Why I Sold On Tuesday's Open, And What It Will Take To Reenter

|
Includes: SPY, VXX
by: Michael Gettings
Summary

I’ve found no better indicator of negative market sentiment than the shape of the VIX futures curve and the rate of change in that shape.

Monday night, while VIX futures were still contango, the degree of contango was fading fast. That is, futures quotes were still rising with longer-dated expiration, but the curve was flattening.

An algorithm calibrating the predictive value of changes in the curve against S&P price performance, indicated a move to cash upon Monday's close.

By Wednesday, the curve was backwardated; the negative sentiment had accelerated.  Losses avoided by the early exit have exceeded 2%.

I'll look for a change of slope (rate of change) as the curve moves from its backwardated shape toward contango in coming weeks; that will signal a reentry point.

Editor's note: Seeking Alpha is proud to welcome Michael Gettings as a new contributor. It's easy to become a Seeking Alpha contributor and earn money for your best investment ideas. Active contributors also get free access to the SA PRO archive. Click here to find out more »

On Monday, May 6, the Dow was down 66 points after starting the morning 400-plus to the downside on China trade worries. More importantly from my perspective, the contango shape of the VIX futures curve made a significant move toward backwardation. On Friday, each of the first three monthly VIX contracts traded at a 5.2% average discount to the subsequent month's contract. By the close on Monday, that discount had fallen to less than 1.8%, and now it is backwardated. Monday's rapid change triggered a sell signal according to my algorithms.

Here is a graphical representation of the degree of contango since the beginning of 2019; negative values indicate backwardated conditions and the black arrows are inserted for emphasis.

Early 2019 has been predominantly contango: Source: Michael Gettings, with data taken from VIXCentral.com

The prevailing contango shape is a normal one - i.e., quotes rise with longer-dated expiration. It typically reflects a healthy environment for rising stock prices. A rising degree of contango is more bullish still, and that comports with the bullish environment of early 2019.

I've found that tracking the contango/backwardation of VIX futures contracts, and focusing primarily on the rate of change in the shape (slope), provides a strong indication of market sentiment.

While the VIX futures curve is normally contango, as equity risks rise, the shape of that curve moves toward backwardation. Applying a calibrated algorithm to this observation has predictive value regarding potential market declines. I use an algorithm that triggers conditionally when a short-term slope (Primary Slope) exceeds a negative threshold, and I look for confirmation when a mid-term slope (Confirming Slope) also exceeds another threshold. When the Confirming Slope also drops sufficiently, it translates to a sell signal.

There are two periods on the early-2019 graph when the rate of change is decidedly negative. The first one is in March and the second is right now. For the rapid decline in March, as indicated by the first black arrow, the algorithm triggered a "sell on the opening" of March 22 with the S&P at 2,845. Two days later positions were reestablished on the opening with the S&P at 2,813, avoiding a 1.1% decline. The 1.1% is not what I aim for, but sometimes small drops morph into big ones, and if I'm not foregoing big gains, I'd rather be safe in such times. I'll have more on that in a moment.

Time will tell if the current sell signal pans out, but conditions and history indicate it's a good probability. To gain confidence in the methodology, I've been running the algorithm for twelve months now, and here are the results:

Six "go-to-cash" periods in the last 12 months:

Source: Michael Gettings, with data taken from Fidelity

The yellow areas represent periods of cash holdings because migration of the VIX curve indicated sell. The methodology identified six "yellow" periods. Five of those six periods produced a net benefit. One period forewent a gain of .2% that otherwise could have been captured by holding stocks. Four periods successfully avoided losses between 1% and 3%, while the December 2018 "yellow" period avoided a 7 % loss by going to cash.

Again, I'm not targeting small single-digit benefits, but if the yellow periods do not forego material gains, I'd rather move to safety. As it looks now, the small gains do add up to favorable impacts. At the end of the year, the managed portfolio produced materially better results than the S&P as indicated by the graph. I would like to back test this methodology over a longer history, but have yet to find a user-friendly source of VIX futures curves.

So, when would I buy back into this market? The Primary Slope referenced earlier is currently at about negative .7% and still falling, well below the threshold for reentry which is close to zero. I would expect it to take a week or more for the "healing" of that metric. I'll track it daily and provide an update when appropriate.

I view the explicit decision-making parameters as proprietary, but I hope this discussion provides sufficient information to be of interest. The methodology seems to hold real promise. Of course, past performance provides no assurance of future results, but the market‑timing prospects look to be very strong. Many believe that market timing is a fool's errand, but if you're not so sure, please follow me and I'll provide periodic updates on performance.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. 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.

Additional disclosure: I trade SPY and other broad ETFs based on the methodology described.