Riding the Volatility Short Bus

There’s been a bit of a hullabaloo of late regarding the resurgence of short volatility trades in equities. This comes at a time when volatility is low by historical standards, which has led some to conclude that the shorting of volatility is causing the low volatility.

Wrong, wrong, wrong–or at least the correlation between the two is insufficient to demonstrate causation. This is another example of the error of reasoning from a price/price change.

The “logic” here is particularly dubious for derivatives (“selling vol” is a synonym for selling options) because they are in zero net supply. Quantity bought is equal to quantity sold. So “short volatility trades come roaring back” also means “long volatility trades come roaring back.” Determining what drives what is not immediately obvious.

Here’s another story. Some believe volatility is to cheap and want to buy it. Or some what to hedge volatility risk, and at the current low levels it appears attractive to do so. So rather than pushing into the vol market, shorts are being pulled in.

You can’t tell which is happening just looking at the level of vol and open interest. And here’s the key thing: derivatives are risk transfer markets. What is determined in these markets are primarily risk premia. And so to see who’s pushing and who’s pulling, you need to look at risk premia.

The long pull story should see the vol risk premium rise, specifically implied vols should rise relative to expected future vols (in the physical measure). A crude proxy for this is implied vols rising relative to realized vols.

The short push hypothesis predicts the opposite. It doesn’t predict that the level of implied vol should fall absolutely: it predicts that implied vols should fall relative to realized vols.

And according to CBOE’s vol maven Mandy Xu, it’s the former and not the latter:

Second, if volatility selling strategies were to blame for the low levels of the VIX index, you would expect the volatility risk premium (VRP) to shrink as the implied-realized volatility spread narrows (i.e. the VRP is what option sellers aim to monetize and thus should decrease as more sellers enter into the market). Instead, what we’ve seen over the past year is the opposite — the S&P 1M volatility risk premium (as measured by the difference between the VIX index vs SPX 1M realized volatility) actually increased quite meaningfully, from 1.5% in 2022 to 3.6% in 2023. Implied volatility may be low, but it’s not trading particularly cheap compared to realized volatility (SPX 1M realized vol went from averaging 24% in 2022 to just 13% in 2023).

Xu also notes that the vol curve has been more skewed to the call wing, i.e., out-of-the-money call vols have been rising relative to ATMs. This could be explained by a demand to hedge against volatility spikes from today’s current levels (a scenario that played out in the “Volmageddon” of 2018). (Pay attention class–by which I mean my real, actual classes–we’ll be discussing that after Spring Break.)

So the short vol crowded trade driving down vols story is certainly logically weak and factually unsupported, and the alternative that the rise in vol-related derivatives notionals is driven by buyers does have support in the data.

Remember children: never argue from a price change. Especially in derivatives markets with arguments based on open interest/volume.

There does appear to be some interesting intermediation going on. There has been a lot of activity in individual stock options (e.g., Nvidia). It appears that some buying pressure has been accommodated by hedge funds selling those options, and buying index options. This is a form of spread trading (in a way not dissimilar from the infamous basis trades) that hedge funds specialize in. It would tend to contribute to higher notionals in index vol positions–both long and short. There is also spreading between short-dated (especially “zero day”) options and longer-dated options.

Regardless of the underlying driver here (index vol bets or individual stock vol bets) it does appear that the surge in volatility trading has been buyer driven, not selling driven, as indicated by the behavior of risk premia.

Sadly, it is the Bank for International Settlements (BIS) that has pushed the “vol shorts have pushed down implied vols (absolutely–but not relative to realized vols). So it’s fair to say that the BIS is riding the volatility short bus. Which is embarrassing.