The Analysts' President

{Q} Aren’t analysts’ estimates notoriously over-optimistic?

{A} Not necessarily overoptimistic. They are best judgments by individuals following these companies. So they are subjective in nature, but by looking at the estimates, the consensus and, most importantly, range or dispersion of the estimates, one can determine a great deal about how much the analysts really understand a particular company.

Companies that consistently meet the earnings expectations of the analysts are stocks that will perform very well.

{Q} Do you risk missing a company whose management just doesn’t believe in giving clear earnings guidance?

{A} No, that isn’t a factor. The ‘sell side’ analysts [i.e., those who work for brokerage houses] have to produce estimates. So regardless of whether a company issues guidance on a quarterly or an annual basis, or no guidance at all, the analysts still have to feed their models, and consequently adjust them over time to the available data.

{Q} But wouldn’t that discriminate against that company being selected for your portfolio because it could end up creating a greater level of dispersion in earnings estimates?

{A} Not necessarily, because as earnings estimates are established and revisions are made, at some point the range around the consensus number will narrow. So we don’t see any evidence that companies that provide less guidance or no guidance perform any differently than those that provide very good guidance. At the end of the day, it’s how that audience of analysts uses whatever information is available for their models.

{Q} How many analysts do you need to fuel your model?

{A} We need at least three analysts following a company for five years. Now, it need not be the same three analysts. We just need that number of data points.

{Q} Can there be too many analysts?

{A} Nope. It doesn’t make any difference. Actually, if you talk to directors and management of publicly traded companies, they will tell you that notwithstanding the number of analysts—4 or 14—there are always a couple of analysts who will know the business extremely well, and the others will have a range of knowledge or understanding. It really doesn’t make a great deal of difference in our model, because we are looking at the trends in estimate revisions, the absolute change in estimate revisions, and the dispersion or lack thereof. For any company that has analysts’ coverage, we can tell you who the most accurate analyst is and who the most influential analyst is—and they are seldom the same. We give a double weighting to the most accurate. And that’s all within our algorithm. We’ve standardized all this so we can score these stocks based on the factors I just mentioned in a logical, consistent manner. It’s hard, cold, and objective, with no emotion whatsoever.