Schaeffer's Trading Floor Blog

The VIX and the Case of the Missing Bulls

AAII data is consistent with what we've seen with the VIX

by 2/7/2014 7:30 AM
Stocks quoted in this article:

Amazing what a couple of weeks will do to sentiment. This, from Bespoke:

In the last week of 2013, bullish sentiment on the part of individual investors rose to its highest levels since January 2011 (55.06). Now in the span of just six weeks, the bullish camp has been halved. According to the weekly survey of bullish sentiment from the American Association of Individual Investors (AAII), bullish sentiment declined from 32.2% last week to 27.90% this week. This represents the lowest reading for bullish sentiment since last April.

The volatility in this data series is pretty intense. It wasn't that long ago we were seeing multi-year highs in bullish sentiment.

Now, I wouldn't get carried away with any one indicator. AAII conducts a survey; it says nothing of what its members are actually doing. I tend to go more with numbers that measure where money is actually going as opposed to where it says it might go.

Having said that, it's certainly a sentiment consistent with what we've seen in the volatility markets. As we noted earlier in the week, VIX-Everything is booming in terms of both volume and the price of what everyone's actually buying.

VVIX is the most confusing index ever invented. It measures the volatility of CBOE Volatility Index (INDEXCBOE:VIX) options. Vol of Vol! It's a little misleading at times, given that there's a large skew in VIX options. The higher the VIX strike, the higher the implied volatility. VVIX uses the VIX methodology to calculate itself. Thus, as VIX itself rises, higher-strike VIX calls become closer to the money, resulting in a greater weight in the VVIX calculation. As such, VVIX will generally lift as a function of VIX itself rising, regardless of whether VIX options have actually seen an increase in their implied volatility.

Got that? You won't be tested. Nickel version is that a rise in VIX in and of itself causes a rise in VVIX.

So, setting aside that disclaimer for a sec, the rise in VVIX is still impressive. From Jan. 23 to the VIX peak on Monday, VVIX rallied from 78.25 to as high as 109.67. It's safe to infer that investors have paid up for all those VIX calls that traded on Monday.

Throw it all together, and it's clear the public has flipped on the "Fear" switch over the past couple of weeks. It's a contrary tell, but they haven't exactly been whipsawed so far. Even with yesterday's volatility pullback, the iPath S&P 500 VIX Short-Term Futures ETN (VXX) is still up about 10% since VIX got overbought on Jan. 24. My "system" trade that goes long SPDR S&P 500 ETF (NYSEARCA:SPY) when VIX closes 20% above its 10-day simple moving average and then closes when VIX closes back below the 10-day would have closed just yesterday. The average holding period amidst a "Low VIX" backdrop is 5.4 trading days. This one took nine trading days, just short of the highest readings (11 days). It was successful 27 of the last 32 times, but not this one. It lost 0.79%.

It's tough to analyze these VIX pops without the benefit of hindsight. But early indication is that this is one of the uglier and longer-lasting shakeouts amidst an otherwise strong market. Hard to believe, seeing as so far, it's not that long-lasting or particularly ugly.

Obviously, it could ultimately become the first warning sign that the Bull has ended, but again, we won't know that for a while.

Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.


permanent link
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT

Partner Center

© 2014 Schaeffer's Investment Research, Inc. 5151 Pfeiffer Road, Suite 250, Cincinnati, Ohio 45242 Phone: (800) 448-2080 FAX: (513) 589-3810 Int'l Callers: (513) 589-3800 Email: service@sir-inc.com

All Rights Reserved. Unauthorized reproduction of any SIR publication is strictly prohibited.

Market Data provided by QuoteMedia.com | Data delayed 15-20 minutes unless otherwise indicated.