When it comes to Forex, most people think of it as a short-term trading option, rather than long-term investing one. This is partly due to the popularity and high visibility of retail Forex trading, which is geared heavily towards short-term positioning, but there are also good reasons for the thinking of Forex as a short term strategy. The biggest reason is that major currencies tend to mean-revert over long periods of time, with even most major trends collapsing after a few weeks, or months at most.
That being said, there are plenty of great Forex investing strategies for investors, if they know where to look. Here are three of the major types of Forex investment strategies offered by banks and investment houses.
Fair Value Strategy
This involves determining the purchasing power parity of currencies in an attempt to identify which currencies are objectively overvalued or undervalued. For example, if $10 worth of currency in country A buys more goods than $10 worth of currency in country B, it would be said that country A’s currency is overvalued and country B’s currency is undervalued. The undervalued currency would be bought and the overvalued currency sold in the hope that the exchange rate would move in that direction.
This seems like a logical strategy. However, applying this strategy consistently in recent years would have led to results that were breaking even at best, and losing heavily at worst. This strategy simply does not work as an overarching investment system.
The Carry Trade
This involves going long of currencies with relatively high interest rates, and short of currencies with relatively low interest rates. This is based upon the theory that the interest rate differentials will cause flows of money from the currencies with the lower rates to those with higher rates. As well as moving the price and creating a profit, there is also the potential to benefit from the positive overnight interest payments.
A common methodology is to use 3m LIBOR rates of different currencies.
Claim up to $26,000 per W2 Employee
- Billions of dollars in funding available
- Funds are available to U.S. Businesses NOW
- This is not a loan. These tax credits do not need to be repaid
A back test was conducted by looked at the seven major global currencies (USD, JPY, EUR, GBP, AUD, CAD, and CHF) over a 10 year period, from March 2003 until March 2013.
The test was simple: at the beginning of each month, the three currencies with the highest 3M LIBOR rate were bought, the three currencies with the lowest 3M LIBOR rate were sold. At the end of each calendar month, that month’s position was exited.
The results do not include the actual interest payment differentials that would have been earned. The results are simply based upon the changes in the exchange rates between the currencies and do not include transaction costs.
10 Year Return | 64.32% |
Annualized Return | 5.09% |
Average Monthly Return | 1.05% |
Median Monthly Return | 1.84% |
% Winning Months | 52.46% |
Max Consecutive Winning Months | 6 (Aug-09) |
Max Consecutive Losing Months | 7 (Feb-09) |
Max Drawdown | 340.25% |
Standard Deviation | 19.71% |
It could be said that this strategy tends to be profitable under “normal” market conditions. Note that the extremely large draw down occurred during the 2008/09 financial crisis.
Momentum
There are two types of momentum strategies. There is “time series momentum” which buys an asset if its price is going up and sells the asset if its price is going down. Secondly, there is “comparative momentum,” where the strongest performer within an investment universe is bought and the weakest sold. As recent academic studies have found the most profitable holding period that can be applied to Forex momentum studies is 1 month, this was used as a holding period in all the back tests.
Hypothetical trades were made in the direction of the pair with the strongest directional movement over periods of time ranging from the previous 1 month to the previous 1 year, and exited each position 1 month later. A universe of the 24 most widely traded currency pairs/crosses was used. The results were as follows:
1M | 3M | 6M | 9M | 12M | |
11 Year Return | 35.55% | 57.07% | 32.22% | -23.45% | -32.76% |
Annualized Return | 2.80% | 4.25% | 2.65% | -2.08% | -2.87% |
Average Monthly Return | 0.27% | 0.44% | 0.25% | -0.19% | -0.27% |
Median Monthly Return | 0.14% | 0.57% | 0.01% | -0.21% | -0.25% |
% Winning Months | 48.09% | 54.96% | 48.86% | 45.80% | 43.51% |
Max Consecutive Winning Months | 6 (Jun-09) | 6 (Nov-05) | 6 (Jul-11) | 6 (Jul-09) | 6 (Jul-11) |
Max Consecutive Losing Months | 8 (Jan-14) | 6 (Nov-03) | 11 (Nov-12) | 6 (Nov-12) | 8 (Aug-12) |
Max Drawdown | 32.24% | 28.89% | 34.12% | 72.79% | 60.41% |
Standard Deviation | 4.27% | 4.03% | 3.95% | 4.25% | 5.06% |
It can be seen that comparative momentum judged by periods of 6 months or less has provided a winning edge, with the 3 month period performing particularly strongly.
Time series momentum is not comparative. An asset is exhibiting time series momentum based upon its own current price relative to its historical price.
The back test used the same universe of 24 currency pairs/crosses, entering hypothetical trades in each and every pair in the direction of its most recent directional movement over periods of time ranging from the previous 1 month to the previous 1 year, and exited each position 1 month later. The results were as follows:
1M | 3M | 6M | 9M | 12M | |
11 Year Return | 205.36% | 344.40% | 105.03% | 125.74% | 72.92% |
Annualized Return | 10.68% | 14.88% | 7.08% | 8.27% | 5.63% |
Average Monthly Return | 1.54% | 2.63% | 0.82% | 1.01% | 0.60% |
Median Monthly Return | 2.37% | 2.60% | 5.34% | 6.09% | 5.21% |
% Winning Months | 54.14% | 51.88% | 54.88% | 53.38% | 51.13% |
Max Consecutive Winning Months | 5 (Jul-13) | 6 (Sep-09) | 6 (Jan-13) | 6 (Jan-13) | 6 (Jan-13) |
Max Consecutive Losing Months | 5 (May-05) | 5 (Jun-12) | 5 (Sep-11) | 6 (Jan-09) | 5 (Mar-08) |
Max Drawdown | 282.09% | 255.00% | 378.27% | 378.27% | 405.91% |
Standard Deviation | 39.09% | 37.62% | 31.24% | 31.30% | 31.54% |
It can be seen that time series momentum has tended to produce an even greater winning edge than comparative momentum during the period of the back test. However, as the test involved “trading” 24 currency pairs every month, the results are extremely volatile and every time period except the 3 month strategy showed a maximum drawdown greater than the 11 return. It is also significant that there is little difference between the results of the 6 month to 12 month tracks, and that these strategies initially produced better returns than the shorter time periods prior the financial crisis of 2007/08.
In both the comparative and time series momentum studies, the three month period has proved the most profitable overall over time.
Conclusion
Regarding widely used quantitative Forex investment strategies that do not make use of fundamental data beyond interest rates, and price series, it seems clear that over recent years, simple momentum investment strategies have performed most strongly. Measures certainly need to be taken when implementing such strategies to minimize draw down.
