Barron's had an ETF Special Report which consisted of two articles including a roundtable of sorts with four people in various roles in the industry. Unfortunately there was very little from the roundtable that was useful, maybe they asked the wrong questions, but there was the following quote that I think is worth blowing up:
Active managers have struggled to add value. There is only so much alpha to go around. For every manager who wins, there has to be an offsetting losing bet.
This was in response to a comment that active sector selection is not a "great way to add alpha." At times, a tremendous amount of alpha can be added from active sector selection but not always of course.
The idea of finite alpha comes up occasionally and as I have said in the past this is an academic point - not one from the real world in my opinion. The idea of an "offsetting losing bet" whether it is in sectors or anything else is especially funny in an article on ETFs. If every investor who uses sector funds decided at the same time they all wanted zero exposure to the healthcare sector then what would happen? There would be an onslaught of redemptions and the AUMs of all of the funds would go to zero (save for the existing short positions). Where ETFs are concerned there doesn't have to be an investor on the other side.
As for offsetting losing bet for stocks, there are countless examples where this is not true. When you sell your Pfizer (PFE), the person on the other end of your trade (if it is a person) is buying Pfizer, right? What if he is buying to close out his short? So you'd both be flat. What if the person buying your Pfizer is doing so to hedge some sort of other position?
It may not be suitable for you to use sector funds, individual stocks or spend much time trying to add alpha, or maybe it is, but where the permutations of stocks and option combos are limitless, alpha is not finite.
