Sunday, 26 August 2012

Interpreting Mean Reversion: Mispricing Or Time-Varying Equity Risk Premiums?

This is the final article in the series on Equity prices, returns and mean reversion (preceded by Mean Reversion and Equity Prices: Background Research and Implications for Investments, Cyclical Variation in Equity Prices, Interpreting Mean Reversion: Mispricing in Irrational Markets and Interpreting Mean Reversion: Time-Varying Equity Premium). This article discusses whether mean reversion in equity prices and returns stem from mispricing or changes in the equity risk premium through time.

Interpreting Mean Reversion in Equity Prices: Mispricing Or Time-Varying Equity Risk Premiums?

The above literature review shows that market anomalies such as out performance of value stocks over glamour stocks, investor overreaction and past extrapolation of growth rates can be explained in terms of mean reversion. Consider the following: Growth stocks achieve (part of) their high valuations through extrapolation of high past growth rates into the future leading to high returns over shorter periods. In the long run, the mean reverting nature of profitability causes profits to slow and valuation falls to more normal levels, inducing mean reversion. At the other end of the spectrum, past losers, which are often value stocks, provide superior returns because the market slowly realizes that earnings growth rates for value stocks are higher than it initially expected. In fact, mean reversion in equity returns may occur because investors appear to ignore the fact that the aggregate market’s profitability (return on equity) over longer horizons is itself mean reverting (Fama et al (2000)). Conceptually, the De Bondt et al (1985) contrarian strategy is also based on (and is probably profitable because of) the idea that stock indexes may revert to means over long horizons, and that returns are forecastable from past price formations (Balvers et al (2000)). There does not appear to be a common consensus among financial economists as to which of the two main theories; mispricing in irrational markets or time-varying equity risk premiums, explains mean reversion. The issue is a delicate one. If the markets are irrational, it would be damning evidence against the very foundation of financial theory; that markets are efficient and that asset prices follow a random walk. This controversy has split financial economists into two camps: (i) Those believing that markets are rational and that mean reversion is caused by time-varying rational equity risk premiums and (ii) those holding that mean reversion is a reflection of irrational markets and investors. The former belief is closely tied to the Chicago school of thought (e.g. Fama) and the latter to the emerging behavioural finance view (e.g. La Porta, Lakonishok, Shleifer and Vishny). Noteworthy is the former camp’s application of the equity risk premium. It is viewed as a number that makes the equation (the CAPM or some other asset pricing model) balance. In this sense it is derived very much like an accountant works out the equity on the balance sheet; direct entries to the equity account is (almost) never made and equity is automatically derived from the difference between assets and liabilities, i.e. it is the number that makes the accounting equation hold. As for the other camp, high returns are occasionally perhaps too quickly attributed to irrational markets when in fact the market may be exante rational with respect to the asymmetric information available at that point in time. Both camps offer tremendous insight into the issue of equity price formation and the solution is probably some fair mix of the two. In summary, there is supportive evidence that markets are not always rational and that behavioural finance theories are prudent additions to the efficient market theory. In light of the findings that value stocks are actually less risky than glamour/ growth stocks (De Bondt et al (1989), Lakonishok et al (1994), La Porta (1996), Dreman et al (1997) and La Porta, Lakonishok, Shleifer and Vishny (1997)), based on numerous measures of risk (e.g. standard deviation, performance in adverse economic states and bull- and bear market betas), this is damning evidence to the former view that the higher return is compensation for assuming higher risk. This evidence leans in the direction that time-varying equity risk premiums and share prices sometimes diverge from that warranted by economic fundamentals.

No comments:

Post a comment