Welcome! Please Log In
Essentials Popular Topics Favorite Forums & Blogs Join Discuss to setup a list of your favorite forums/blogs.
Discuss > Single Post
Investor behavior
Robert T 07-12-2007, 12:17 PM | Post #2411673 | 

I have read some of the 'behavioral' literature (don't like the word as IMO all finance/economics is about behavior - a better description of the different explanations of the value premium may be 'good' or 'rationale' behavior vs 'bad ' or 'irrational' behavior). While some of the work is interesting I still side with the risk story (i.e. on average over time the value premium is mainly due to higher fundamental risk).

In their paper Contrarian Investment, Extrapolation, and Risk - LSV suggest:

- the value premium is too large to be explained by rational pricing (should have been arbitraged away so its existance must be due to irrational pricing)

- investors extrapolate past earnings growth too far into the future (i.e. overprice high growth 'glamour stocks' and under-price out-of-favor value stocks which leads to the value premium)

- that the lack of co-variation between the value premium and equity premium is a sign of irrational pricing (co-variation is taken to mean a consistent premium for additional risk beyond the market).

Fama and French in Multifactor Explanations of Asset Price Anomalies responded to each of these findings in the following way (as least my understanding):

- An arbitrage opportunity would arise if the standard deviation of HmL was small, but it is similar to the SD of the equity premium and the size premium. So if we are concerned that the value premium is too high to be explained by rational pricing we should also be concerned about the equity premium.

- The value premium persists for longer than the time taken for mean revision of earnings growth (i.e. if investors had over-priced (high-growth) glamour stocks then when actual earnings revert one would expect prices to decline and hence the value premium to decline - but this doesn't appear to have been the case).

- Covariance with the market is not a sufficient measure of risk. Industries fluctuate between strengthen and distress with are not linked to overall market performance (e.g. the automotive industry today) so its not surprising that the value premium does not co-vary with the market.

Re: the portfolios - I simply construct them from the FF factor benchmark data rather than from the FF portfolio benchmark data.


Originally posted in thread: 59459
View Complete Thread
Reply Quote
  • Favorites
  • Flag
  • Contact
    © Copyright 2019 Morningstar, Inc. All rights reserved. Please read our Terms of Use and Privacy Policy.