What Is a Carry Trade? Yen Carry Trade, Examples, and Risks
By Forex & Brokers DeskUpdated Nov 10, 2026

Definition
A carry trade borrows in a low-interest-rate currency or asset to hold a higher-yielding one, earning the rate gap while taking on the exchange rate risk between them.
Key takeaways
- A carry trade earns the difference between a high interest rate and a low one, and it carries the exchange rate risk between the two currencies.
- The expected return is roughly the interest rate differential plus or minus the change in the exchange rate, minus costs. Exchange rate moves usually decide the outcome.
- In theory, interest rate parity should cancel the advantage. Historically, high-yield currencies have often done better than the forward rate implied. Economists call this the forward premium puzzle.
- Carry trades tend to collect small, steady gains and suffer occasional sharp losses, especially when many traders exit at once.
- Central bank decisions in either currency can change the differential. In the funding currency, a rate hike can set off an unwind, as the Bank of Japan's 2024 moves showed.
A carry trade is a strategy in which a trader borrows, or takes a short position, in a currency or asset with a low interest rate and uses that funding to hold a currency or asset that pays a higher rate. The profit target is the gap between the two rates, called the interest rate differential. In forex, the low-rate side is the funding currency and the high-rate side is the target currency. The trade earns that differential for as long as it stays open, and it can make or lose much more than that through exchange rate moves.
The best-known version is the yen carry trade. Japan held interest rates near zero for long stretches, so traders and investors borrowed in yen and put the money into higher-yielding assets abroad. The weakness of the strategy showed up in 2008 and again in August 2024. When the yen rises sharply, months of collected interest can disappear within days.
What Is a Carry Trade?
“Carry” is the income or cost of holding a position over time, apart from any change in its price. A bond’s coupon is carry. So is the storage cost of a barrel of oil, and so is the interest credited or charged on an open currency position overnight. A carry trade is built to collect positive carry: the trader earns more from the asset held than they pay on the funding.
The strategy exists in many markets. Investors borrow at low short-term rates to buy longer-term bonds, and some fund equity positions with cheap credit. Forex is the market where the idea is most explicit, because a currency pair is always long one currency and short another. Each currency has its own policy rate, so every forex position has a built-in rate differential:
- Positive carry: you are long the higher-rate currency and short the lower-rate one. You collect the differential.
- Negative carry: you are long the lower-rate currency and short the higher-rate one. You pay the differential.
A carry trade deliberately takes the positive side and holds it, usually for weeks, months or longer. The trader is betting on two things. The rate differential has to stay favorable, though it does not need to stay fixed. And the exchange rate must not move against the position by more than the interest collected.
Historically popular funding currencies have included the Japanese yen and the Swiss franc, both from economies that kept rates very low for long periods. Target currencies have included the U.S. dollar, the Australian dollar and the Mexican peso. Which currencies fill each role depends on where central banks set rates at the time, so the list changes as monetary policy changes.
The term “carry trade” should not be confused with the older phrase carrying trade. In economic history, a carrying trade was the business of shipping goods between foreign ports on behalf of other countries. The Dutch merchant fleet of the 17th century is the classic example. That meaning concerns transport, not interest rates.
How a Carry Trade Works
The same exposure can be built in several ways. The return drivers are the same in each, but the mechanics, costs and leverage differ.
The cash version: borrow, convert, invest
The textbook form has four steps:
- Borrow in the funding currency at its low interest rate.
- Convert the borrowed money into the target currency at the spot exchange rate.
- Invest it in an interest-bearing asset in the target currency, such as a deposit or a short-term government bill.
- At the end, convert back into the funding currency and repay the loan with interest.
The investor keeps whatever is left. If the exchange rate is unchanged, the profit is close to the rate differential. If the target currency has strengthened, the investor gains extra when converting back. If it has weakened, part or all of the interest income goes toward buying back the funding currency.
The spot forex version: rollover
Retail traders rarely borrow yen from a bank. They open a spot forex position with a broker, and the carry arrives as rollover, also called the swap. Each night that a position stays open past the broker’s cutoff, the broker credits or debits interest based on the rate differential between the two currencies in the pair, minus the broker’s own markup.
Some practical points follow from how rollover is calculated:
- Rollover is charged on the notional value of the position, not on the margin deposited. Leverage multiplies the carry relative to the trader’s capital, and it multiplies price risk by the same factor.
- A broker’s long and short rates for the same pair are not mirror images. Because of the markup, a trader may receive less for the positive side than they would pay for the negative side, and on narrow differentials even the “positive” side can end up as a small debit.
- Spot FX conventionally settles two business days after the trade date, so brokers book several days of rollover on one weekday to cover the weekend. Each broker’s swap schedule sets that day.
- Swap-free accounts remove rollover, so a carry trade is not possible on them.
The derivatives version: forwards, futures, swaps and options
Hedge funds and other professional speculators commonly build carry exposure with derivatives: currency forwards above all, alongside futures, currency swaps and options. A trader who wants long-dollar, short-yen carry can enter a forward contract to buy dollars with yen at a set future date. The forward rate already includes the interest differential, as explained below, so the trader profits if the spot rate at maturity is more favorable than the contracted forward rate. Economically, this matches the cash version: the rate differential plus the spot move.
Derivatives make high leverage easy and need little upfront cash. That efficiency is one reason carry positions can grow very large relative to the capital behind them.
The return formula
For any of the three versions, the approximate return on the position’s notional over a holding period is:
Carry trade return ≈ (target rate − funding rate) + % change in the target currency against the funding currency − transaction and financing costs
Assumptions behind this approximation: rates are annualized and scaled to the holding period, interest is not compounded, and the exchange rate change is measured from entry to exit. For return on the trader’s own capital, multiply the notional return by the leverage used. Leverage applies to the currency move and to the interest equally.
The differential is the target currency’s rate minus the funding currency’s rate. Exchange rate movements do not change its sign; they change the total return.
Interpreting a Carry Trade
The formula raises an obvious question. If the differential is known in advance, why doesn’t everyone collect it risk-free? The answer runs through interest rate parity and one of the most studied anomalies in international finance.
Covered interest rate parity
Covered interest rate parity says the forward exchange rate must reflect the interest rate gap between two currencies. For a pair quoted as base/quote:
Forward rate = Spot rate × (1 + quote currency rate) ÷ (1 + base currency rate)
The rates here are scaled to the contract’s term. Take a hypothetical USD/JPY spot rate of 150, a one-year dollar rate of 5% and a one-year yen rate of 0.5%. The one-year forward rate is 150 × 1.005 ÷ 1.05 ≈ 143.57. The higher-yielding dollar trades at a forward discount.
So a trader cannot borrow yen, buy dollars, earn 5% and lock in today’s spot rate for the conversion back. The only rate available for locking in is the forward rate, set exactly where the interest gain disappears, so a fully hedged carry trade earns roughly nothing beyond the funding cost. If it didn’t, traders would run the risk-free arbitrage until prices moved back into line. Carry is only available to someone who leaves the currency exposure open.
Uncovered interest rate parity
Uncovered interest rate parity (UIP) is a stronger claim. It says that without any hedge, the high-yield currency should, on average, depreciate by about the amount of the rate differential. The forward rate would then be an unbiased forecast of the future spot rate. If UIP held, carry trades would earn zero on average, with the interest income offset by currency losses.
The forward premium puzzle
Historical data has not cooperated. Across many currency pairs and decades, high-interest currencies have often depreciated less than the differential implied, and sometimes they appreciated. The forward rate has been a biased predictor of the future spot rate, and that bias has tended to favor the carry trader. This is the forward premium puzzle, also called forward bias, and it is the main reason carry trades have shown positive average excess returns over long samples.
Researchers have offered several explanations, and none has settled the question:
- Risk premia. Investors may demand extra compensation for holding currencies exposed to economic or political risk, beyond what simple rate gaps reflect. Critics say this restates the puzzle more than it explains it.
- Crash risk. Carry returns may be payment for bearing rare but severe losses when markets panic and funding currencies spike.
- Central bank policy. Interventions and policy expectations can shift currency demand in ways current rates do not capture.
- Institutional factors. Large institutions with fixed mandates or hedging needs may trade currencies regardless of parity conditions.
- Behavioral factors. Herding and overreaction to news can bias currency flows.
- Market microstructure. Order execution and the spread of information can produce persistent patterns in order flow.
- Global imbalances. Long-running capital flows tied to trade and investment can keep sustained pressure on certain currencies.
More recent work adds conditions. A 2024 study of selected advanced-economy currency pairs found that forward bias appeared mainly when interest rate differentials were positive. Periods when a normally higher-rate currency’s rate fell below the other’s tended to coincide with crises, and in those periods the usual bias could disappear or reverse. Switches between these regimes are often abrupt, so forward bias can stop working exactly when traders most need it.
The shape of carry returns
Carry trades lose differently from how they win. Interest accrues a little at a time, while currency losses can arrive in large, sudden jumps when funding currencies surge during market stress. Carry returns therefore show crash risk and a negatively skewed distribution: many small gains and a long tail of large losses, or “up the stairs, down the elevator.” Such a strategy can look stable for years, and its volatility figures can understate its true risk, because ordinary volatility measures assume something closer to a bell curve. The concept of skewness explains why average return and standard deviation describe it poorly.
The Yen Carry Trade
The yen carry trade is the most widely cited case of the strategy because the conditions behind it lasted so long. After its asset bubble burst in the early 1990s, Japan faced weak growth and deflationary pressure. The Bank of Japan (BOJ) cut rates to near zero by the late 1990s and introduced a negative policy rate in 2016. For roughly a quarter century, the yen was one of the cheapest currencies in the world to borrow.
That made the yen a natural funding currency. Hedge funds, banks and other institutions borrowed yen or shorted it through derivatives, and put the money into higher-yielding assets such as U.S. Treasurys, Australian and New Zealand dollar deposits, emerging-market bonds and global equities. Japanese households also took part by investing savings in higher-yielding foreign currency accounts and products. Pairs such as AUD/JPY, NZD/JPY and USD/JPY became shorthand for the trade among currency traders.
The scale was large enough to affect exchange rates and asset prices outside Japan. By 2007 the yen carry trade was estimated at around $1 trillion. Because so much of the trade ran through leveraged and derivative positions, no single figure captures it precisely, which is one reason its unwinds have repeatedly caught markets by surprise.
What happens when the yen carry trade unwinds
Unwinding means closing carry positions. In a conventional yen-funded trade:
- Investors sell the assets they bought with borrowed yen, such as foreign stocks or bonds.
- They use the proceeds to buy back yen and repay the loans, which raises demand for the yen.
- Traders holding derivative positions close or offset them, which has the same effect on the exchange rate even without a literal asset sale or yen loan.
The process feeds on itself. As the yen rises, other carry positions lose money, and margin calls and risk limits force more exits, which push the yen higher still. Meanwhile, the assets bought with borrowed yen face selling pressure in their own markets, so an unwind can move global stock and bond markets.
2008: the crisis unwind
The global financial crisis produced the classic example. As asset prices collapsed in 2007 and 2008, leveraged investors rushed out of yen-funded positions. The yen appreciated by roughly 19% against the U.S. dollar from January 2008 to January 2009. Several major carry target currencies fell by roughly 20% to 35% against the yen between September 2008 and January 2009. For a leveraged trader collecting a few percent a year in carry, moves of that size erased years of interest income and, in many cases, the capital behind the positions.
An earlier episode, in October 1998, showed the same dynamic on a smaller scale. USD/JPY fell sharply over a few days as leveraged funds cut positions during the market stress that surrounded the Russian default and the troubles at the hedge fund Long-Term Capital Management.
2024: a policy-driven unwind
The 2024 episode started with Japanese monetary policy rather than a global credit crisis. Facing higher inflation, the BOJ ended its negative interest rate policy in March 2024, then raised rates again at the end of July and signaled reduced government bond purchases. Higher yen funding costs and a rising yen caught many leveraged holders of short-yen positions off guard.
As the yen climbed against the dollar, traders closed carry positions, adding to yen demand and to selling in riskier assets, while disappointing U.S. economic data added to the pressure. On August 5, 2024, Japan’s Nikkei 225 fell 12%, its second-largest percentage drop on record, and the S&P 500 fell 3%, its largest one-day decline in almost two years. Both indexes rebounded the next day. Analysts at major banks suggested the unwind among speculative investors was only partly complete. By the end of the month, the Federal Reserve chair had signaled a September rate cut, which would narrow the dollar-yen differential from the other side.
The episode showed both sides of the central bank effect. A funding-currency hike raises the cost of the trade and tends to strengthen that currency, and a target-currency cut reduces the income. If both happen together, the trade is squeezed from both ends.
Example of a Carry Trade
The figures below are hypothetical and chosen to show the mechanics. They are not current rates or any broker’s terms.
A one-year cash carry trade
Assumptions: USD/JPY spot at 150, a yen borrowing rate of 0.5% a year, a one-year dollar deposit rate of 5%, simple interest, no transaction costs or taxes.
- Borrow ¥15,000,000.
- Convert at 150 to get $100,000.
- Deposit for one year at 5%, ending with $105,000.
- Yen debt at year end: ¥15,000,000 × 1.005 = ¥15,075,000.
How the trade ends depends entirely on the exchange rate at the end of the year:
- Rate unchanged at 150. $105,000 converts to ¥15,750,000. After repaying ¥15,075,000, the profit is ¥675,000, about $4,500 at 150, or 4.5% of the original notional. This is the rate differential in action.
- Dollar strengthens to 160. $105,000 converts to ¥16,800,000. Profit: ¥1,725,000. The currency move added more than the interest did.
- Yen strengthens to 140. $105,000 converts to ¥14,700,000, which is ¥375,000 short of the debt. The trade loses money despite earning 5% interest.
The break-even exchange rate is ¥15,075,000 ÷ $105,000 ≈ 143.57. A yen appreciation of about 4.3%, from 150 to 143.57, wipes out a full year of carry. Compare this with the covered interest parity calculation above: the one-year forward rate with the same inputs is also 143.57. This is no coincidence. Locking in the forward rate would guarantee the break-even outcome and remove all the profit. The carry trader is betting that the spot rate in a year will be above the forward rate.
A leveraged spot forex position
Assumptions: a trader deposits $10,000 as margin and opens a long USD/JPY position with a notional of $100,000 (10:1 leverage) at 150. The net rate differential, after the broker’s markup, is a hypothetical 4.5% a year in the trader’s favor.
- Daily carry: $100,000 × 4.5% ÷ 365 ≈ $12.33 a day, about $4,500 over a year if nothing changes. That is 45% on the $10,000 margin.
- Price risk: each 1% move in USD/JPY changes the position’s value by roughly $1,000, which is 10% of the margin. A 4.5% drop in the pair cancels a full year of carry.
- Larger move: if USD/JPY falls from 150 to 140, the loss is $100,000 × (140 − 150) ÷ 140 ≈ −$7,143, about 71% of the deposit. That is roughly 19 months of the hypothetical carry, lost to a move of under 7%.
Long before a loss that size, most traders would face a margin call or an automatic stop-out, which would close the position near the bottom and lock in the loss. Leverage scales the interest, and it scales every pip of price movement just as much.
Why Carry Trades Matter
They move capital across borders. When rate differentials are wide and markets are calm, carry demand can push money into high-yield currencies and their bond and equity markets, and the funding currency tends to weaken. These flows can lift asset prices in target countries and keep the funding currency cheaper than domestic conditions alone would suggest.
They tie markets together. A shock to the funding currency travels to the assets it financed. In August 2024, a policy change in Tokyo reached U.S. equities within days, and the 2008 unwind amplified the crisis through currency markets.
They act as a gauge of risk appetite. Carry positions are popular when investors are willing to take risk and volatility is low. When stress rises, the same positions are closed. Traders therefore watch carry-sensitive pairs, such as yen crosses against higher-yielding currencies, as one indicator of market sentiment.
They complicate central bank policy. Rates set for domestic reasons also change the incentives for carry traders abroad. A funding-currency central bank that starts raising rates after a long period near zero can move exchange rates far more than the rate change itself, because it alters positions built up over years.
Risks and Limitations of a Carry Trade
The carry trade has done well in calm markets and badly in turbulent ones. The main risks fall into two groups: those that hit an individual position, and those that hit the whole market at once.
Position-level risks
- Exchange rate risk. This is the dominant risk. A move in the funding currency’s favor larger than the accumulated interest produces a net loss, and currency moves can be many times the annual differential.
- Interest rate changes. A funding-currency hike or a target-currency cut narrows the differential, and expectations can move currencies before any decision is announced.
- Devaluation and emerging-market risk. High-rate emerging-market currencies often pay that rate because of inflation, political risk or weak external finances. A sudden devaluation, hyperinflation or political crisis can depreciate the target currency far faster than its high yield can make up for.
- Economic data. Surprises in inflation, GDP growth and trade balances can shift both currency values and rate expectations.
- Geopolitical shocks. In risk-off episodes, money tends to move into safe-haven currencies such as the U.S. dollar, Swiss franc and yen. When the funding currency is itself a safe haven, a shock can hurt the trade on both sides at once.
- Costs. Rollover markups, spreads and hedging costs reduce the net carry, and protective options held for long periods can eat much of the differential.
- Capital controls and convertibility. Some high-rate currencies cannot be freely moved offshore, so exposure may only be available through non-deliverable forwards or offshore markets, whose prices can diverge from onshore rates. China’s split between the onshore and offshore yuan is one example.
Market-wide risks
- Crowding. When many investors hold the same carry position, a shift in sentiment can start a rush for the exit.
- Leverage and forced unwinding. Margin calls in a sharp move force selling, which adds to the move.
- Liquidity black holes. In extreme stress, as in 2008 or the early weeks of the pandemic, liquidity can almost vanish. Falling prices attract more sellers instead of buyers, and exits happen at much worse prices than traders planned for.
- Algorithmic amplification. Automated trading models can add to selling when volatility jumps, which speeds up the unwind.
The worst losses tend to come when several of these hit together: a rate change, a safe-haven rush, crowded positioning and thin liquidity. Each makes the others worse.
Behavioral traps
The return pattern encourages specific mistakes. Long runs of steady income breed overconfidence and extra leverage just as positions become crowded, and fear of missing out pulls traders in late, after the easy currency gains have happened. In an unwind, those who exit first do best, while traders who react to every scare risk closing positions that would have recovered. The strategy resembles selling insurance: the premium comes in regularly, and the claims arrive together.
Managing Carry Trade Risk
No technique removes the exchange rate risk without also removing the carry. Risk management aims to keep losses survivable when the unfavorable tail arrives.
- Size positions for the currency move, not the interest. Size each position so that a large adverse move costs an acceptable share of the account, using a tool such as a position size calculator.
- Use leverage sparingly. As the leveraged example shows, modest leverage can turn a routine currency move into a near-total loss of margin.
- Set exit rules in advance. A stop-loss caps the loss on a normal move. In a disorderly unwind or a weekend gap, it becomes a market order once triggered and can fill well beyond its trigger price.
- Hedge selectively. Forwards remove the carry along with the risk. Options can cap the downside while keeping some carry, but their premiums rise with volatility, which is exactly when protection is wanted.
- Diversify across pairs, with care. Spreading carry across several funding and target currencies reduces dependence on one central bank, but in a broad risk-off episode carry positions often fall together.
- Watch the drivers. Track central bank policy in both countries, inflation trends, and signs of crowding or rising volatility. A funding-currency central bank moving toward hikes is the clearest warning sign in the yen trade’s history.
Carry Trade vs. Related Concepts
Several terms sound similar and get mixed up.
Carry trade vs. arbitrage. Arbitrage locks in a profit from a price difference with no market risk. A carry trade depends on leaving the currency exposure open. Hedged with a forward, it earns roughly nothing.
Carry trade vs. cost of carry. In futures and commodities, cost of carry refers to the net cost of holding the underlying asset until delivery, such as storage, insurance and financing, minus any income it pays. It explains the gap between spot and futures prices. A carry trade is a strategy that seeks to profit from net positive carry, while the cost of carry is a pricing input.
Positive vs. negative carry. Positive carry means the position earns more than it costs to hold, and negative carry means the opposite. A trader can hold a negative-carry position deliberately, for example shorting a high-yield currency because they expect it to fall, but they pay the differential every day they wait. Shorting the higher-yielding currency is sometimes called a reverse carry trade.
Carry trade vs. “carry forward” in stock markets. In some equity markets, particularly in India, “carry forward” means holding a leveraged or derivative position open beyond the current session or settlement period instead of closing it the same day. The term concerns position duration, not an interest rate differential.
Carry trade vs. carry in quantitative finance. Quant strategies use “carry” as a general factor: the return an asset would earn if its price did not change. That includes currency rate differentials, bond yield and roll-down, and futures roll yield. The currency carry trade is the best-known example of this broader carry factor.
The Bottom Line
Carry trades matter most to forex traders and investors who hold leveraged positions open for weeks or months, because the interest collected is small next to the currency risk. The yen unwinds of 2008 and August 2024 showed that a sharp rise in the funding currency can erase months or years of carry very quickly. Sizing positions for the currency move, using leverage sparingly and watching central banks on both sides of the pair help keep losses survivable.















