I'm currently working on a paper about the value premium. Right now I'm wondering if there's any reason why calculating the long-short portfolio return as a difference between the high B/E portfolio return and the low B/E portfolio return wouldn't make sense. For other anomalies like size (Banz, 1981) or specific return reversal (Rosenberg, 1985), the authors did this. However, in very the same paper, Rosenberg didn't construct such a long-short portfolio for the value effect. Therefore my question, is the any reason this could be wrong for the value premium which I can't see right now?

Additionally, is this only applicable for equal-weighted portfolios or can such a long-short portfolio be also constructed with?

Thanks a lot!

  • $\begingroup$ Yes, long high B/E and short low B/E is often considered as a strategy in the literature. Even if Rosenberg, Reid and Lanstein do not exactly follow this procedure, they do have both negative and positive weights in their portfolio; it is a little more complicated how they do it, as they attempt to neutralize other factors. The way Fama and French do it is simpler and can be considered the standard nowadays. $\endgroup$
    – nbbo2
    Jan 6, 2020 at 17:42
  • $\begingroup$ alright, thanks a lot! $\endgroup$
    – user43224
    Jan 6, 2020 at 17:53


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.