Gamma Rays...and Falls

Well.  The scariest night of the year is almost upon us (take your pick of October 31st or November 3rd), and it definitely feels more trick than treat in the markets this week, doesn't it?  Monday was off a couple percent, no bounce on Tuesday, and then down another 3.5% Wednesday.  A little bounce yesterday, but all given back again so far today.  Frustratingly, Wednesday's selloff was another one of those days where literally everything sold off.  Stocks, bonds, gold..everything.  To put that into perspective: stocks, bonds, and gold all down on the same day has happened twice.  Ever.  And both those times were back in March of this year.

Which kind of makes sense...the catalyst for the selloff is a little murky, though it seems like Europe (at least Italy, Spain, France, Germany, UK, Belgium, Czech Republic, and Poland) reinstating various lockdown measures is largely to blame.  Domestically, optimism around a fresh stimulus deal seems misplaced (though does that actually surprise anyone a week before the election?), and the election races seem to be tightening.

A few weeks ago, the default positioning on elections felt like expectations of a "Blue Wave" result next week, and markets had been rising ahead of an expected proverbial opening of the government spending floodgates.  Now, it feels like there is a reconsideration of that outcome.  Markets hate uncertainty.  The worst possible outcome (for markets) is a contested election that has to spend months working its way through the courts.  Tighter polls make that outcome marginally more likely, so perhaps what we're seeing in this selloff a broad deleveraging of risk assets.  Or maybe it's the opposite of what we saw back in Q2 and is a bit of a negative gamma squeeze.

In the spirit of Halloween, we'll take this newsletter in less of the obvious Bruce Banner/Incredible Hulk direction and more of a The Gamma People/Creature with the Atom Brain direction as we dive into what exactly gamma is and how it moves markets.

Gamma is essentially a second derivative of stock prices.  Okay, let's back up a second.  We've written about options before.  Options are contracts to buy or sell 100 shares of an underlying stock at some set price at some set point in the future.  Options can be considered a type of leverage, since you are able to effectively control 100 shares of some stock for significantly less than the price of 100 shares.  (As an example, 100 shares of Apple would cost you about $11,000 as of this writing.  An at-the-money monthly call option for 100 shares of Apple is less than $500.)

It used to be that options were primarily used for hedging.  Asset managers would go long a stock (or short), and then could use options to hedge the position.  Like, long the S&P500 but also buy some put options in case the market goes down.  

But something interesting happened earlier this year - back in July, for the first time ever, the volume of options trades was greater than the volume of actual stock trades.  A lot of this seems to be a knock-on "RobinHood" effect of new day traders.  We wrote about the massive increase in retail trading, and it seems like the options market is no exception.  Fun fact: as of this summer, 13% of all options trades were for 1 contract.  Stimulus checks were for $1,200.  One contract of Apple is about $500.  Just sayin'.

Anyway, while retail money was getting into options for the cost leveraging aspect, there was a whale that was also moving into options in a big way, but for a slightly different reason.  The whale was Softbank, and the reason was gamma.

Let's back up once more.  The change in the price of an option contract given the change in price in the underlying stock is called delta.  If the option is way far out of the money, delta will be small (close to 0), because the price of the option won't move much at all even if the stock price itself bounces around.  If the option is way far in the money, delta will be closer to 1 and jump around basically as much as the stock price itself.  For options that are at the money, delta is usually around 0.5, meaning the price of the option moves about half as much as the price of the underlying stock.  

Gamma is the rate of change of delta. 

Remember the crazy market meltup this summer, especially in the tech names? Here's how Softbank did it:

Step 1: Buy a lot of stock in specific tech names    
Step 2: Buy a looooot of short-dated, out of the money call options on those same names.  Like, $1B lot.  Keep in mind this is the options market, so that $1B is probably equivalent to $20B or so in stock value.
Step 3: Then you open the box.

Kidding.

Actual Step 3:  Sit back and watch the profits roll in.

What profits you may ask?  Well, Softbank's massive call option buying effectively left the various dealers/market makers that sold them the calls in the first place short.  To hedge this short position, they had to buy the underlying stock (at a ratio determined mostly by the delta on the options they sold to Softbank).  But the size here was such that the amount of buying pressure from the hedging that needed to be done pushed the price of the stock up.  Which raised the price of the call options, which increased the delta of the options positions, which necessitated more stock buying to hedge the position, which raised the price of the stock, which raised the price of the call options, which increased the delta, which necessitated further stock buying...if you give a mouse a cookie, as they say. 

The subtle brilliance here was that the names Softbank was primarily running this strategy in make up about 40% of the NASDAQ.  So the hedging on the part of the dealers could be done not just in individual names, but in the index as a whole!  And so you saw the runup "spill over" into the broader market, even though it was concentrated in a few big tech names.

That yellow line is the NASDAQ year-to-date.  Purple is S&P500, blue is Dow Jones.  That spike through August into September is what we're talking about.

You've had some attempts at copycat trades once this strategy was revealed, but it's tough to do if you're not Softbank-esque in size.  The concern that we have is the reverse scenario: say that Softbank wants to hedge against uncertainty in the election outcome.  If enough protection is sought, dealers are left "long" the market.  Which they hedge by - yep, selling stocks.  Which raises the price of put options, which makes delta go up, which necessitates more hedging (by selling this time), which drops the price of stocks, which makes put option prices go up, which...you get it.    

Now, suppose it's not Softbank, but rather everybody.  All those single-options-contract traders buy a put to hedge a market drop post-election.  Cumulatively, it has the same effect.  There's a fair bit of irony in the idea that everyone hedging against the market going down actually makes the market go down.  It's like...waiting seven years for the perfect full moon Blue Moon Saturday Halloween…and then getting Covid. 

Sigh.  Oh well, ghost stories and horror movies galore tomorrow!  Happy Halloween everyone!