NVDA + Jackson Hole = pin trade

NVDA

We have a series of theoretically exciting events coming up, but my feeling is that we are more likely to go nowhere over the next week in G10 FX. Throughout most of my career, I was always biased to look for directional trades and exciting catalysts and imminent large moves but over the course of the last 10 years, I have become more willing to embrace the honest, but less-exciting truth that quite often markets go nowhere and you can make money expressing this view just as easily as you can make money betting that there’s going to be a huge move.

In other words, I have gone from structurally and synthetically long vol to something closer to neutral. The “nothing is going to happen” view doesn’t work that great on Twitter (or in a paid newsletter), but my experience is that honesty and unbiased views are worth more to my subscribers than razzle dazzle.

If you think about the coming week (now until 02SEP, which is next Wednesday), there are some decent events on the docket. Let me go through them one by one.

NVDA today. Nvidia has been an exception to the aphorism that trees do not grow to the sky. It looked like a potentially 1999-style overblown picks and shovels play, trading at 40X sales in 2023, but it grew into its valuation and now trades at “only” 20X sales.

The more relevant point for trading, though, is that as the company has grown into its valuation, the volatility around earnings has dissipated significantly. This makes sense as the story was not well understood in 2023 and it is extremely well-understood now. Expected moves have gone from 9%/10% to 5.4%.

And as that expected move has come down, the realized vol has underperformed the implied.

Markets have become inured to soggy NVDA price action through and after earnings.

And naturally, as NVDA earnings become less of a watershed event for the stock, the rest of macro shrugs, too. See this:

I used yellow boxes to show times when macro went the opposite direction suggested by NVDA, but the real takeaway here is that the macro impact of NVDA earnings has declined and has become close to noise. There seems to be some small impact on yields, and obviously the indexes move mechanically if NVDA moves. USDJPY is not listening to NVDA.

So that’s NVDA: Probably not much market impact on macro. If you must trade NVDA, I would sell put spreads because the market is very used to the stock going down after earnings and so the base case here is that the options market is still pricing in too much vol and the end result is NVDA closes something like $215 tomorrow at 4 p.m. after whipping around and mostly going nowhere.

The next big event is Warsh at Jackson Hole. There are two types of Jackson Hole confabs: The big market moving monetary policy ones, and the boring academic ones. Given Warsh’s preference to avoid forward guidance, there is almost no chance this becomes the old “we are signaling a hike in September” type of thing that was common in the Draghi, Bernanke, Powell eras. The title of the conference “Financial Innovation: Implications for Payments and Policy” doesn’t lend itself to any monetary policy surprises, either. The skill move from Warsh would be to present a 15-minute speech about how new payment rails will impact the global and USD payment systems in coming decades.

While Warsh did the performative hawkish thing at the June FOMC meeting, the data has not cooperated and so for him to reprise that inflation-fighter vibe is hard to imagine. But for him to turn dovish so quickly looks pretty stupid, especially with the Treasury attempting to rein in back-end yields. So net/net, I think Jackson Hole will go down as a non-event.

To give you a sense of past Jackson Hole outcomes, here’s a graphic.

Most of these moves are on Friday, but 2015 is Saturday, 2020 is Thursday.

On top of Jackson Hole, you also have corporate month end and real money month end, but these should roughly offset as corporates buy USD tomorrow and then real money sells USD early next week. Stocks are up 3% or so in the United States, which is about the right size of a move to offset corporate USD demand. It’s a push. Meanwhile, bonds are doing nothing despite all the angst driven by round number bias as the U.S. debt crosses the arbitrary $40T mark. Bonds got through three auctions and a lot of media hype and are still in a tight range.

While it’s not particularly sexy or directional or macro, I like selling 1-week vol in EURUSD and USDJPY. There is a simple and limited-loss way to do so.

Sell 1-week 1.1650 EURUSD straddle and buy 1.1600 put and 1.1700 call. You receive around $350,000 per 100 million EURUSD and your worst case is minus 50 pips, so the max loss is $150,000. This chart shows the P&L. And similarly…

Sell 1-week 159.30 USDJPY straddle and buy 158.80 put and 159.80 call. You receive around $250,000 per 100 million USDJPY and your worst case is minus 50 pips ($313,000), so the max loss is $63,000.

Equity options traders have cool names for structures like this (e.g., iron butterfly) but in FX we are pretty boring, so we don’t. I will call this “the pin trade” as my dream is that both EURUSD and USDJPY pin the short strikes next Wednesday.

---

Final Thoughts

1. There is an absolute FLOOD of AUDUSD and AUDCAD bullishness this morning. This does not mean it can’t go up, but it’s good to know. My emails and chats are experiencing a three standard deviation influx of bullish AUDUSD and AUDCAD ideas. The CPI last night and the upcoming dividend flows are part of it, plus it has simply traded fairly well in the face of falling Chinese growth and soggy iron ore prices. Feels like a risky entry point with this much bullish cacophony.
2. Interesting chart from the Financial Times.

The kids from Alabama won again yesterday and they play another elimination game today at 7:00 p.m. Canada plays at 5:00 p.m. In a world where sports coverage feels like all money all the time, late August is the time you can still see people playing a sport for the love of the game in Williamsport.

The English word “hap” means chance, luck, or an unexpected occurrence. From the Old Norse word “happ” meaning “chance” “good luck” or “fortune”. Common derivative words include:

Happy (favored by hap)

Haphazard (using a derivation of the Arabic word al-zahr (or az-zahr), meaning “the dice”)

Hapless (unlucky, lacking hap)

Mishap (bad luck)

Perhaps (Inshallah)

Happen (to occur by chance)

Happenstance (a blend of luck + circumstance)

 

Hap Tip : Alvin Chopra

good luck ⇅ be nimble