am/FX: Drawn Out Battle

Subscriber-only

FX intervention doesn’t just instantly flip the trend, usually.

I think we are close to the place to buy a dip in USDJPY but I acknowledge that it’s scary.

by Brent Donnelly

September 22, 2022

Round number bias

Current Views

Go long USDJPY @ 140.33

(limit order, trade not open yet)

Stop loss 138.88

Take profit 144.44

September 22, 2022

Drawn out battle

The BOJ is going one way and the MOF is going the other. I am surprised and intrigued at the lack of coordination. The intervention was large and will buy some time but USDJPY is unlikely to reverse for real until global yields peak. Historically, USDJPY generally has worked eventually but it was a long, drawn-out battle over months or even years.

Here are the interventions in 2010, 2011, and 2012. You can see the market impact was large but temporary. The moves were about 3% on the normal interventions, not including the one after Fukushima. The high to low move last night was 3.6%, so it looks similar to those interventions back in the day.

USDJPY interventions in 2010, 2011 and 2012

My long USDJPY idea was right in theory but wrong in real life as my take profit was 145.98 and the high was 145.90. A good argument for spreading your TP out a bit to avoid the false precision of a particular exact level.

It is generally rare for central bank intervention to instantly turn a currency pair. Usually it requires long and repeated interventions and even then, sometimes it doesn’t work. For a complete chronology of JPY FX intervention, this Reuters article is good.

USDJPY becomes extremely difficult to trade in intervention regimes because the reasons it was going up are still valid, but rallies are always under threat by the next round of MOF USDJPY selling. It becomes a cat and mouse game where everyone tries to guess the MOF’s next move. This instance is particularly interesting because the BOJ is still easing aggressively and thus pushing the other way.

The 1998 USDJPY collapse is seared into the memory of anyone that was trading at that time. Intervention preceded but did not cause that epic collapse. Here’s how it played out in 1998:

USDJPY interventions in 1998

Just before last night’s intervention, the SNB hiked a bit less aggressively than hoped for. Long CHF has been a favorite view of speculators and CTAs for ages and so the unwind in CHF and the JPY intervention give you a chart that looks like this:

CHFJPY hourly since November 2021: Only a flesh wound

So what do we do now?

I don’t think the USDJPY intervention will immediately end the trend in USDJPY. If you want to buy USD, I think 139.50/140.00 will be the first big support because that was the key pivot in July and September as you can see here:

USDJPY hourly back to June 2022

That old resistance should now become support. 140/145 should be the equilibrium range. I am going to take a stab at buying down near 140.00 with a stop below 139.00. Full details in sidebar. It is not safe to assume the MOF will win on the first try as this is likely to be a drawn out saga.