The volatility index "VIX" has been plunging lately. In fact, it is in the midst of a historic collapse. And it's leading people to conclude that a market crash is right around the corner. Is that a fair assumption? First, the details.
Over the past six weeks following Brexit, the VIX has tanked 55% and fallen each consecutive week. It's the first time since the financial crisis and only the fifth time ever that the VIX has fallen six consecutive weeks, according to Schaeffer's Investment Research.
Notably, the last time this occurred was in August of 2008. We all remember what happened just after that. The other three occurrences of this pattern in VIX brought mixed results. A long VIX losing streak in 2001 brought in a 10% market correction over the next three months. The VIX indicator was a decent market timing signal in this situation. However, similar declines in VIX in 1992 and 2003 failed to trigger any reaction; the S&P 500 (NYSEARCA:SPY) continued to advance in both those cases.
Regardless, it's a mixed and limited data set that includes two big market declines. And there's certainly intuitive reason to think that a relaxed or overly calm market is more likely to be hit by an out-of-the-blue correction.
However, I think people give the volatility index way too much credit. It's often called the "fear gauge" and is expressed as though it were some near-magical indicator. But that is to give it power it probably hasn't earned.
The VIX is, simply, a mathematical calculation of option prices. Specifically, it takes a weighted average of prices for S&P 500 index call options and put prices over the next thirty days. It blends these pricing inputs to create the VIX figure - the one you see on CNBC - that indicates how much volatility the market expects over the next month.
People (or rather, institutional funds) tend to buy S&P 500 options, particularly puts, as insurance in case of a big swing. Thus, higher demand for puts (and calls to a lesser degree) will cause the VIX to rise, while softer demand will cause the index to fall.
We've had the VIX for many years now. However, only since the financial crisis have people elevated it from a rather arcane market statistic to being a supposedly top-notch sentiment indicator. There are several problems with VIX.
For one, there are numerous seasonal factors that affect VIX. Volatility tends to be lower when market volumes are lower. Around holidays and in the summer. Christmas in particular leads to lower volatility and the well-known December "holiday effect" where VIX is usually near its 52-week lows. There are also some issues with VIX due to the way options pricing work. It's technically true to say people are less "fearful" because it's the holidays, but clearly, that's not what a proper "fear gauge" would be measuring.
However, the biggest issue is that VIX is simply not designed to measure fear. It's far from clear that put and call options pricing is the best way to determine market sentiment. Volatility very specifically measures one thing - if there is more volatility (wider range in S&P 500 prices) over the month, then VIX buyers win. If there is less, VIX buyers lose. It's a very specific question that VIX measures, and there's great risk in generalizing that one pricing question into a broad statement of market health.
Right now, VIX is extremely low. People are suggesting this indicates that a crash may be coming. However, the simpler explanation is that VIX is really low because the market has experienced really low volatility lately. Look at this:
Over the past month, the market has done absolutely nothing. That's a basically 1% trading range for an entire month. Excluding Brexit, nothing worth hedging has occurred since February. Actual realized volatility (the way the formula is calculated) is in the mid-single digits now. The current 11 VIX reading suggests market participants think market volatility will almost double from what we just observed over the past month.
That's probably a fair assessment. It's unlikely markets will be as quiet as July. However, until Labor Day, there's no reason to expect any sudden drops either. The VIX curve reflects this, with August and September VIX at very low levels, while the rest of the board is at more normalized levels:
There's nothing in this board - at all - that suggests the market is being irrationally complacent. It is still pricing in a swift move upward in volatility from what we witnessed in July. Yes, 12 is a low figure; however, look at the market lately. Remember, volatility pricing is set by S&P 500 options - buyers of S&P 500 options are getting killed as the market trades sideways for weeks on end. It's no wonder volatility reflects the market you currently have - real money is changing hands.
At some point, the market will get more sharp again, and volatility will rise to reflect that. But I've seen little indication that volatility is the leading indicator; it seems to generally follow what the rest of the market is doing. As market conditions change, people start buying more insurance, and VIX starts rising. But there are many indicators that can indicate changes in market behavior, VIX isn't necessarily the best.
Let me present today's VIX curve against this same August day three years ago, in 2013:
As you can see, both today (green line) and August 2013 (black line) shared a similar shape. We actually have higher volatility from October onward today than we did in 2013 - this market is more concerned than 2013 was about the coming two quarters. This market has slightly lower volatility in August than 2013 - because our market has been asleep for a whole month now - but on the face of it, these are very similar VIX curves.
2013 saw the market breaking and running to new highs, just as 2016 is now doing. If you looked at this very low VIX set-up in 2013, here's what happened next for the S&P 500:
The 2013 market did dip a bit to close out August, but there was no big correction as a result of the low VIX figure. The S&P 500 was able to continue higher without issue. For buyers of VXX on that day in August with the same volatility curve as we have now, here's what happened:
Yes, there was a bit of a pop in VXX, but it was nothing dramatic, and by year-end, VXX was another 25% lower.
I see people suggesting that going long volatility is a "no-brainer" and that volatility can only rise from here. And it's true, the VIX is unlikely to stay at 12 for all that long - it tends to mean revert to 15 fairly frequently. However, that won't make money for VXX or the other volatility ETFs - they constantly erode due to the contango in the VIX futures curve you can see above. With contango nearly at 20% a month at present, VXX will lose approximately 20% of its value over the next month (and UVXY almost 40%!) if the S&P 500 merely trades sideways.
VIX is not a proper fear gauge; don't assume the market is going to plunge simply because VIX is low. I expect a correction before year-end, but that's independent of VIX. I believe VIX has minimal value as a leading indicator.
As for the long volatility trade (NYSEARCA:UVXY) (NYSEARCA:VXX), there are two correct positions: short, or not short (there's an exception my subscribers know about, but even it doesn't involve buying VXX or UVXY). Going long is never the correct play as long as contango is in effect. Despite the volatility ETFs having just taken a pounding and being forced to reverse split, they will continue heading lower in coming months.
Disclosure: I am/we are short UVXY.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.