am/FX: A Good Time of Year to Buy FX Vol 16AUG22

A good time of year to buy FX vol

This is a pretty good time of year to buy FX vol. It makes sense because:

1.      Stocks trade weak in September as you can see here. Higher equity vol means higher FX vol, ceteris paribus.

2.      Jackson Hole has stirred the pot in some years.

3.      FX punters return from summer and buy vol to press their bets and boost leverage into Q4. Speculative engagement is very high in September.

Here’s how FX vol moves over the course of the year:

Cumulative weekly change in JP Morgan FX vol index, by week of year

Since 1992

That chart is nice, but if you take out 2008, it looks much less nice as you see here:

Cumulative weekly change in JP Morgan FX vol index, by week of year

Since 1992, excludes 2008

Still, there is a pattern for FX vol to bottom and rally week 30 to week 40 of the year and we’re in week 34 right now. Slicing things differently, here’s 4-week rolling change in AUDUSD, by week (again excluding 2008 because it distorts everything).

Rolling four-week change in AUDUSD, by week of year

Since 1990 (excluding 2008)

Absolute value of change

We are in week 34 right now, the perfect time to buy some FX vol while everyone is sleeping.

USDCNH volatility looks particularly attractive to me. USDCNH is the kind of trade where momentum is everything because it realizes quite a lot at times, then sits there and does nothing for ages. So, you need a methodology to pick the timing. Right now, it looks like a good time to take a shot because the pair has been doing nothing in a bull flag for ages and just broke out yesterday as you can see here:

USDCNH textbook flag break

Rate cuts and bad data in China, too. There are six states of volatility:

Of those six, I like to buy low and rising volatility best. USDCNH vol fits that:

USDCNH 1-month volatility in 2022

Simple trades like buy 1-month 6.85 USDCNH vanilla call or buy 3-month 6.90 in USDCNH look logical to me. Here is the average seasonality of USDCNH 1-month volatility:

Cumulative weekly change in 1-month USDCNH vol, by week of year

Since 2011

You could also apply this same analysis to NZDUSD vol or USDCAD vol. The kiwi story could be interesting as the RBNZ meets tonight and they could be getting close-ish to a point where the end of hikes is finally in sight. CAD could have a very tough few months going forward as the three main engines of Canadian growth have stalled, big time. Strong US growth, high oil prices, and booming real estate are all in the rearview for Canada. 3-month 1.33 digital in USDCAD costs around 20% or so and seems logical (ballpark price off 1.2905 spot).

There are other long FX vol ideas that make sense. Choose your weapon.

EURUSD

The EURUSD take profit hit at 1.0166. I’m still generally bearish but feel there is no reason to get involved down here right now. The tech setup worked perfectly, but the cross-market stuff isn’t suggesting further downside from here as rate differentials are flatlining, BTPs are quiet, and Euro Stoxx are not sending any signals. I am flat like a crepe.

Final thought

For my educational Substack, I’m doing a series of surveys to see if there is any correlation between conviction level and accuracy of forecasts. Here are the results of last week’s survey (conducted August 11). The main takeaways are: Everyone is bearish stocks, EURUSD, and gold. Nobody has a clue on copper. Sample size = 217. I will be running this for a while and hope the experiment will yield something interesting with regard to the relationship between confidence and accuracy in market forecasting.

Have a caffeinated day.

good luck ⇅ be nimble

---

Timmy’s!

Click here to subscribe to am/FX