Fear, Greed, Volatility and Opportunity

Here's a thought. An algorithm is coming soon too.

Fear is well....a simple word in English which has the same meaning in finance / economics. Greed, as Gordon Gekko put it, for the lack of a better word, is good. States of mind as they may be, we can think of a collective measure of fear / greed across all market participants. This is the volatility index or the VIX. While it is computed using option prices, and a bit of complicated math, it suffices to say that it does a pretty good job as the index of fear.

There is something however, which is not as reliable as the fear index, and that is fear itself. While fear is a part of the survival tool kit that humans carry, it turns out that most of the time, the amount of fear in the minds of market participants seems excessive, especially in equity markets. What do I mean by excessive? I mean that, more often than not, what people are afraid of doesn't materialize. In addition, fear seems to rise (sending VIX surging) on reasons that are less than well founded. What am I getting at though? A few simple truths, and as a biproduct, a trading strategy.

VIX is meant to measure people's expectation of future volatility and it does. Just that people sometimes expect unreasonably high volatility going forward. This expectation of the apocalypse, is usually triggered when the underlying (S&P or FTSE or any other index) sees a sudden crash in price. The crash is sometimes driven by news, sometimes by rumor and sometimes by nothing. Is this rational? It is, to the man on the street. It shouldn't be to someone who understands the markets. A crash in traded price of the index is an event in the past, however near. It shouldn't affect the entire outlook of the future.

Let's plug in some numbers to understand this better. The historical average of VIX tends to be around 18%, which is close to but higher than the historical, annualized volatility that is realized - 16% (standard deviation computed using daily returns). Assuming 250+ trading days in a year, this transalates to 1% volatility per day {daily volatility = (annualized volatility) / (square root of the number of trading days in a year)} . Let's say the VIX jumps to 32% in a couple of days. For this kind of fear to get realized, the daily volatility should become 2%. That's a very low probability thing. Even if it does happen (black swan event), it is unlikely to sustain for more than a couple of months or so.

Why is high VIX unsustainable? As mentioned earlier, VIX eventually should come close to converging with realized volatility, to be a meaningful measure. Because of this constraint, VIX exhibits a mean reverting behavior. It is a statistically safe bet (or statistical arbitrage) to count on VIX to return to its mean. It can get higher before turning down, it might take time, but it is almost guaranteed to revert to its historical mean. But, as I explained earlier, VIX has this unreasonable (yet statistically sound) connection with price fluctuations in the near past. So, if VIX has to go back down from unrealistically high levels, it takes the underlying to trade higher and higher. So if you happened to buy the underlying when there was blood on the street, mean reversion in VIX has made you wealthy by the time fear has subsided. But do keep in mind. This may not work for individual stocks. Individual stocks have bounced back in the past. However, there are many cases of cheap going cheaper. Also, do remember that even with the index, things might get worse before they get better. So it is important to be willing to cut your losses somewhere.

Great! When is a good time to exit though? Honestly, I don't have a one size fits all answer. VIX unfortunately hasn't gone too far below its historical average. Nor does VIX always show smooth trends in the long run. So the exit trigger is probably something other than VIX. It is also a function of the trading / investment philosophies and goals.

Is there any other way to profit from the behavior of VIX other than muster the courage to love the smell of blood? Indeed! The behavioral economics around implied volatility leads to a big bunch of interesting, theoretically valid statistical arbitrages. But here are some things that need to be paid attention to, before executing on these opportunities.

a) Understand what creates the anomaly.

b) Explain why the anomaly has to get corrected.

c) Prove the existence of the anomaly and its correction by back testing.

d) Paper trade the opportunity.

Interestingly, you can use neural networks and other machine learning techniques to understand these behaviors and craft trading strategies.

Keep in mind you can't trade the "underlying" when it comes to VIX, only its futures. That makes for a very frustrating experience for all of us who at some point realized the spectacular opportunities that you've outlined, then come to the frustrating realization that there's not a great way to exploit it. Eventually you see that you're faced with the brutal headwind of the normal time decay in the VIX futures, which is unfortunately almost exactly balanced by the occurrences where the futures make a huge jump and sit in backwardation for a while, so you can't reliably profit on either side of the trade over the long term. That doesn't keep me from returning to the idea and reading any posts that come along with interest though, there's got to be something in there somewhere!