 23-22, Swiss Finance Institute, Zurich, Switzerland.
- Kloks, Peteris, Edouard Mattille, and Angelo Ranaldo. 2024. “Hunting for Dollars.” Research Paper 24-52, Swiss Finance Institute, Zurich, Switzerland.
- Kloks, Peteris, Patrick McGuire, Angelo Ranaldo, and Vladyslav Sushko. 2023. “Bank Positions in FX Swaps: Insights from CLS.” BIS Quarterly Review (September): 17–31.
- Kubitza, Christian, Jean-David Sigaux, and Quentin Vandeweyer. 2025. “The Implications of CIP Deviations for International Capital Flows.” Working Paper 3017, European Central Bank, Frankfurt am Main, Germany.
- Kumar, Abhishek, Sushanta Mallick, Madhusudan Mohanty, and Fabrizio Zampolli. 2023. “Market Volatility, Monetary Policy and the Term Premium.” Oxford Bulletin of Economics and Statistics 85 (1): 208–37.
- Lambert, Colin. 2023. “What Happens When a Primary FX Venue Goes Offline?” The Full FX, October 4, 2023.
- Liao, Gordon Y., and Tony Zhang. 2025. “The Hedging Channel of Exchange Rate Determination.” Review of Financial Studies 38 (1): 1–38.
- Mackenzie Smith, Robert. 2015. “Thomson Reuters Matching Suffers Outage.” FX Markets, July 1, 2015. https://www.fx-markets.com/technology/trading-systems/2415705/thomson-reuters-matching-suffers-outage
- Nenova, Tsvetelina, Andreas Schrimpf, and Hyun Song Shin. 2025. “Global Portfolio Investments and FX Derivatives.” Working Paper 1273, Bank for International Settlements, Basel, Switzerland.
- Ranaldo, Angelo, and Paolo Santucci de Magistris. 2022. “Liquidity in the Global Currency Market.” Journal of Financial Economics 146 (3): 859–83.
- Ranaldo, Angelo, and Fabricius Somogyi. 2021. “Asymmetric Information Risk in FX Markets.” Journal of Financial Economics 140 (2): 391–411.
- Schrimpf, Andreas, and Vladyslav Sushko. 2019. “FX Trade Execution: Complex and Highly Fragmented.” BIS Quarterly Review (December): 39–51.

*Source: CHAPTER 2 RISk ANd RESILIENCE IN ThE GLOBAL FOREIGN ExChANGE MARkET.*

---


_Source: https://www.imf.org/-/media/files/publications/gfsr/2025/october/english/ch2.pdf_