streetauthority.com
Steven Poser
Editor
The ETF Authority
New York, NY
Technical analysts love to lean against the crowd. It is a built-in reaction and an interesting psychological phenomenon. I am rarely comfortable if too many people agree with me. At the same time, it is incredibly difficult to make a trade when you are all alone. This is exactly what sentiment is all about.
Every trend is comprised of a period where only the lucky few are in on it (some call this the "smart money"). For example, at the lows in October 2002, almost everybody was writing about how the rally was a mere bear market bounce. I still believe the rally is, but it is a huge one, and it is probably not over yet. Finding buyers then, or even at the March 2003 bottom, would have been tough to do.
The next step in a trend takes place when the so-called "technical traders" jump on the bandwagon. The whole world has not yet caught on, but volume starts to pick up. If you're in the early stages of a bullish rally, then the market (or the particular stock) is making higher highs and higher lows. The opposite is true if the trend is down. Prices will actually start to move higher (in a rally) more quickly -- meaning that momentum is on the rise. Soon after, the "crowd" kicks in. It is not a bad thing to be on the same side of the crowd at this point. The move is not over yet.
The final part of the run sees everybody jumping in feet first, and not even looking. Look at all the sellers after September 11, 2001, or at the bottoms in October 2002 and March 2003. Or even better, the buyers at the March 2000 top. Usually, you have almost everybody playing from the same side, be it long or short. People are fully invested and there's nobody left to buy (in a bull market) or to sell (in a bear market). This is when contrarians -- those that like to trade against the crowd -- make their money.
The above description, by the way, is largely a partial explanation of Dow Theory. If you are interested in Dow Theory, then pick up Technical Analysis of Stock Trends by Edwards and Magee. (You can purchase this book at a discount by clicking here.)
So, how does one measure sentiment? There are many ways. I will discuss a few of these below.
Surveys and Polls: There are surveys everywhere you look that ask investors, traders, analysts and portfolio managers what they think about some investment or other. The ones that measure portfolio managers' bullish or bearish leanings are best known. CNBC talks about them all the time. I take part a short-term oriented one that is published by the former head of Knight-Ridder's Chicago bureau. Some come out weekly, some monthly, some even daily (I am not so sure daily sentiment indications are terribly meaningful).
In order to really use the data generated by a particular survey, you first need to understand that survey and to know the levels at which its extremes have typically foreshadowed changes in market trends. Also, you must keep up on the survey itself. Some surveys gain and lose participants frequently, which can make comparisons to previous time periods useless. Some are very short-term oriented, while others are longer-term in nature.
Regardless of how the survey is constructed, the idea remains the same: Most of the time the data will tell you little. However, at historic extremes the survey might accurately forecast an imminent trend reversal. I would never suggest making a trade based exclusively on one of these polls, but it is a piece of evidence worth considering.
Asset Allocation:
What is Wall Street telling you? If most major brokerage houses are recommending that you allocate 70% of your portfolio to common stocks, then that is usually a sell signal. If they are at 50%, then it is often a buy signal. Richard Bernstein at Merrill Lynch tracks that information, but you can probably find out what most big brokerage firms are recommending for free on their web sites. The idea is that most companies rarely push the stock market portion of a portfolio above 70-80%, or below 50%. For those big-picture bears, the asset allocators have been near 70% throughout much of the bear market. We will most likely not have a final bottom until nobody cares anymore and allocations fall down near 50% again.
Options Volatility:
If you've been reading this newsletter for awhile, then you've seen me talk about the S&P 500 VIX (Volatility Index) and the similar Nasdaq-100 VXN. These measure the implied volatility of at-the-money options on these indices (see investorwords.com for definitions of these terms). Typically, very high volatility can signal a bottom in the stock market, while low volatility may warn of an impending top. However, volatility can be seasonal. It gets low in the summer months, so you must always be aware of those patterns.
Note also that the underlying indices can change. For example, the Nasdaq-100 removed a slew of technology sector stocks last year. The changes have made that index less volatile due to its more docile make-up. Therefore, one would expect lower volatility in general from that volatility index.
Newsletter Writers Bull/Bear Levels:
Several prominent publications that track the percentage of newsletter writers that are bullish or bearish. Unfortunately, newsletters come and go, and I am not convinced that these types of indices make a good sentiment indicator. However, if all newsletter writers (or nearly all) are bullish or bearish, then I would stand up and take notice. Hopefully, this one will not be with everybody else!
Rydex Fund Ratios:
Rydex has bull and bear funds as well as bond market funds. Tracking how much money is in each of these funds has, in the past, provided good short-term to medium-term buy and sell signals. However, I've heard some talk on Wall Street recently that hedge funds have been using the Rydex Funds as hedging vehicles, which has altered the nature of this method.
Mutual Fund Cash Flows:
Trimtabs and others publish, with a delay, how much money is flowing into and out of stock, bond and money market funds. Losses of momentum or extreme inflows/outflows may precede a major change in trend.
Mutual Fund Cash Levels:
These data are reported with a delay. However, if a bond fund is 100% in bonds, then unless new money flows into the fund, it will not have any more money to allocate to bonds. Since funds represent a large portion of market activity, if cash levels are very low, then there is probably very little room for the market to post further gains. If cash levels are high, then portfolio managers are being conservative and are probably under-invested. Any rally will force managers into making major purchases as they attempt to "catch up" to their benchmarks.
In summary:
Sentiment indicators are supposed to help you determine what the crowd thinks. When you're in the middle of a trend, the crowd is usually correct. However, they get on board late and leave late. Sentiment indicators can help you identify what the crowd is thinking. I would rarely ever trade just based on a sentiment signal, but when coupled with other technical signposts, you can often make a lot of money if you understand how to use them.
If you want to read about manias and sentiment, then I suggest Charles Mackay's classic: "Extraordinary Popular Delusions and the Madness of Crowds." (You can purchase this book today by clicking here.) |
