Sunday, March 18, 2007

The Bearish Equity Put-Call Ratio: What To Make Of Extreme Options Sentiment

In the wake of the sharp market drop on 2/27, the bearishness is so thick you can cut it with a knife. When markets are making orderly new highs--as was happening for months prior to the drop--I rarely get emails asking my opinion about market direction. As soon as the market plunged, however, I started getting requests from readers as well as media. Posts showed up in my inbox from True Believers heralding the end of the market as we know it. That got my attention. There are no foolproof strategies in trading, but fading market idiots comes pretty close.

But all of that is subjective, as a certain barefoot speculator would chide. What do the data tell us?

The equity put-call option data tell us that bearish sentiment is off the chart. Literally.

My data show that we have seen 12 of the last 15 trading sessions in which the volume in equity put options has exceeded the volume in equity calls. To put that into perspective, from January, 2001 through mid-February, 2007, we only had a total of 44 days (out of 1527!) in which put volume exceeded call volume among equities.

Moreover, we've had several days recently (February 27 and March 13 stand out) which have seen higher put-call ratios than *any* day during the 2001-2003 bear market. Indeed, during that bear market, we *never* saw a 20-day put-call ratio exceed 1.0, as it does at present.

What makes this surprising from my perspective is that all of this bearishness is occurring without us being down so much as 10% on the major averages. In my recent video, I review the price and money flow for eight major S&P sectors. From the vantage points of both price and flow, the recent drop so far has been no greater than any other correction during the past several years. But it certainly has engendered greater fear among options participants.

Going back to 2001, such fear has had bullish implications. When we've had a single put/call reading above 1.0 (N = 44), the next 20 days in the S&P 500 Index (SPY) have averaged a gain of 2.09% (33 up, 11 down). That is quite a bit stronger than the average 20-day gain of .16% (899 up, 628 down) for the entire sample.

If we just go back to 2004, we find 30 instances of equity put/call readings above 1.0. Twenty days later, SPY has been up by an average of 1.48% (23 up, 7 down). In Vegas, I'd take those odds.

I've seen a few historical comparisons between the current market period and past ones. Most of those comparisons look for similarities between this market and past market tops. In some ways, however, from a sentiment vantage point, this market reminds me of 1987 *after* the huge drop. It was such an anomalous event and major decline that it turned sentiment on a dime. In the Investors' Intelligence survey, for example, the percentage of bears to bulls shifted from 17:60 in August to 45:28 in October. It also led to further selling in the near term and restrained performance for much of 1988. Ultimately, however, we look back on that period as a time when extreme bearishness had it wrong.

Perhaps, however, the better analogue would be the market of the mid 1990s. Then, as recently, we had record low volatility. We then climbed to new highs in 1995, just as we made new highs in 2007, but occasional market drops took their toll on sentiment. For instance, we saw bears outnumber bulls in the Investors' Intelligence survey after market corrections in November, 1995; August, 1996; and April, 1997. This tendency for even relatively mild corrections to engender strong bearish sentiment was a major part of what kept that bull market going.

In the recent market we see a significant pickup in bearish sentiment in the AAII poll--from 27% bears and 58% bulls in January to 45% bears and 33% bulls most recently--but not so far in the Investors' Intelligence survey. If we look at weekly data for all listed options, we have to go back to July, 2004 to find an equivalent, elevated ratio of put volume to call volume. Before that, we saw similar elevated levels in February, March, May, and November, 2003--all good buying opportunities for investors.

One last observation. The VIX bottomed at 8.89 intraday on December 27, 1993. The market correction drove the VIX higher to 28 in early April, 1994 and a later correction late in 1994 took us back toward 20. From that point forward, we continued with sub-20 VIX levels throughout 1995 and into early 1996 as the market moved steadily higher. When we did finally break out to consistently higher VIX levels in March, 1996, it was in an environment of upside volatility, not market weakness. Moral of the story? A bottom in VIX does not at all necessarily imply a major market top. It hasn't taken much price weakness to get the bears growling. This time may be different and--as I recently stressed--I need to see dollar volume flows and solid price action show me that we can put in a bottom here. Still, I don't think market history is on the bears' side.

6 comments:

wcw said...

Sentiment indexes I like:
- the ISEE. Viz http://www.iseoptions.com/marketplace/statistics/sentiment_index.asp
- the Yale surveys. Viz http://icf.som.yale.edu/confidence.index/

On the ISEE, you'll see that non-marketmaker options trades are as bearish as they've been for a while, but they are hardly "off the charts". They've been as bad as that index has shown since its inception in October 2002, but the readings are not breaking to never-before-seen-lows. Viz http://www.bignose.org/~wcw/isee.16mar07.png

At the sentiment survey, you'll see very supportive numbers on valuation confidence, which is as it should be. Indexes are nowhere near as extended versus earnings as they were ~2000.

For what it's worth, you might consider making formal models and running some toy regressions, rather than doing your average-only analysis. With the ISEE, for example, I find strong statistical evidence for counter-sentiment trading -- but with tiny r-squareds and very small coefficients. There appears to have been a contrary sentiment effect these last few years, but it seems to have been awfully small.

Market history is indeed not on the bear side, but based on put-call sentiment alone it isn't really on the bull side, either.

Brett Steenbarger, Ph.D. said...

Hi WCW,

Thanks for the perspective on options. Do check out Jason Goepfert's most recent special report on option activity. He also is noting significant extremes in bearish sentiment--

Brett

Michele said...

Ah, the VIX. I just took a look at the weekly VIX for the last two years. Every time the VIX jumps off its upper Bollinger Bands as it has in the last two weeks, it heads right back down. As a technician, I note that the last five episodes of huge spikes in the VIX were followed by classic dark cloud covers and then a much lower VIX the next week. I call last week a reversion to the mean of the big drop from a week before but I'm looking for the VIX to end this coming week lower and so will go out on a limb and cautiously call the SPX higher. Time will tell...

Brett Steenbarger, Ph.D. said...

Thanks for the comment, Michele. Yes, we've had some pretty volatile VIX action. I've generally found that volatility in the VIX itself occurs closer to market bottoms than tops.

Brett

fear_and_greed said...

Although sentiment has undoubtedly grown more bearish recently, there may be other factors at work. Effective in just two weeks, new SEC regulations will actually encourage investors to buy puts so that they can extent their margin (see http://www.themoneyblogs.com/optionsandvolatility/my.blog/new-margin-rules-for-option-positions.html).

For as much talk of high put/calls, I'm a little cautious that I haven't heard anyone acknowledge the recent SEC changes. Fear is definitly prevalent, but be wary when a lot of people are bullish because they think that a lot of people are bearish.

Brett Steenbarger, Ph.D. said...

Hi Fear and Greed,

Thanks for the heads up on the rules change. It will be interesting to see how this affects the option volume data. So far, the extremes in put/call ratio have turned out to be great buys.

Brett