Volatility Takes a Vacation
The stock market traded in an extremely tight range last week as price volatility continued to contract. Bond yields fell slightly while crude oil prices inched closer to the $60 per barrel level.
The S&P 500 started the week at 1448.39 and finished the week at 1438.09 for a whopping weekly change of -0.7%. This lack of movement is quite rare and is the smallest weekly change by the "500" since August 1992. We looked all the way back to 1990 so there were almost 900 weeks of price rate-of-change (ROC) data. This is the second recent week of extremely low price volatility of late as the S&P 500's ROC during the week of January 19 was only -1.6%. This was the fifth lowest weekly change since 1990.
Price volatility, both historic and recent, is a fascinating subject, but widely misunderstood by the media and the average investor, in our view. There are many ways to look at price volatility, and we think this is where some of the confusion comes in. You can look at intraday volatility, measuring the difference between the days price high and price low. Other measures of price movement can be analyzed on a daily, weekly, or monthly basis.
We think the size of a move is also misunderstood in the context of historical data. When the Dow Jones industrial average is up or down 100 points in a day, it is a move of less than one percent. Yet the media would have you believe that the market is soaring.
In addition, the DJIA is a price-weighted index, so stocks that are priced the highest influence it the most. Currently, IBM is the highest priced stock in the DJIA, at about $100 per share. A one percent move in IBM is only one point but equates to an eight point move in the DJIA because the DJIA's divisor is only 0.1248. So why is a 1 percent move in IBM not such a big deal but a 1 percent rally in the DJIA celebrated by the media? Oh, I forgot about TV ratings.
Stepping back a bit, we can observe that price volatility on a monthly basis has been declining since 2002. During 2002, which represented the bottom of the last bear market, the highest monthly volatility was seen in September when the S&P 500 fell just a bit over 11%. This was followed by the largest gain that year with an 8.65% surge in October. That year had seven months of at least a 5% move for the index. Moving forward to 2006, the largest monthly loss was only 3.1%, and in fact, it was the only monthly loss for the year. The largest monthly gain during 2006 was only 3.15%. Looking at 2004 through present, monthly price volatility has only ranged between 3.9% and -3.4%.
Historically, low volatility has been a bull markets best friend, but not always. Price volatility was very low in the first half of the 1990's, but expanded in the second half as the bull market really got in gear. Whether that happens this time is anyone's guess, but many times, periods of low volatility are followed by long periods of rising volatility, and periods of compressed stock returns are followed by periods of expanding (both positive and negative) stock returns.
What is confounding about market analysis is the time it takes for the market to change trend when the majority of the indicators you follow are predicting a change in trend. Market tops, whether they are intermediate- or long term, seem especially hard to pinpoint until well after the fact. As a technician, you have a good feeling that something is about to change to verify your work, and then you get another week of marginally higher prices.
What is even more difficult to predict is a consolidation in an ongoing trend, without getting a pullback. The consolidation takes the place of a pullback or correction and all the momentum indicators that are overbought, or the internal divergences that are being seen, or all the high bullish readings of market sentiment are corrected by time, and not by price weakness. It has been our experience that these are rare but we will be cognizant of a strong upside breakout by all the major indexes.
Instead of looking at the sentiment indicators that are overbought, or at bullish extremes and calling for a pullback, we will look at one that is actually forecasting a rally. This is another conundrum for the technician. Why are so many sentiment indicators showing high levels of bullishness and complacency, while this one is showing a lot of fear? Well, we don't know, we just forecast what the majority of our indicators are telling us.
The NYSE short-interest ratio is simply monthly short interest on the entire exchange divided by the average daily volume of the NYSE for the last month. We use this, as many sentiment measures, as a contrarian indicator. Since 1990, this indicator has done a good job of calling both the intermediate-to long-term trends in the stock market. From the beginning of 2002 until August, 2006, this ratio has stayed in between 4.5 and 6.7. In September, the ratio hit 7.0, the highest since July, 1998, which actually turned out to be a good time to be short. Following the high September reading, the NYSE short interest ratio backed off a bit, but was back up to 6.8 in January. This indicator has tended to be at high levels during market corrections, not during major advances.
Bond yields pulled back last week before turning higher on Friday. The 10-year Treasury yield fell to 4.71% on an intraday basis Thursday but climbed Friday, finishing the week at 4.78%. Thursday's reversal occurred right at a couple of pieces of resistance. Both the 50-day and 200-day exponential moving averages sit around the 4.72% level. In addition, chart resistance from the yield highs at the end of last year, come in at the 4.72% zone.
The pullback in yields has equated to a 38.2% retracement of the move from 4.4% to 4.9%, and may suggest that the counter trend move is complete, and that the uptrend in yields is resuming. A break above the 4.9% level would then bring the 5% to 5.25% zone into focus.