Many investors purchase investment newsletters. Depending on the letter, the cost can be anywhere from $50 to thousands of dollars per year. In return, the newsletter writer offers investment advice about which stocks to purchase and whether or not to be invested in the stock market at all. These letters may come in the mail every week or once a month. Some of the advice is delivered through email, Web sites, and telephone bulletin boards. Some popular examples are The Granville Market Letter, Louis Rukeyser's Wall Street, MPT Review, The Ruff Times, and The Value Line Investment Survey (discussed above).
Is this investment advice useful to investors? Academic studies find that, in general, newsletter recommendations do not outperform the S&P 500 Index. Newsletter recommendations that do perform well usually hurt their net performance because of the high trading costs. That is, many recommendations change so frequently throughout the year that they incur a high cost due to transaction costs. Indeed, the newsletters produce anywhere from a couple to 50 trading signals per year.14
To reduce transaction costs, an investor can invest in the portfolio recommended by the newsletter and then hold it for a year, rebalancing with the new recommendations after the year. Can investors profit from the newsletter recommendations if they enact this type of strategy? Harvard economist Andrew Metrick examined the recommendations of 153 different newsletters during the sample period of July 1980 to December 1996 from this perspective.15 On average, an investor would have only beaten the market in seven of the 17 years using these recommendations annually. Overall, newsletter portfolios underperformed the stock market. Of course, some newsletters do outperform the market. But is this because they are great advisers, or because they are lucky? Remember that if 153 newsletters are making predictions, some are going to do well by sheer luck. Indeed, the low number of newsletters that outperform the market are about the number one would expect through random chance (see Chapter 3, "Patterns and Predictions").
Consider the contribution of the newsletter Investors Intelligence. Intelligence subscribes to 140 other newsletters. From these newsletters, it creates a sentiment index by observing how many of the newsletter writers are bullish about the stock market and how many are bearish. Dividing the percent bullish by the sum of the percent bullish and bearish creates the index. For example, on May 5, 1995, 40.3% of the newsletters were bullish and 34.5% were bearish. The sentiment index was computed as 53.9 (=40.3/[40.3+34.5]).
How useful is the recommendation of newsletters and this sentiment index? By examining the sentiment index from 1963 to 1995 and the subsequent return on the S&P 500 Index, the predictability of the index was tested.16 The conclusion was that the sentiment index had no predictability when examining the S&P 500 Index return over a one-month, six-month, or 12-month holding period.
However, Intelligence argues that the sentiment index is most useful when it is at extreme levels (very high or very low). It is interesting to note that Intelligence believes this sentiment index is a contrary indicator. That is, it believes that its readers should do the opposite (or contrary) of the group of newsletter recommendations. When the sentiment of the newsletters is very bullish, Intelligence argues that we must be at a market top, thus it's bearish. In other words, Intelligence believes that the other newsletters' advice is only useful if you do the opposite of what they recommend! Intelligence may be onto something here. If you examine market returns after the periods with the lowest 30% of sentiment index level, the average monthly return for the following four months is 0.58%. The average return after the highest sentiment levels is 0.46%. Doing the opposite of the consensus of the newsletters is beneficial (though not by much).
However, that is not how investors appear to use the recommendations of newsletters. One study compared the sentiment of newsletters and the flow in and out of index mutual funds. The flow of money from investors into the Spartan Market Index fund, the Spartan U.S. Equity Index fund, and the VIP Market Index fund increases when newsletter sentiment is bullish.17 Outflow increases when sentiment is bearish. Investors appear to use the newsletter recommendations as is, instead of as a contrary indicator.
One other interesting behavior should be noted about the newsletter writersthey tend to be trend followers. After four weeks of a high return on the market, a migration occurs of newsletter recommendations from bearish to bullish. Every 1% increase of the market during the month causes 1.2% of the newsletters to switch to a bullish perspective. This trend-following behavior is an example of the representativeness bias discussed in Chapter 2.
Followers of newsletter advice lose out in two ways. First, the annual fee paid to subscribe to the newsletter can be substantial, especially considering that the advice of the newsletters has no predictability for stock returns. Second, by following the advice, an investor moves away from a good, well-diversified, long-term strategy and gets into the trading mentality: a mentality that magnifies the psychological biases.