The NBER's Working Group on Behavioral Finance met in Chicago on April 20. NBER
Research Associates Robert J. Shiller, Yale University, and Richard H. Thaler, University of
Chicago, organized this program:
Christopher K. Polk and Paola Sapienza, Northwestern University, "The Real Effects of
Investor Sentiment"
Discussant: Jeremy C. Stein, NBER and Harvard University
Nicholas C. Barberis, NBER and University of Chicago, Andrei Shleifer, NBER and Harvard
University, and Jeffrey Wurgler, New York University, "Comovement"
Discussant: Robert J. Shiller
Sendhil Mullainathan, NBER and MIT, "Thinking Through Categories"
Discussant: Jesus Santos, NBER and University of Chicago
Tim Loughran, University of Notre Dame, and Jay R. Ritter, University of Florida, "Why has
IPO Underpricing Increased Over Time?"
Discussant: Ivo Welch, NBER and Yale University
Amiyatosh K. Purnanandam and Bhaskaran Swaminathan, Cornell University, "Are IPOs
Underpriced?"
Discussant: Alon Brav, Duke University
Mark Grinblatt, NBER and University of California, Los Angeles, and Bing Han, University
of California, Los Angeles, "The Disposition Effect and Momentum"
Discussant: Harrison Hong, Stanford University
Do inefficiencies in the capital markets have real consequences? Or, are they simply
wealth transfers from noise traders to arbitrageurs? Polk and Sapienza study firm business
investment and find a positive relationship between investment and each of their three measures
of mispricing (after controlling for investment opportunities and financial slack.) Consistent with
their predictions, they find that firms with higher research and development intensity (suggesting
less transparency and longer periods of information asymmetry) have investment that is more
sensitive to mispricing.
Barberis, Shleifer, and Wurgler distinguish three views of comovement among
different traded securities. The traditional "fundamentals" view explains the comovement of
securities through positive correlations in the rational determinants of their values, such as cash
flows or discount rates. "Category-based" comovement occurs when investors classify different
securities into the same asset class and then shift resources in and out of this class in correlated
ways. "Habitat-based" comovement arises when a group of investors restricts its trading to a
given set of securities, and then moves in and out of that set in tandem. The authors model each
type of comovement, and then assess them empirically using data on stock inclusions into and
deletions from the S&P 500 index. Index changes are noteworthy because they change a stock's
category and investor clientele (habitat), but do not change its fundamentals. The authors find
that when a stock is added to the index, its beta and R-squared with respect to the index increase,
while its beta with respect to stocks outside the index falls. The converse happens when a stock
is deleted. These results broadly support the category and habitat views of comovement, but not
the fundamentals view. More generally, these non-traditional views may help to explain other
instances of comovement in the data.
Mullainathan presents a model of human inference in which people use coarse
categories to make inferences. "Coarseness" means that, rather than updating continuously as
suggested by the Bayesian ideal, people update or change categories only when they see enough
data to suggest that an alternative category fits the data better. This simple model of inference
generates a set of predictions about behavior. Mullainathan applies this framework to produce a
simple model of financial markets which produces straightforward and testable predictions about
the predictability of returns, comovement, and volume.
In the 1980s, the average first-day return on initial public offerings (IPOs) was 7 percent.
The average first-day return doubled to almost 15 percent during 1990-8, before jumping to 65
percent during the internet bubble years of 1999-2000. Part of the increase can be attributed to
changes in the composition of the companies going public. Loughran and Ritter attribute much
of the increase in underpricing, though, to previously latent agency problems between
underwriters and issuing firms. They argue that the increase in valuations over time has caused
issuers to be more complacent about leaving money on the table.
Purnanandam and Swaminathan study the valuation of initial public offerings (IPOs)
using comparable firm multiples. In a sample of more than 2000 IPOs from 1980 to 1997, the
median IPO is overvalued at the offer by about 50 percent relative to its industry peers, they find.
In the cross-section, overvalued IPOs earn 5 percent to 7 percent higher first day returns than
undervalued IPOs, but earn 20 percent to 50 percent lower returns over the next five years.
Overvalued IPOs temporarily exhibit higher sales growth rates but persistently earn lower profit
margins and return on assets than undervalued IPOs over the next five years. This suggests that
any projected growth opportunities implicit in the initial valuation fail to materialize
subsequently. These results are not consistent with asymmetric information models of IPO
pricing and rather support behavioral theories based on investor overconfidence.
Prior research shows that many investors have a lower propensity to sell stocks on which
they have a capital loss. This behavioral phenomenon, known as "the disposition effect," has
implications for equilibrium prices. Grinblatt and Han investigate the temporal pattern of stock
prices in an equilibrium that aggregates the demand functions of both rational and disposition
investors. The disposition effect creates a spread between a stock's fundamental value -- the
stock price that would exist in the absence of a disposition effect -- and its market price. Even
when a stock's fundamental value follows a random walk, and thus is unpredictable, its
equilibrium price will tend to underreact to information. Spread convergence, arising from the
random evolution of fundamental values and updating of the reference prices, generates
predictable equilibrium prices. This convergence implies that stocks with large past price run-ups and stocks on which most investors experienced capital gains have higher expected returns
than those that have experienced large declines and capital losses. The profitability of a
momentum strategy, which makes use of this spread, depends on the path of past stock prices.
The authors find that stocks with large aggregate unrealized capital gains tend to have higher
expected returns than stocks with large aggregate unrealized capital losses; this capital gains
"overhang" appears to be the key variable that generates the profitability of a momentum
strategy. When this capital gains variable is used along with past returns and volume to predict
future returns, the momentum effect disappears.
