B.Descriptional TestsWhen using a multiple, it is always useful to have a sense of what a high value, a lowvalue or a typical value for that multiple is in the market. In other words, knowing thedistributional characteristics of a multiple is a key part of using that multiple to identifyunder or over valued firms. In addition, you need to understand the effects of outliers onaverages and unearth any biases in these estimates, introduced in the process of estimatingmultiples.Distributional CharacteristicsMany analysts who use multiples have a sector focus and have a good sense ofhow different firms in their sector rank on specific multiples. What is often lacking,however, is a sense of how the multiple is distributed across the entire market. Why, youmight ask, should a software analyst care about price earnings ratios of utility stocks?Because both software and utility stocks are competing for the same investment dollar,they have to, in a sense, play by the same rules. Furthermore, an awareness of howmultiples vary across sectors can be very useful in detecting when the sector you areanalyzing is over or under valued.What are the distributional characteristics that matter? The standard statistics –the average and standard deviation– are where you should start, but they represent thebeginning of the exploration. The fact that multiples such as the price earnings ratio cannever be less than zero and are unconstrained in terms of a maximum results indistributions for these multiples that are skewed towards the positive values.1Firms that adopt different rules for reporting and tax purposes generally report higher earnings to theirstockholders than they do to the tax authorities. When they are compared on a price earnings basis to firms
9Consequently, the average values for these multiples will be higher than median values2,and the latter are much more representative of the typical firm in the group. While themaximum and minimum values are usually of limited use, the percentile values (10thpercentile, 25thpercentile, 75thpercentile, 90thpercentile, etc.) can be useful in judgingwhat a high or low value for the multiple in the group is.Outliers and AveragesAs noted earlier, multiples are unconstrained on the upper end and firms can haveprice earnings ratios of 500 or 2000 or even 10000. This can occur not only because ofhigh stock prices but also because earnings at firms can sometime drop to a few cents.These outliers will result in averages that are not representative of the sample. In mostcases, services that compute and report average values for multiples either throw outthese outliers when computing the averages or constrain the multiples to be less than orequal to a fixed number. For instance, any firm that has a price earnings ratio greater than500 may be given a price earnings ratio of 500.