In the week of January 25th, 2021, Gamestop (ticker symbol $GME) rose by several multiples due to a massive short squeeze. A synopsis of the saga can be found on Wikipedia here.
This caused a huge furore in the markets, bringing the term “short squeeze” into the mainstream financial social media. This phenomenon, where a stock rises by so much that people who have short sold a stock end up facing large mark-to-market losses due to a rise in the stock price and are forced to close their position at whatever the available market price, is not new – another relatively recent example was the infamous Volkswagen short squeeze (you can read about this here.
The mechanics are simple: a short seller has to borrow shares from someone else for a fee before they sell them on the open market, hoping to profit when they buy the stock back at a lower price. If the stock goes up after they lose they will have to buy the shares back at a loss, and depending on their margin arrangement with their broker, if they don’t want to buy the shares back they will have to post extra collateral to back the higher dollar value of their shorted shares. If they run out of margin collateral their position will be liquidated and forcefully bought back on market to close out their position.
There is also the risk of recall – since the stock has skyrocketed, the lender they borrowed from might want to sell their shares at these prices and ask for their shares back. This forces the short seller to buy back the shares to deliver, which in turn also pushes the market price higher.
By now there have been many articles online discussing how this phenomenon happened to a company like $GME. This article will instead discuss how an options market maker would react in response to a short squeeze scenario such as this and why it doesn’t apply to stock index options or cryptocurrency options on Deribit.
The Market Maker’s mandate
There is much misconception in the crypto space about what a market maker’s job is. A market maker’s job is to make markets (quote bid/ask prices) in securities/derivatives such that they profit from other market participants crossing the bid/ask spread while their own portfolio risk remains flat. This is much more easily achieved when you are able to trade a variety of instruments that have very similar risk exposure. For example, if you see that BTC trades at 35000 on Binance and you quote 34980/35020 on Bitstamp, someone lifts (buys) your offer at 35020 and you buy it back on Binance at 35000 – you make a profit of $20 (before fees) after you have hedged your portfolio. Of course, this is not free money – if you are unable to buy 35000 on Binance because the BTC price moved up you can lose on this strategy.
The upshot of this is that market makers effectively act to make markets efficient and are compensated for this by capturing these small temporary arbitrages between products.
The idea that seems to pervade cryptotwitter is that market makers manipulate the price by suppressing the price by going short large amounts of BTC at key price levels to change the market psychology to sell lower and buy it all back and go long again at a lower level – this is patently preposterous because you only have to be wrong about the market direction once and might go broke. As a market maker your mandate is to be as risk neutral as possible to protect yourself from the possibility of going broke. It’s possible that some market making firms might be willing to accept some degree of risk with a net delta strategy on top of pure risk neutral market making but the more they deviate from risk neutrality the less they would be able to truly classify themselves as a market maker.
Options market making
So how does this work in the world of options where options have so many expirations and strikes? As an options market maker, your ideal scenario is someone buys and sells you the identical option in quick succession, crossing your spread both ways. You then end up with no position at all and have collected the spread in profit – happy days.
This rarely happens though due to supply and demand. So what do you do if you are quoting an option and someone buys some calls from you?
Firstly, there is the delta risk of the option – the easiest and most liquid way to hedge this is by directly buying the underlying delta in the underlying stock or future.
Secondly, there is the fact that you have a variety of non-linear option risks in your portfolio associated with the fact that you are short a specific option. Without going into a discussion of option portfolio greeks (measures of various aspects of risk associated with the nonlinearity of option payoffs), we will just note that this is where spreads come in! You can understand this better in our article here.
As an options market maker, your objective if you cannot unwind your specific position is to end up with a portfolio of successively long or short option strikes, as close to each other as possible, so that your position ends up looking like a big string of butterfly spreads and therefore your overall risk is limited. With a long time to expiration, the butterfly spread prices don’t change significantly on a move in the underlying, and so you can wait until closer to expiration when the option premiums are lower to close out your strike specific exposure.
So what does this mean on a practical level for options? If someone buys a lot of one option and therefore market makers net sell (say the 36000 Feb BTC call), it’s unlikely that the market maker can immediately buy back the Feb 36000 call, so it’s likely they try to buy either the Feb 35000 or the 37000 call to partially hedge. Depending on the net options flow against market makers, the price of all options will move somewhat in tandem as a result as the market makers balance out their risks. If they can’t buy enough Feb options to hedge then they might start buying March options, and so on such that Feb changes can impact the price of other expirations and vice versa.
Short squeeze impact on options MM
For equity options such as $GME in a short squeeze scenario, options market makers have additional risks:
- Stock moves around very quickly so if you quote options tight then you can very easily miss your delta hedge price by a long way and lose money.
- The stock is very volatile, so the correct implied volatility is very high (IV on $GME Jan 29th calls was over 900% annualised on Jan 28th). In that situation, if you are net long options at that implied vol level, you need the stock to realise that level of volatility until expiry until expiration otherwise your position will net lose on burning theta. Conversely, if you are net short options there and the stock does move more than that IV level implies, you will lose more money than you collect in theta. As a result, the only correct position is as flat options as possible.
- The higher volatility means that other market makers are also unwilling to quote tight, which makes it much harder for you to get a good options hedge for any trade that you do. This makes it very hard to be flat options.
- Since the stock has rallied so much, you may actually run out of listed option strikes to hedge. So for example stock starts at $75 and the highest strike option is $300. You happen to be short the $300 strike option, and the stock rallies to $350 overnight. You will be short that payoff convexity with no way to get it back until new strikes are listed.
- If your net options portfolio delta is long, in order to maintain a flat delta you need to be short the underlying stock. In a short squeeze scenario it may be impossible for you to borrow the shares, and even if you can find the shares to borrow the borrow rate will skyrocket. This means that it will be difficult to trade short delta contracts vs long delta contracts (e.g. synthetic futures will have to drop in price to reflect the difficulty and cost of going short stock). If you happen to be long synthetic futures already in your portfolio then the mark to market value of your futures will drop and you lose money.
- If the options are American options (so the holder can exercise the option early before expiration, as is the case with US stock options), holders of the call options can exercise them early to get stock back. This means if you are short the call option you will have to deliver the stock when this happens, and if you don’t have the stock to deliver you will have to get it somehow while maintaining delta neutral. This contributes further to the short squeeze.
Note that for Deribit crypto options, points 5 and 6 above are generally not an issue, because the options are European and also because there is already plentiful liquidity in BTC and ETH futures and swaps across a number of venues that make an equity-style short squeeze relatively difficult to engineer.
This was a brief discussion on what options market making is and how a Gamestop-style short squeeze impacts options market makers in general. Some market makers may have managed to make great profits in these conditions and some may have lost their shirts, but either way the trading environment is highly challenging for many reasons – have a heart for them in continuing to provide liquidity for the market in the face of these conditions rather than turning their machines off!