The entire dissertation is devoted to modeling and estimating asymmetric search. The first essay examines the impact of firm heterogeneities on equilibrium pricing behavior in an online market where an information gatekeeper charges click-through fees, and rationalizes the observation that some firms that persistently charge high prices none-the-less advertise prices at comparison sites. Consistent evidence of asymmetric pricing is found in data. Using a two-step GMM estimator, I obtain structural estimates of the "effective" number of competitors, the proportion of customers using the price comparison site, and the welfare gains the site generates for consumers. The second essay studies asymmetric price dispersion between online and off-line channels and asymmetric pricing across traditional retailers, e-tailers, and multichannel retailers. The asymmetric channel use by the sellers and the different customer compositions online and off-line result in asymmetric pricing across the three types of sellers. The online channel has relatively lower prices but not necessarily less price dispersion. Data collected from a leading price comparison site are consistent with the predicted asymmetric pricing between e-tailers and multichannel retailers. The third essay introduces an efficient GMM estimator based on distribution function and studies its asymptotic properties as well as finite-sample performance. For a search model, Monte Carlo study demonstrates the new method is more efficient than Bayesian estimation based on theoretical quantiles with no sampling error. Then I extend the estimator to conditional distribution functions, derive its asymptotic properties, and study its finite-sample efficiency in linear models.
展开▼
