Skip to main content
Currency Pairs in Depth

The Japanese Yen as a Safe Haven

Pomegra Learn

Why Does the Yen Rally More During Carry-Trade Unwinding Than Other Safe-Haven Currencies?

The Japanese Yen holds a paradoxical position in global forex markets: it is the safe-haven currency most likely to rally during equity-market crashes and geopolitical crises, despite Japan being the world's third-largest economy (by nominal GDP) with persistent fiscal deficits, a shrinking population, and declining economic growth rates. This seeming contradiction resolves through understanding the yen's unique role as the funding currency for leveraged carry trades worldwide. For two decades, Japanese interest rates have remained at or near zero—the Bank of Japan (BoJ) has maintained rates at 0–0.10% since the mid-1990s, explicitly unwilling to tighten policy. This near-zero funding cost created an irresistible arbitrage opportunity: investors could borrow unlimited yen at 0.05%, convert to USD or AUD or NZD at 2–5%, and pocket the interest-rate spread. At any time, an estimated $500 billion to $1 trillion in gross leveraged yen carry trades exist globally—borrowed yen funding longs in higher-yielding currencies. When market stress spikes and margin calls trigger, carry traders face a forced unwinding: they must buy yen to close their borrowing positions, creating concentrated demand for the yen regardless of Japan's economic fundamentals. A USD/JPY crash from 110 to 100 during a crisis (9% yen appreciation in 2–3 weeks) occurs not because Japan became safer relative to the U.S., but because carry traders are forced to cover. This mechanism makes the japanese yen safe haven behavior the most predictable and profitable safe-haven trade during acute market stress—but it also makes the yen vulnerable to reversal when risk appetite returns.

Quick definition: The Japanese Yen serves as a safe-haven currency during crises primarily through carry-trade unwind mechanics: when investors face margin calls and reduce leverage, they must buy yen to repay borrowed positions, creating concentrated appreciation pressure regardless of Japan's economic fundamentals or BoJ policy stance.

Key takeaways​

  • Carry-trade unwind mechanism: An estimated $500B–$1T in leveraged yen carry trades exist globally; when VIX >25 or credit spreads widen >100 bps, forced unwinding triggers concentrated yen buying
  • Zero-rate funding advantage: The BoJ's 20+ year maintenance of near-zero rates created a permanent arbitrage: borrow yen at 0.05%, lend USD/AUD/NZD at 2–5%, capturing 200–500 basis points annually
  • Non-economic safe-haven property: Yen strength during crises is mechanical (carry-trade unwind), not fundamental (economic safety); yen rallies even when Japan's economy deteriorates faster than other nations
  • Correlation with carry-trade positioning: JPY strength correlates more tightly with hedge-fund leverage and carry-trade positioning than with Japan's economic data; when leverage falls 20%, JPY typically strengthens 3–5%
  • BoJ tightening risk: The BoJ's possible future rate hikes would compress the carry-trade advantage, potentially weakening the yen's crisis-appreciation characteristics and destabilizing trillions in carry positions

The Carry-Trade Structural Foundation​

The japanese yen safe haven status rests on a deceptively simple fact: Japanese interest rates have remained near zero for 25+ years, while virtually all other developed economies have maintained 1–5% rates. In 2007–2008, a carry trader could borrow yen at 0.50%, convert to Australian dollars at 6–7% rates, and pocket 5.5–6.5% annually—with minimal funding costs. As late as 2023, the BoJ was still holding rates at 0.10% while the Fed offered 5.25–5.50% and the RBNZ offered 5.50%, creating 500-basis-point spreads.

This rate advantage is not accidental. The BoJ deliberately maintained zero rates as a monetary policy framework designed to stimulate the Japanese economy and combat persistent deflation. A 25-year experiment in ultra-loose monetary policy created the unintended consequence of making the yen the world's favorite funding currency. When the BoJ finally began considering rate hikes in 2023 (raising rates from -0.10% to 0.25%, still near-zero by global standards), yen carry traders became nervous: even a 0.50% rate hike would compress their 500-basis-point advantage to 450 basis points, potentially worth $5–10 billion in carry returns per percentage point of moves across all leveraged positions.

The mechanics of yen carry trades are straightforward: (1) A hedge fund borrows 10 billion yen at 0.10% (annual interest: 10 million yen = ~$75,000 in USD terms). (2) It converts the yen to 100 million USD (at 1 yen = 100 USD), depositing USD in a 5.0% money-market fund (annual interest: 5 million USD). (3) Net gain: 5 million USD – 75,000 USD = ~4.9 million USD annually, a 4.9% return on the 100 million USD deployed. (4) With 10:1 leverage, the fund earns 49% annual returns on its equity capital (4.9 million USD / 10 million USD initial capital). When leverage is high, these returns compound rapidly; a $1 billion hedge fund with 10:1 leverage earns $490 million annually from the carry trade alone—no alpha, no risk-taking, pure mechanical arbitrage.

This machine runs smoothly until margin calls arrive. When a crisis spikes volatility (VIX >25), brokers immediately increase margin requirements on leveraged positions. A hedge fund with 10:1 leverage might see margin requirements increase from 10% to 15%, requiring it to immediately raise $5 billion additional collateral. Rather than raise new capital (impossible during crises), funds must sell positions and de-leverage. They sell the funded currency (USD, AUD, NZD, etc.) and buy yen to close the borrowed positions.

The Carry-Trade Unwind Process and Forced Buying​

When multiple hedge funds unwind simultaneously (a typical crisis scenario), the forced yen buying becomes mechanical and predictable. Consider a simplified scenario: 50 hedge funds, each running $1 billion in yen carry trades with 10:1 leverage, collectively holding $500 billion in borrowed yen. A sudden margin call requiring 30% position reduction forces them to collectively buy 150 billion yen to close positions. This 150 billion yen buying hits the FX market concentrated in a 3–10 day period, creating a vertical move in USD/JPY.

In practice, the unwind happens in stages:

  1. Initial shock (Day 1–2): VIX spikes above 25; early-warning traders begin unwinding, creating 2–3% USD/JPY declines in the first 2 days.
  2. Broker margin calls (Day 3–5): Brokers increase margin requirements; mid-stage carry traders receive margin calls and begin unwinding, creating 4–6% cumulative declines.
  3. Cascading liquidations (Day 6–15): Carry traders face losses from the unwind itself (if they're forced to buy yen when it's spiking, they lock in losses), triggering additional forced liquidations of other positions to raise cash. This cascade can produce 10–15% USD/JPY declines in 10 days.

The March 2020 COVID crisis provides a textbook example. USD/JPY was trading at 112 on February 28, 2020. On March 2, news of aggressive corporate bankruptcies and pandemic spread triggered a VIX spike from 17 to 32. By March 9, USD/JPY had fallen to 105 (6.3% move in 7 days). By March 16, as corporate bankruptcies accelerated and equity markets crashed 15%, USD/JPY reached 101 (9.8% move in 14 days). The entire cascade—from initial shock to maximum dislocation—occurred in 2 weeks, with the most violent moves in the first 5 days.

During this unwind, yen carry traders experienced catastrophic losses: those with 10:1 leverage watched their capital erode by 50–70% in one week. A $1 billion fund with 10:1 leverage and $500 million in carry losses (2–3% moves on $10 billion notional positions) faced capital losses of 500 million / 1 billion = 50% in days. Realizing these losses accelerated unwinding as funds forced-liquidated positions to raise capital and shore up investor redemptions.

Measuring Carry-Trade Positioning and Unwind Risk​

Traders can estimate carry-trade positioning by monitoring several metrics:

Yen Basis Swaps: The cost to swap yen against USD for specific periods reflects carry-trade funding demand. When basis swaps widen (moving from 0 to +50 basis points), it signals carry traders are funding heavily. When they compress (moving from +50 back to -20), unwinding is underway. A 70-basis-point swing in yen basis swaps over 2–3 weeks signals imminent unwind risk.

Yen Funding Rates in Offshore Markets: Tokyo money markets publish overnight and term yen funding rates (TONAR/TIBOR). During normal periods, these trade flat to slightly positive (reflecting the carry-trade funding cost). During crises, these rates spike sharply positive as carry traders scramble to buy yen, and brokers widen bid-ask spreads due to the buying pressure. A 50-basis-point spike in TIBOR 1M rates signals acute unwind.

Gross Yen Borrowing in the Tokyo Money Market: The BoJ publishes weekly data on yen borrowing by non-resident entities. When this metric rises for 2+ weeks, carry-trade positioning is growing. When it falls 10%+ in a single week, unwind is imminent.

Implied Volatility in JPY Options: When 1-month implied volatility (IV) on USD/JPY options spikes from 8–10% to 15–20%, option markets are pricing in expected yen strength. A spike above 18% suggests unwind risks are elevated.

Cross-Asset Correlation: When the correlation between USD/JPY returns and equity-market returns turns sharply negative (falling below -0.50), carry-trade unwind is active. During calm periods, this correlation stays near -0.10 to -0.30 (modest negative); during crises, it crashes toward -0.70 to -0.90.

All four metrics rarely spike simultaneously; typically, one or two precede a full unwind by 1–2 weeks, providing traders with a 5–10 day warning window to adjust positioning.

The August 2015 China Shock and August 2019 Yen Flash-Crash​

The August 2015 Chinese devaluation shock provides a clear example of carry-trade unwind mechanics. On August 11, 2015, Chinese authorities unexpectedly devalued the yuan by 3%, signaling potential currency warfare and emerging-market distress. Global equity markets collapsed: the S&P 500 fell 8% in 4 days. Within 24 hours of the devaluation announcement, carry traders began unwinding.

USD/JPY fell from 124.0 on August 11 to 119.0 by August 14 (4% decline in 3 days), then continued to 116.5 by August 21 (6.5% decline in 10 days). During this unwind, yen funding rates (TIBOR 1M) spiked from 0.20% to 0.80%, a 60-basis-point increase signaling acute funding stress. Yen basis swaps widened 80+ basis points as carry traders scrambled to find yen funding to close positions. The unwind lasted 2–3 weeks; by early September, positioning had stabilized and USD/JPY recovered to 119.

The August 2019 "flash crash" demonstrated similar mechanics but with an additional currency-intervention element. On August 6, 2019, after the Fed cut rates (sparking recession fears), carry traders began unwinding. Within 24 hours, USD/JPY crashed from 108.5 to 104.5 (3.8% in hours). During the crash, yen funding completely froze: brokers stopped offering yen in FX swap markets because the demand was overwhelming. The crash was so violent that the BoJ explicitly intervened in the FX market, selling yen and buying USD/currencies to provide liquidity and slow the descent. The BoJ's intervention (rare and dramatic) demonstrated how severe carry-trade unwinding can become without central-bank relief.

BoJ Policy and the Future of Yen Carry Trades​

A critical structural risk to yen carry trades is the BoJ's possible future rate normalization. If the BoJ were to raise rates to 1.0%–1.5% (still historically low but double current levels), the carry-trade advantage would compress sharply: borrowing yen at 1.0% and lending USD at 5.0% yields 400 basis points, not 500. At scale, a 100-basis-point compression costs carry traders $10–20 billion annually in foregone returns—enough to trigger voluntary position reductions and potential market instability.

The BoJ's March 2023 hawkish pivot (signaling possible future rate hikes) immediately triggered USD/JPY weakness: USD/JPY fell from 130 to 126 in 4 weeks as carry traders began reducing positions proactively. The fear was that the BoJ would eventually hike rates to 0.50%, then 0.75%, then 1.0%, progressively making yen carry trades uneconomical. When the BoJ did hike to 0.25% in July 2023, further yen strength followed, with USD/JPY trading as low as 124.

However, the BoJ faces a dilemma: hiking rates too quickly risks triggering a chaotic unwind that destabilizes global financial markets. The BoJ knows that $500B–$1T in yen carry trades exist globally; hiking rates abruptly would force simultaneous unwinding, creating a potential JPY appreciation spiral (as yen strengthens, margin calls accelerate, forcing more unwinding, creating further yen strength—a vicious feedback loop). The BoJ has therefore chosen an extremely gradual hiking path, raising rates only 0.25% in a year despite inflation running 2%+. This glacially slow normalization is intentionally designed to allow carry traders to gradually adjust positions rather than forced-liquidate.

This creates an asymmetry: the BoJ wants to gradually normalize rates but is constrained by carry-trade stability concerns. Carry traders, facing ever-narrowing spreads, are proactively unwinding voluntarily. The result is a potential scenario where USD/JPY gradually trends weaker over 2–3 years (from 145 to 100+) not due to crisis-driven unwind, but due to structural carry-trade reduction in anticipation of future BoJ hikes.

The Yen's Hybrid Role: Carry Currency Vs. Safe Haven​

The japanese yen occupies a unique hybrid position: it is simultaneously (1) a carry-trade funding currency (creating weakness during calm periods as investors fund carry trades, pushing USD/JPY higher), and (2) a safe-haven currency (creating strength during crises as carry traders unwind, pushing USD/JPY lower). This dual role creates an unusual volatility pattern:

  • During calm periods (VIX <15): Carry traders are accumulating leverage; USD/JPY drifts higher (200–300 basis points annually as the spread widens). Yen is weak.
  • During moderate volatility (VIX 15–20): Positioning stabilizes; USD/JPY consolidates in narrow ranges. Yen is neutral.
  • During stress periods (VIX >25): Carry traders unwind; USD/JPY crashes 5–15% in 2–4 weeks. Yen is strong and appreciating.

This pattern has held consistently. From 2009–2019 (a 10-year calm period), USD/JPY drifted from 90 to 110, a 22% yen weakness over 10 years. From 2019–2020 (a crisis period), USD/JPY crashed from 111 to 101, a 9% yen strength in 6 months.

The implication: traders should approach the yen differently depending on regime. During calm periods, shorting yen (long USD/JPY carries) generates steady returns until the inevitable crisis. During stress periods, being long yen (short USD/JPY) is the profitable position. This creates a "barbell strategy": maintain small long yen positions continuously (5% portfolio allocation) and scale to 15–20% during warning signs of carry-trade unwind.

Real-World Examples: Crisis-Driven Yen Appreciation​

2008 Lehman Collapse (September–December): USD/JPY fell from 104 to 90 in 12 weeks (13% yen appreciation), the most severe carry-trade unwind of the 2000s. Yen basis swaps widened 200+ basis points; yen funding rates spiked to 3.0%+. The BoJ eventually had to intervene in the FX market to provide yen liquidity. Leverage in hedge funds compressed 30–40% as positions unwound.

2011 European Debt Crisis (August–September): When Greece's sovereign debt concerns spread to Italy and Spain, USD/JPY fell from 77 to 72 (6.5% yen appreciation in 4 weeks). Carry traders unwound EUR/JPY positions (which had been large shorts in yen, creating long carry trades). The unwind was less severe than 2008 because the crisis was more localized.

2015 China Devaluation (August 11–21): USD/JPY fell from 124 to 116.5 (6.1% yen appreciation in 10 days). Carry-trade unwinding was acute and fast. Within 3 weeks, unwinding was complete and USD/JPY recovered to 119 as the crisis stabilized.

2020 COVID-19 Pandemic (February 28–March 16): USD/JPY fell from 112 to 101 (9.8% yen appreciation in 14 days). The unwind was severe due to pandemic shock triggering margin calls across all leveraged funds globally, not just carry traders. The BoJ had to provide unlimited yen liquidity through repo facilities and FX swap lines.

2022 U.K. Gilt Market Disruption (September 23–30): After the UK announced aggressive fiscal stimulus, gilt yields spiked. GBP/JPY fell from 180 to 163 (9.4% yen appreciation in 7 days), as carry traders unwound GBP/JPY shorts (longs in yen). USD/JPY fell more modestly (from 145 to 141, a 2.8% decline) because the crisis was sterling-specific.

Why Yen Strength During Crises Doesn't Reflect Economic Reality​

A critical insight: yen appreciation during crises is entirely mechanical (carry-trade unwind) and has no correlation with Japan's economic health. In March 2020, Japan's economy contracted 6.3% (more sharply than the U.S. at -3.4%), yet JPY strengthened dramatically. In August 2015, Japan's economy was stagnating with 0.3% growth, yet JPY appreciated 6% against the dollar during the China shock.

This disconnect reveals that yen safe-haven status is an accident of policy, not fundamental strength. The yen is a safe haven because:

  1. The BoJ maintains zero rates (structural feature, not by choice)
  2. Carry traders fund in yen at scale ($500B–$1T)
  3. Crises trigger forced unwinding (mechanical feature)
  4. Yen strength occurs, whether or not Japan is economically strong

If the BoJ were to hike rates to 3–4% tomorrow (matching developed-economy peers), the yen would become a risk asset, not a safe haven. Carry traders would stop funding in yen; instead, they'd borrow yen at 4% to lend in other markets. The safe-haven role would transfer to whichever currency offered the lowest rates and most stable economy. This is why BoJ policy is so critical to understanding yen dynamics: the BoJ's zero-rate commitment is the entire foundation of yen safe-haven behavior.

Common Mistakes in Yen Trading​

Mistake 1: Assuming yen strength means Japan is economically strong. Yen strength during crises reflects carry-trade mechanics, not economic fundamentals. Japan could be in recession while yen is appreciating, because the forcing function is carry-trade unwind, not growth differentials.

Mistake 2: Trading yen fundamental correlations during crises. The yen's correlation with equity markets shifts from -0.10 to -0.70 during crises. Traders using historical correlations get blindsided. During crises, use mechanical unwind indicators (yen funding costs, basis swaps, positioning data) rather than fundamental factors.

Mistake 3: Shorting yen with tight stops during calm periods. Shorting yen carry (long USD/JPY) seems attractive for capturing 2–3% annual carry returns. However, crises can produce 10–15% reversals in weeks. Without hard stops, carry shorts can result in catastrophic losses. Use 4–5% stops.

Mistake 4: Assuming BoJ intervention will always prevent acute yen strength. The BoJ intervenes in yen, but only rarely and with limited effectiveness. In August 2019, the BoJ's intervention slowed the unwind but didn't stop it. Carry traders' forced unwinding overwhelms BoJ selling in most cases.

Mistake 5: Neglecting the BoJ's rate-hike trajectory. If the BoJ normalizes rates to 1.0%+ over 3–5 years, carry-trade advantages shrink, and long-term yen depreciation accelerates. Traders shorting yen for "free carry" should monitor BoJ rate expectations monthly.

FAQ​

Is JPY stronger than USD during crises?​

Not always. During financial systemic crises (Lehman 2008), both USD and JPY appreciate together. During carry-trade unwinding specifically, JPY appreciates more than USD (USD/JPY falls more than USD indices fall). The March 2020 COVID crisis showed both: USD strength against risk currencies (EUR, AUD, BRL) and additional JPY strength against USD (USD/JPY fell 9% while USD indices rose modestly).

What is the maximum carry-trade position size globally?​

Estimates range from $500 billion to $1.5 trillion in gross notional yen carry trades. The exact size is unknowable because much of it is off-the-books (leverage through non-standardized swaps, unregulated brokers, sovereign wealth fund positions). The 2019 and 2020 crises suggested the number is closer to $1 trillion based on unwinding speed and funding-rate spikes.

If the BoJ raises rates to 2.0%, would the yen still be a safe haven?​

Probably not. At 2.0% BoJ rates vs. 5% Fed rates, the carry trade becomes much less attractive ($1 billion carry trade captures $30 million annually instead of $50 million). Carry traders would gradually reduce positions. Over 5–10 years, yen would likely depreciate as carry funding dries up. The yen's safe-haven role depends on zero-rate maintenance.

How do I differentiate between carry-trade unwind (good for long yen) and fundamental yen weakness (bad for long yen)?​

Monitor the yen basis swaps and TIBOR funding rates. If they spike (basis swaps >100 basis points, TIBOR up 50+ basis points), carry-trade unwind is active and long yen is profitable. If they remain stable and yen weakens, then it's fundamental yen weakness and shorting yen is correct.

Should I hedge my Japan equity portfolio with yen strength?​

Yes. A 5–10% portfolio allocation to long yen or long USD/JPY calls serves as an excellent hedge for Japan equity holdings. During market crashes, the yen hedge captures 5–15% gains while equities fall 20–30%, reducing net portfolio losses by 5–10 percentage points.

Can the BoJ's negative-rate policy strengthen the yen paradoxically?​

The BoJ used negative rates (-0.10%) from 2016–2023, which theoretically should have weakened the yen (negative returns on yen deposits). However, the negative rates didn't eliminate the carry-trade advantage because even at -0.10%, borrowing yen and lending USD at 5% captures 500+ basis points. The yen remained weak during this period (USD/JPY rallied from 100 to 145), consistent with carry-trade funding.

Summary​

The Japanese Yen serves as a powerful safe-haven currency during equity-market crashes and financial crises, not because Japan is economically strong (it faces aging population, fiscal deficits, and low growth), but because the Bank of Japan's 25-year commitment to zero interest rates created a structural carry-trade funding advantage. Billions of dollars in leveraged yen carry trades exist globally; when margin calls trigger during crises, carry traders are forced to simultaneously buy yen to close positions, creating concentrated appreciation pressure that produces 5–15% moves in 2–4 weeks. Traders can identify imminent yen strength by monitoring yen basis swaps, TIBOR funding rates, and cross-asset correlations; spikes in funding costs signal carry-trade unwind is underway. The key insight: the yen's safe-haven status is entirely structural and dependent on BoJ policy; if the BoJ normalizes rates to developed-economy levels, the yen would transition from safe haven to risk asset, fundamentally disrupting a multi-trillion-dollar carry-trade ecosystem.

Next​

→ The Swiss Franc as a Safe Haven