AAPLLoadingMSFTLoadingNVDALoadingGOOGLLoadingAMZNLoadingMETALoadingTSLALoadingAMDLoadingNFLXLoadingCOINLoadingPYPLLoadingSPYLoadingQQQLoadingIWMLoadingJPMLoadingVLoadingDISLoadingKOLoadingJNJLoadingNKELoadingXOMLoadingASMLLoading

FreeUUPJul 20, 2026, 5:43 PM2 min read

The dollar smile is flattening, and the middle is the weak part

Amara OkaforDemoMacro, rates & FX

The dollar index tends to strengthen at both ends of a smile: when the US economy clearly outperforms, and when the world panics and runs for safety. In between, when global growth is steady and nobody is scared, the dollar softens. The middle of that smile looks wider than it has for a while.

UUP

Delayed

Life in the middle

The two strong-dollar conditions are both fading at the margin. US growth is still solid but less exceptional relative to the rest of the world, and there is no acute crisis driving haven demand. That leaves the dollar in the soft middle, where capital looks for returns abroad.

The middle can last a long time. It can also end abruptly if either tail reasserts itself.

What I'm watching

US growth relative to Europe and Asia
Global risk appetite, the switch for haven demand
Capital flows into non-US equities

Where I stand

Watching. The smile is flattening in the dollar's disfavour, and the tails can return faster than the middle develops.

What would change my mind

A clear pickup in US outperformance, or a genuine risk-off shock. Either pulls the dollar back to one of the strong ends of the smile.

The author holds no position in the instruments discussed. Opinion. Not advice. Not a financial promotion approved under FSMA s.21. Quote data is delayed. Past performance is not a reliable indicator of future results.

Comments

27 comments

Greta Rivas·Jul 20

The second point is the one most people are going to skip over.

2 replies

Amara OkaforAuthor·Jul 20

Following on from that —

1 reply

Viktor Reinholt·Jul 20

Right, but consider the second-order effect.

1 reply

Dana Cho·Jul 20

Which is exactly why the timing matters.

1 reply

Priya Nair·Jul 20

Fair, though that assumes the base case holds.

1 reply

Sam Whitfield·Jul 20

And if it doesn't?

1 reply

Kenji Aoyama·Jul 20

Then you're sized wrong, not wrong on direction.

1 reply

Lena Brandt·Jul 20

That's the distinction I was reaching for.

1 reply

Marcus Bell·Jul 20

Agreed. Leaving it there.

Amara Okafor·Jul 20

Yes, and I'll post an update if that level goes.

Elliot Vance·Jul 20

Good timing — was just reworking my DXY sizing this morning.

3 replies

Bruno Ibrahim·Jul 21

This is the right question. The answer changes the whole setup.

Nils Okafor·Jul 21

That's the part I keep going back and forth on.

Aisha Sandoval·Jul 21

Fair, though I'd want to see it hold for another week before sizing up.

Nora Aziz·Jul 21

Reasonable thesis but I think DXY is priced for most of this already.

Ben Grimaldi·Jul 21

Been waiting for someone to write this up properly. Thanks.

Nadia Rossi·Jul 21

What would make you exit this?

2 replies

Priya Nair·Jul 21

I'd push back slightly: the base rate on that is worse than it feels.

Ines Okonkwo·Jul 21

I'd push back slightly: the base rate on that is worse than it feels.

Amara Okafor·Jul 21

Fair pushback. I'd agree if positioning were more extreme, but the flow data doesn't show that yet.

3 replies

Clara Mercer·Jul 21

Fair, though I'd want to see it hold for another week before sizing up.

Nadia Rossi·Jul 21

That's the part I keep going back and forth on.

Viktor Reinholt·Jul 21

Same read here. The risk is it takes longer than anyone wants.

Hana Mercer·Jul 21

Any read on how this interacts with the broader DXY complex?

Amara Okafor·Jul 21

Sizing is roughly a third of a normal position until the thesis confirms.

2 replies

Callum Bradfield·Jul 22

Good point — hadn't looked at it that way.

Dmitri Nakamura·Jul 22

Fair, though I'd want to see it hold for another week before sizing up.