The authors propose a theory of security market under- and overreactions based on
two well-known psychological biases: investor overconfidence and biased
self-attribution. Their theory implies that investors overreact to private
information signals and underreact to public information signals. The authors
show that short-run positive return autocorrelations can be a result of
continuing overreaction. These short-run effects, however, must eventually be
followed by a long-run correction. Thus, short-run positive autocorrelations can
be consistent with long-run negative autocorrelations.
In recent years, a body of evidence on security returns has presented a sharp challenge
to the traditional view that securities are rationally priced to reflect all publicly
available information. Numerous studies have found security market underreaction to
public announcements concerning stock splits, tender offer and open market repurchases,
analyst recommendations, dividend initiations and omissions, seasoned issues of common
stock, earnings surprises, previous insider trades, and venture capital share
distributions. Other studies have discovered short-term momentum and long-term reversal
patterns, excess volatility in prices relative to changes in fundamentals, and abnormal
stock price performance in the opposite direction of long-term earnings changes. The
authors offer a behavioral model based on imperfect investor rationality in an effort to
help explain some of the anomalous patterns that exist in securities markets.
The model is based on investor overconfidence and variations in confidence arising from
biased self-attribution of investment outcomes. The authors assume that investors will
systematically underestimate the forecast error surrounding their estimate of a
security's intrinsic value. According to the model, this behavioral flaw causes
investors to overestimate the precision of their private information signals but not the
information signals publicly received by all. Therefore, investors, and consequently
stock prices, will overreact to private information signals and underreact to public
signals.
The model also reflects that as individuals observe the outcomes of their actions, public
information confirming the individuals' actions results in increased confidence but that
information disconfirming the actions of individuals causes confidence to fall only
modestly, if at all. This finding suggests that public information can trigger further
overreaction to a private signal. This continuing overreaction causes momentum in
security prices, but this momentum is eventually reversed as further public information
gradually draws the price back toward fundamentals. Thus, biased self-attribution
implies short-run momentum and long-term reversals.
The authors' theory offers an alternative explanation to the phenomenon of post-event
abnormal returns following the release of public information. The authors argue that
market underreaction to new public information is neither a necessary nor a sufficient
condition for such event-based anomalies. Rather, these abnormal returns may be
generated by continuing investor overreaction to pre-event private information. For
example, post-earnings-announcement drift may be a continuing overreaction to pre-event
information triggered by the earnings announcement to pre-event information.
The authors' work also contrasts with the extant noise-trading approach to security
markets. The noise-trading model generally posits that variability in prices arises from
unpredictable trading that is unrelated to valid information. In contrast, the authors'
model is based on the premise that an important class of mistakes by investors involves
the misinterpretation of valid private information. Therefore, the model endogenously
generates trading mistakes that are correlated with fundamentals. The advantage of this
structure is that it provides predictions about the behavior of asset prices that depend
on the particular cognitive error assumed.
Moving beyond the confines of the formal model, the authors expect the effects of
overconfidence to be more severe in less liquid securities and assets, such as real
estate. The authors also argue that return predictability will be strongest in firms
with the greatest information asymmetries, which, in turn, implies greater
inefficiencies in the stock prices of small companies.