Like our prior set of charts, the following VIX graph from Birinyi Associates discusses a measure of sentiment hitting an extreme measure -- they call it "extreme volatility."
How is that defined? Anytime the VIX doubles over 50 days:
"Over the last 50 trading days the CBOE Volatility Index (VIX) is up over 100% which is an occurrence we have seen in only five other periods...The chart of the S&P 500 highlights the first day of each of the prior periods where this occurred. Each of these prior occurrences have coincided with market weakness, which was then followed by positive returns."
Let's look at their chart, which I annotated below:
The first two of these 5 examples of extreme volatility may have been followed by positive returns, but they didn't lead to a significant snapback anytime soon (left side of graph, un-numbered):
However, the 3 most recent prior examples of extreme volatility did, and I have annotated them with numbers 1, 2 and 3 (This week's example is marked "?").
Here's how they break down:
Point 1: August 1998 Sell off (due to LTCM blow up) led to much more upside
Point 2: post 9/11 Sell off led to a 40% plus Nasdaq rally, which ultimately failed and made new lows
Point 3: Late 2002 lows, led to a small pop, and a successful retest pre-Iraq Invasion
Point ? is this week.
Note that the prior 3 VIX events all had a response from the Fed of massively increasing liquidity: Rates were lowered, money supplies were increased. However, at present, liquidity is being removed from the system -- not added. That is consistent with our expectations for a small bounce 3-6 weeks, and not the beginning of a brand new cycle.
Short term, this VIX move looks parallel to three most recent versions; Longer term, however, I suspect it will have less legs than the 1998 or 2002/03 versions.
TrackBack URL for this entry:
Listed below are links to weblogs that reference Extreme Volatility:
I just read about that in the WSJ and came here because I wondered what your response would be. You make a very good argument.
I think this is one of those stats that leads nowhere. If you restrict the measure to x number of months the market went up y %. The value of x is chosen by determining whatever results in the highest y. More self-affirming data mining.
I'm not convinced things will get as bad as you predict, but the more I hear from the bulls, the more cautious I become.
There's another piece in the WSJ today about emerging markets being much better positioned financially now than they were in 1998. Which is obviously true. But the present problem is not their finances but their dependency on American demand for their exports. There won't be a melt-down, but that's not a reason to put money in markets that are still up 60% in the past two years and have the prospect of tightening margins.
The bulls seem to be throwing every possible justification out there hoping a few prove true.
Posted by: Bob_in_MA | Jun 15, 2006 9:59:43 AM
The comments to this entry are closed.