Let's say that today beta only accounts for 8% of the stock's variability, i.e. its R squared is 0.08.

If investors use beta to determine discount rates and calculate NPV, etc., can this 8% increase to, say, 12% solely because beta was used to make previous investment decisions?

Ideally I think the answer should be no, because if its R squared increases, that causes some kind of internal bias. But I don't know how to even start to think about it.

  • $\begingroup$ Are you assuming that investors use the same model? Which model? $\endgroup$ – David Addison Feb 2 '18 at 22:31
  • $\begingroup$ Yes, the same model. Assume CAPM. I might also formulate the question as: if all investors use different valuation models, does that reduce the efficiency of beta/CAPM? Because the internal flaws of each valuation method will be reflected in the market now and these are unlikely to be reflected in beta. $\endgroup$ – user27257 Feb 3 '18 at 14:45
  • $\begingroup$ Okay. So you are using CAPM to calculate the discount rates, but are you also using CAPM to calculate asset prices? NPV implies you assume that investors also behave according to expected future cash flows. I think it's an important distinction because if all investors used only CAPM to price assets, then a low beta security would also imply low expected return (i.e., it would be expected to increase in either volatility or correlation versus the benchmark). On the other hand, securities priced via DCF begin to look increasingly attractive as beta decreases. $\endgroup$ – David Addison Feb 3 '18 at 18:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.