Understanding Systematic FX Strategies
This is the second installment of our three-part series exploring the underperformance of currency markets and strategies since the great financial crisis. This week, we compare the mechanics of the three major FX strategies against the theory that markets should efficiently achieve interest rate parity across currencies.
By: Verdad Research
There are three major currency trading strategies: carry, momentum, and value. Each strategy uses a different signal to borrow one currency and buy another, capturing both the interest rate differential between the two countries and changes in the nominal exchange rate. The carry strategy derives its signal from interest rate differentials, momentum derives its signal from trends in exchange rates, and value derives its signal from the real purchasing power of currencies.
Theoretically, these currency trading strategies should not offer any returns. Academics have a theory they call Uncovered Interest Rate Parity, which holds that any gains captured through differences in country-level interest rates should be counteracted by an equivalent change in the exchange rate. However, data show that there have been prolonged deviations from this theory - in fact, positive spreads have actually predicted exchange rate appreciations for much of foreign exchange market history. This dynamic has allowed traders to successfully execute and capture gains from currency trading strategies over various time periods. Here we backtest the strategies based on these signals and evaluate their historical performance.
We built simple models for each of the carry, momentum, and value strategies and tested each of them across 21 currency pairs against the US dollar. We chose to limit our strategy survey to the US dollar for two primary reasons. First, we find the results are more interpretable. Second, by holding out most cross pairs, we establish a hold-out sample, or data we can use to validate our findings.
The carry strategy seeks to capture the interest rate differential between two countries. If prevailing interest rates in Japan are 1% but approach 19% in Turkey, investors can borrow yen to buy lira and harvest the yield differential, or carry, between the two countries.
The momentum strategy is like momentum strategies in other asset classes and offers a systematic way to bet that strengthening currencies will continue to strengthen and weakening currencies will continue to weaken.
The value strategy seeks to take long positions in currencies that are undervalued relative to their purchasing power and short positions in currencies that are overvalued relative to their purchasing power.
Figure 1 shows the portfolio statistics for each of these three strategies. From 2000 to 2010, risk-adjusted returns were high. During this period, carry delivered a 0.21 Sharpe ratio, momentum a 0.75 Sharpe ratio, and value a 0.14 Sharpe ratio. From 2010 to present, however, the tides reversed strongly, and each of these strategies has delivered negative returns.
Figure 1: Portfolio Statistics for Momentum, Carry, Value (Jan 2000–Present)
Source: Thomson Reuters Datastream
Each of these strategies, despite operating with different signals, are highly correlated. Figure 2 shows a correlation matrix. Notably, carry and value strategies follow each other quite closely. Academic research has found that value strategies, which seek to buy currencies with depreciated real exchange rates, tend to also be currencies with high carry. Momentum is highly correlated as well, though slightly less than value and carry.
Figure 2: Correlation Matrix (Jan 2000–Present)
Source: Thomson Reuters Datastream, Analysis: Verdad
Looking at the underlying holdings, each strategy tends to build concentrated positions in high-carry emerging market currencies. Intuitively, this makes sense: the currencies of risky economies should offer higher rewards to their holders. However, research indicates that the carry is compensation for global risk and unwinds sharply with positive increases in volatility. Or, as the saying goes, returns to high-carry currency strategies go up by the stairs but down by the elevator. Figure 3 shows the negative relationship between yield and carry strategy drawdowns. This relationship shows that, as investors chase higher yields, they expose themselves to the potential for ever-larger drawdowns.
Figure 3: Carry vs. Drawdowns (2000–Present)
Source: Datastream, Analysis: Verdad
Figure 4 frames this relationship in a slightly different light: high-yielding EM currencies have exhibited a negative historical skew while low-yielding developed-market currencies exhibited a positive one.
Figure 4: Carry vs. Skewness of Returns (2000–Present)
Source: Datastream, Analysis: Verdad
Investors generally look for positive, not negative skew, and thus might logically prefer the low-carry to the high-carry strategy, despite the potential for higher returns in the higher-carry currencies. It’s worth noting that value and carry performed particularly poorly in 2008, as shown in Figure 5 below.
Figure 5: Performance of Carry, Value, and Momentum Strategies (1999–Present)
Source: Datastream, Analysis: Verdad
In sum, carry, value, and momentum strategies have been a mixed bag for investors. 2000–2010 was an attractive period for FX investing, and each strategy captured attractive risk-adjusted returns. Since 2010, the story has been quite different, with each strategy delivering negative returns—a victory for Uncovered Interest Rate Parity. When we conclude this series next week, we’ll explore a multi-factor short strategy that takes advantage of these strategies’ tendency to go down by the elevator.