The main factor behind the rally off the March low, at least if you listen to the media, is that both earnings and economic releases have not been all that great, but at least they've beaten expectations.  It's that whole going-from-bad-to-less-bad thing that often signals a turnaround.

We can very clearly see that dynamic in the following chart.  It shows the percentage of S&P 500 companies beating earnings expectations (as compiled by Bloomberg) and also the Citigroup Economic Surprise Index which is a three-month average of the number of worldwide economic reports that have come in better than estimates.

Both data sets show a major recovery from their troughs, and have actually rebounded back towards the upper ends of their ranges.  In fact, neither one tends to get much higher than the readings they recently recorded - there's a bit more room, but not much.

The red dots on the S&P price line show other times when both series were at levels equal to our current juncture.  During the bull market, the S&P dipped over the next few weeks or so each time, but obviously rebounded quickly thereafter.  The only instance during this bear market was in the summer of 2008 right before the "fit hit the shan".

With a recovery in earnings and economics (again, relative to expectations), it's no surprise that implied volatility has decreased as well.  As traders become more comfortable with the future, they see less need to protect against future uncertainty.

That, of course, moves in cycles and proves to be a reliable contrary indicator at extremes.  Yesterday was interesting not necessarily because of an extreme, but because while major indices like the S&P 500 were busy moving to new two-month highs, the VIX index of implied volatility did not move to a new two-month low.

We've touched on these kinds of divergences in the past without much of a conclusion, but it's always a popular topic for questions, so let's take another look.

There have been 20 times when the S&P was in a bear market and we saw this kind of divergence.  Three days later, the S&P was positive only 7 times, averaging a return of -0.5%.  That's worse than any-time returns during a bear market.

If we look at the two primary bear markets since the inception of the VIX, it's pretty clear that these divergences have led to sub-par performance going forward.

First, the 2000 - 2002 bear market:

I included the few months leading up to the actual peak to show that these conditions existed right at the peak as well as during the meat of the decline.

It's a similar scenario for the current bear as well:

These setups were fairly rare during the bear markets - but probably not because the divergences are so rare, mainly because it's pretty tough to reach a two-month high during a bear market.

The reason I suggest that is because when we look at non-bear market periods, the number of these divergences explodes.  If I showed the bull-market periods, then I'd be sitting and drawing those little red arrows on the charts all the way through the weekend.

Since the inception of the S&P 500 SPDR (SPY), there have been 423 trading days when the S&P hit a two-month high but the VIX did not set a two-month low.  It was almost as common to see days with this divergence as without from 1995-1999 and 2003-2006.

 Disclaimer  |  Privacy Policy