Seasonality In The Forex Market - Kathy Lien

When it comes to trading the currency market, you might often find yourself making one of two choices: pro dollar or anti dollar. As a component of 90% of all currency transactions, the U.S. dollar has long been the primary driver of fluctuations in exchange rates. Most traders will analyze the future direction of the dollar by using either fundamental or technical analysis or a combination of both. However, few people realize that the time of the year could also play a role in how the U.S. dollar behaves against various currencies. The study of technical analysis is all about analyzing past price activity through the use of indicators. There are many different ways to do this, which explains why there are so many different types of technical indicators out there.
Filtering Out Noise
Despite this, many traders may not realize that there is no clearer way to analyze past price behavior than to look at the price activity itself without the noise of indicators. When you do that, you will find the presence of seasonality in some currency pairs, which is when there is a predictable change that repeats every year at the same period in time. For example, did you know that in eight out of the past 10 years (between 1997-2006), the U.S. dollar rose in the month of January against the euro? Or that in nine out of the past 10 years, the U.S. dollar rose against the Japanese yen? Although there are no guarantees that historical patterns will repeat themselves, the fact that a pattern has been repeated 80-90% of the time makes it statistically significant. This information can be very valuable when you are trading. In this article, we'll go over why seasonality is an important concept in the forex market and why, although it only occurs in rare instance, it should not be ignored. (For related reading, see Capitalizing On Seasonal Effects.)

July: A Positive Month for USD/JPY The strongest seasonality example is the one in USD/JPY, which can be seen in Figure 1 below. In 90% of the samples (or nine out of the past 10 years), USD/JPY ended the month of July higher than where it started. It is difficult to pinpoint an exact reason for why USD/JPY tends to behave this way in the month of July, although it could be related to the end of the first quarter in Japan or the beginning of the second half of the year in the US. Either way, this instance of seasonality is very strong and one that is worth keeping in mind if you find some short USD/JPY trades during the month of July. The presence of seasonality may encourage you to take a smaller-than-usual short position or to avoid longer term short USD/JPY trades during this period.

Source: FXCM

Figure 1

August: USD/JPY Gains Made in July are Often Erased Seasonality is also present in USD/JPY during the month of August. Although the seasonality is not as strong as the seasonality in July, it is still one of the strongest examples in the currency market. You can see in Figure 2, below, that a good portion of the gains earned in July were erased in August. In fact, a look at other yen crosses quickly reveals that in a calendar year August tends to be the strongest month for the Japanese yen across the board. In other words, other currencies such as U.S. dollar, euro and British pound have a strong tendency to fall against the yen in August.

Source: FXCM

Figure 2

September: A Very Positive Month for the GBP/USD The presence of seasonality can also be seen in September in the GBP/USD. Based the analysis of past data, in eight out of the last 10 years (between 1997-2007), the British pound rallied against the U.S. dollar in the month of September. With most of Europe taking the entire month of August off for their summer holidays, it can be argued that the British pound rallies in the month of September because European traders re-awaken from their summer slumber and begin to trade actively in the markets once again. However, this relationship seems to have broken down in 2005 and 2006. The two back-to-back months of losses that occurred in those years calls the future reliability of this specific seasonality into question.

Source: FXCM

Figure 3

January: A Negative Month for EUR/USD Cases of seasonality can also be found in other currency pairs such as the EUR/USD and GBP/USD. The behavior of the U.S. dollar against the euro in the month of January, for example, shows strong seasonality. In eight out of the past 10 years, the U.S. dollar has rallied in the month of January. We have seen the same in 2007, which would improve this statistic to nine out of the past 11 years (between 1997-2007). The reason that we typically see this behavior is because many companies and funds tend to repatriate their money back to their local country at the end of the year to dress up their balance sheets. In the beginning of the year, money is once again sent abroad for new investment purposes. Because everyone starts with a blank slate as far as profits and losses go, their main focus is to initiate new positions. The U.S. market is one of the most liquid markets in the world, which explains why a lot of that investment money ends up there.

Source: FXCM

Figure 4

Implications for Traders As traders, there are many ways that you can apply the knowledge of seasonality to improve your trading. If for example, you are trading the GBP/USD in the month of September, as a longer term trader you can look for opportunities using fundamentals or technicals to buy the GBP/USD or to go in the direction of the seasonal trend. As a shorter term trader, you can reduce your holding period if you are taking a trade that is against the seasonal trend or like longer term traders, you can focus on primarily looking for long GBP/USD trades. Although seasonal patterns do not duplicate themselves 100% of the time, following seasonality rather than fading it may improve your ability to find high probability trades.

Supplemental Data Below is data gathered by FXCM and used to support the notion of seasonality in the forex market. Notice different behaviors of each currency throughout the calendar year. Each of the numbers represents the percentage change between the open on the first trading day of the month and the close of last trading day.  
 

EUR/USD	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sept	Oct	Nov	Dec
1997	(5.92)	(3.02)	1.22	3.63)	1.07	(0.90)	(5.05)	3.02	3.05	2.92	(1.53)	(1.21)
1998	(1.80)	(0.04)	(1.81)	3.20	0.87	(1.39)	1.70	1.63	4.52	0.87	(2.48)	1.61
1999	(3.19)	(3.14)	(2.07)	(1.78)	(1.61)	(0.69)	3.41	(1.27)	1.12	(1.24)	(4.36)	(0.32)
2000	(3.63)	(0.71)	(0.83)	(4.65)	2.83	1.65	(2.70)	(4.20)	(0.68)	(3.96)	2.82	8.07
2001	(0.58)	(1.39)	(5.09)	1.72	(4.90)	0.43	3.09	4.12	0.14	(1.05)	(0.39)	(0.69)
2002	(3.43)	1.22	0.25	3.30	3.77	6.29	(1.38)	0.49	0.49	0.36	0.38	5.49
2003	2.63	0.23	1.27	2.49	5.40	(1.75)	(2.47)	(2.20)	6.03	(0.56)	3.57	5.19
2004	(0.95)	0.16	(1.35)	(2.73)	1.71	0.09	(1.47)	0.99	2.07	2.90	3.72	2.06
2005	(3.77)	1.46	(2.00)	(0.69)	(4.33)	(1.58)	0.13	1.85	(2.59)	(0.27)	(1.70)	0.52
2006	2.63	(1.94)	1.65	4.31	1.40	(0.16)	(0.13)	0.37	(1.09)	0.61	3.76	(0.33)

USD/JPY	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sept	Oct	Nov	Dec
1997	4.86	(0.67)	2.76	2.65	(8.46)	(1.32)	3.47	2.02	(0.24)	(0.02)	6.16	2.18
1998	(2.35)	(0.38)	5.57	(0.19)	4.57	0.07	4.27	(3.84)	(2.00)	(15.03)	6.64	(7.60)
1999	2.73	2.85	(0.32)	0.65	1.88	(0.35)	(5.49)	(4.44)	(2.96)	(2.09)	(1.99)	0.43
2000	5.07	2.80	(6.82)	4.64	(0.19)	(1.44)	3.20	(2.42)	1.35	0.71	1.42	3.64
2001	1.86	0.69	7.64	(2.33)	(3.46)	4.53	0.25	(4.97)	0.34	2.41	0.87	6.74
2002	2.28	(0.95)	(0.48)	(3.21)	(3.37)	(3.85)	0.08	(1.16)	2.79	0.57	0.07	(3.09)
2003	0.91	(1.48)	(0.08)	0.69	0.40	0.16	0.68	(3.01)	(4.51)	(1.41)	(0.25)	(2.27)
2004	(1.41)	3.31	(4.42)	6.02	(0.83)	(0.74)	2.37	(1.73)	0.80	(3.86)	(2.70)	(0.41)
2005	0.96	0.90	2.41	(2.24)	3.33	2.17	1.43	(1.75)	2.60	2.68	2.92	(1.74)
2006	(0.59)	(1.19)	1.74	(3.36)	(0.91)	1.58	0.20	2.36	0.66	(0.99)	(1.00)	2.81

GBP/USD	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sept	Oct	Nov	Dec
1997	(6.44)	1.90	0.62	(0.85)	0.89	1.38	(1.48)	(1.58)	0.14	3.40	0.64	(2.61)
1998	(1.23)	0.53	1.66	(0.08)	(2.43)	2.14	(2.05)	2.78	1.06	(1.45)	(1.69)	0.73
1999	(1.10)	(2.69)	0.84	(0.17)	(0.51)	(1.54)	2.76	(1.04)	2.68	(0.16)	(2.81)	1.23
2000	0.09	(2.27)	0.76	(2.47)	(3.35)	0.89	(1.08)	(3.50)	1.96	(1.94)	(1.56)	4.79
2001	(1.92)	(1.30)	(2.01)	1.24	(1.03)	(0.22)	0.62	2.03	1.44	(1.18)	(2.10)	2.18
2002	(3.00)	0.47	0.60	2.28	(0.16)	5.22	2.04	(0.87)	1.23	(0.20)	(0.45)	3.42
2003	2.25	(4.56)	0.65	1.01	2.36	1.77	(2.65)	(2.05)	5.24	2.04	1.54	3.81
2004	2.13	2.50	(1.15)	(3.69)	2.99	(0.63)	(0.02)	(1.16)	0.53	1.40	3.85	0.43
2005	(1.89)	2.03	(1.58)	0.97	(4.72)	(1.41)	(1.88)	2.71	(2.20)	0.39	(2.30)	(0.37)
2006	3.36	(1.44)	(0.92)	5.09	2.50	(1.14)	1.11	1.98	(1.70)	1.68	3.08	(0.36)

USD/CAD	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sept	Oct	Nov	Dec
1997	(1.76)	1.63	1.16	0.98	(1.46)	0.36	(0.22)	0.75	(0.48)	1.94	1.05	0.47
1998	1.60	(2.07)	(0.31)	0.85	1.82	0.69	3.14	3.59	(2.19)	0.86	(0.59)	0.31
1999	(1.77)	(0.10)	(0.18)	(3.35)	1.16	(0.74)	2.99	(0.90)	(1.68)	0.25	0.26	(1.85)
2000	0.01	0.13	(0.11)	2.10	1.10	(1.06)	0.42	(0.99)	2.08	1.34	0.87	(2.47)
2001	(0.03)	2.58	2.53	(2.64)	0.18	(1.56)	1.27	1.14	1.91	0.58	(0.95)	1.25
2002	(0.26)	0.79	(0.42)	(1.74)	(2.54)	(1.18)	4.35	(1.60)	1.77	(1.80)	0.47	0.54
2003	(3.62)	(2.30)	(1.09)	(2.54)	(4.43)	(1.82)	4.28	(1.27)	(2.40)	(2.42)	(1.48)	(0.20)
2004	2.19	0.72	(1.98)	4.84	(0.73)	(2.19)	(0.11)	(1.18)	(3.92)	(3.45)	(2.57)	1.24
2005	3.11	(0.49)	(1.90)	3.96	(0.35)	(2.37)	(0.15)	(2.80)	(1.99)	1.69	(1.37)	(0.30)
2006	(2.03)	(0.20)	2.80	(4.42)	(1.35)	1.33	1.39	(2.44)	1.30	0.40	1.64	2.24

Source: FXCM

Conclusion
Although instances of seasonality in the forex market are rare, being aware of them can help traders become more in-tune with the outlook for their currency trades. Seasonal patterns won't always bear out as you might hope, but following these trends should help you improve your trading strategy in most cases.