| |
| |
| |
Working Papers
Characteristics of the investment banking industry, particularly the
extreme concentration
of spreads at exactly 7%, seem consistent with some form of collusion
through which
underwriters can extract surplus from the IPO. I present a model of
investment banking
that, under the assumption of optimal collusion, generates a
distribution of spreads qualitatively
similar to that observed. The model is extended to show that
underpricing and
spread rigidity may arise together, each one reinforcing incentives to
engage in the other.
-
- with Paul Healy, John Ledyard, and Sera
Linardi
In many environments, prediction markets aggregate information and
accurately
estimate the probability of future events. But, these markets have
typically performed
best in ’simple’ situations with many traders and few securities. We
test the standard
prediction market mechanism in a highly ‘complex’ environment with
several securities
and few traders. We compare its performance to three alternative
mechanisms for aggregating
information. In the complex environment the performance of the
prediction
market is dominated by a simple iterative polling mechanism. We analyze
four behavioral
conjectures that explain why the poll performs better in the complex
setting.
- Financial Intermediation, Trust, and Asset
Values
In thin financial markets where intermediation is necessary to
facilitate exchange, the
intermediary may have an informational advantage in addition to his
cost advantage for
acquiring the security. If all information is eventually revealed, the
intermediary may have
an incentive to truthfully reveal the value of the security to a
customer as soon as he learns
it rather than attempting to profit from the asymmetric information.
Whether this incentive
is sufficiently strong depends on the patience of the intermediary and
the probability
that he will have future interactions with the same client. This
probability depends on
the value of the security because securities that decline in value may
cease to trade. I
study a model of repeated interaction between clients and
intermediaries that takes into
account the correlation between value and continuation probabilities.
The model captures
the unresponsiveness of thinly traded securities to bad news and
explains the breakdown in
liquidity following declines in asset values. This fact can help
explain why relatively illiquid
securities, such as those based on subprime mortgages, can experience
apparent bubbles
and crashes.
- Social Capital as Economic Overlap
This paper presents a model of endogenous social capital where location
decisions can
generate the necessary means to sustain cooperative behavior in the
absence of legal in-
stitutions or social conventions. By choosing to locate close to each
other, agents create
public goods that facilitate cooperative behavior on other endeavors.
The model can serve
to explain both initial agglomeration decisions and cooperation in
extra-legal environments,
even in the absence of frequent repetition.
|
|
|
