Measuring Expectations

In the case of oil service companies, in terms of their fundamental outlook, they haven’t changed at all. Earnings estimates have actually increased a little bit for these companies, because the economics of their business don’t really change if oil prices drop. Yet their stock prices came under the same pressure as the rest of the industry. This is the type of overreaction we’re trying to exploit. So in our sector strategies, we have energy services as the top pick.

{Q} What are you trying to avoid within the sectors?

{A}   Right now, utility stocks are something our model is saying is unfavorable, meaning that most of these stocks have increased in price or reached 52-week highs, but their earnings outlook has actually dimmed a little. Part of that has to do with the specific aspects of the company service territory or how they hedged various oil contracts. 

{Q} How does the model work in terms of your sell discipline?

{A} Underreaction to a deterioration in factors in our model is based on mean reversion: It says that the company’s current value is different than its normal value, and that the market will work to correct it . . . . If the stock price starts going above that normal value, our sell discipline would identify a more favorable under- or overreaction elsewhere.

{Q} Do you have an example?

{A} Johnson & Johnson [NYSE: JNJ] is one of the best quality names in the market, and if you were to go back and look at its highs, the over- and underreaction models are very accurate in terms of telling you when to buy and when to sell it. We have actually bought and sold Johnson & Johnson two or three times over the past five years based on that. Higher-quality companies in general tend to trade very predictably, and if you apply this discipline, we believe it produces very good results.

{Q} It sounds like the model might not work as well in less efficient parts of the market.

Yes. If we’re dealing with companies that have 18 or 19 analysts, we can more easily define whether there is an over- or underreaction than we can with a company with only three analysts, because not enough analysts are providing the kinds of robust estimates we need to give us a sense of where ‘normal’ is.

{Q} What worries you the most?

{A} That the nature of the market is changing. That is, if you were to look at the volume on the NYSE, 60 to 70 percent is related to program trading—much of it driven by hedge funds. That, from our standpoint, causes the variance of stocks to be less.

What you see are investors buying energy or selling health care—buying and selling a whole sector as part of large program-driven strategies. Regardless of the stock’s unique characteristics, the valuations get smoothed out in the marketplace, and we don’t get the degree of over- or underreaction we would want to get in order to truly exploit our model.