Forex Carry Trade Backtest: Does Holding High-Yield Currencies Still Work?
# Forex Carry Trade Backtest: Does Holding High-Yield Currencies Still Work?
The carry trade is one of the oldest strategies in forex: borrow in a low-yielding currency, invest in a high-yielding one, and pocket the interest rate differential. But with central banks moving in sync and volatility spikes becoming more frequent, does this strategy still hold up in backtesting?
Historical Context
From 2000 to 2007, carry trades were practically a license to print money. Borrowing in Japanese yen at near-zero rates and investing in Australian dollars at 5-7% produced consistent returns. Then 2008 happened, and carry traders learned the meaning of unwinding risk.
Our Backtest Parameters
We tested a simple carry trade portfolio from 2010 through 2025. The strategy went long the three highest-yielding G10 currencies and short the three lowest-yielding ones, rebalancing monthly based on central bank rate differentials. Position sizing was equal-weighted across all six positions.
The Results Were Mixed
The strategy produced a total return of 4.2% annualized before transaction costs, with a Sharpe ratio of 0.38. While positive, this is significantly below the pre-2008 golden era returns of 8-12% annualized. The strategy suffered particularly during risk-off events in 2015, 2020, and 2022.
Drawdown Analysis
Maximum drawdown was 22.4%, occurring during the March 2020 COVID crash when carry trades unwound violently. The strategy also experienced a 15.8% drawdown during the 2022 rate hiking cycle when previously low-yielding currencies like the USD suddenly became high-yielders, causing portfolio churn.
Adding a Volatility Filter
We introduced a VIX-based filter that reduced position sizes by 50% when VIX exceeded 20 and closed all positions when VIX exceeded 30. This simple risk management overlay improved the Sharpe ratio to 0.61 and reduced maximum drawdown to 12.1%, though it sacrificed some total return.
Emerging Market Carry Enhancement
Extending the universe to include emerging market currencies (BRL, ZAR, MXN, TRY) increased returns to 6.8% annualized but also increased maximum drawdown to 31.2%. The higher yields come with crash risk that cannot be diversified away easily.
Regime Dependence
The carry trade performs best during low-volatility, gradually changing rate environments. Our backtest showed that 78% of total returns came during periods when VIX averaged below 15. This makes the strategy essentially a bet on continued calm, which works until it catastrophically does not.
Transaction Cost Impact
With realistic spread costs for monthly rebalancing, the G10-only strategy return dropped to 3.1% annualized. For retail traders paying wider spreads, the strategy becomes marginal at best. Institutional-grade execution is almost required to make pure carry work.
Conclusions for Modern Traders
The carry trade still generates positive returns over long periods but is no longer the easy money it once was. A volatility-filtered version with conservative sizing remains viable as one component of a diversified forex strategy. However, standalone carry trading without risk management is essentially picking up nickels in front of a steamroller, a cliche because backtesting proves it true.