based on what we've studied so far, you can see that beta is a pretty important topic. You might wonder then, are all published betas created equal? Read on for a partial answer to this question.

We did some checking on betas and found some interesting results. The Value LineInvestment Surveyis one of the best-known sources for information on publicly traded companies. However, with the explosion of online investing, there has been a corresponding increase in the amount of investment information available online. We decided to compare the betas presented by Value Line to those reported by Yahoo! Finance (finance.yahoo.com) and CNN Money (money.cnn.com). What we found leads to an important note of caution.

Consider DuPont, the chemical company. Its beta reported on the Internet was 1.33, which is larger than Value Line's beta of 1.10. DuPont wasn't the only stock that showed a divergence in betas from different sources. In fact, for most of the technology companies we looked at, Value Line reported betas that were significantly lower than their online cousins. For example, the online beta for Cisco Systems was 1.43, but Value Line reported 1.10. The online beta for Adobe (maker of the ubiquitous Acrobat software) was 1.27 versus a Value Line beta of 1.10. Value Line's betas are not always lower. For example, the online beta for eBay was .79, compared to Value Line's 1.00.

We also found some unusual, and even hard to believe, estimates for beta. Tyson Foods had a very low online beta of .11, while Value Line reported Tyson Food's beta as .75. The online estimate for Expedia was .19, compared to Value Line's 1.25. Perhaps the most outrageous reported betas were the online betas for Ghost Technology and Sollensys Corp., with betas of 2,976.08 and -4,705.53 (notice the negative sign!), respectively. Value Line did not report a beta for these companies. How do you suppose we should interpret a beta of -4,705.53?

There are a few lessons to be learned from all of this. First, not all betas are created equal. Some are computed using weekly returns and some using daily returns. Some are computed using 60 months of stock returns; some consider more or less. Some betas are computed by comparing the stock to the S&P 500 index, while others use alternative indices. Finally, some reporting firms (including Value Line) make adjustments to raw betas to reflect information other than just the fluctuation in stock prices.

The second lesson is perhaps more subtle. We are interested in knowing what the betas of the stocks will be in the future, but betas have to be estimated using historical data. Anytime we use the past to predict the future, there is the danger of a poor estimate. As we will see later in the chapter (and in the next one), it is very unlikely that Ghost Technology has a beta anything like 2,976.08 or that Sollensys Corp. has a beta of -4,705.53. Instead, the estimates are almost certainly poor ones. The moral of the story is that, as with any financial tool, beta is not a black box that should be taken without question.