This paper presents a Monte Carlo comparison of the small-sample performance of subsample ordinary least squares, the Heckman-Lee two-stage estimator, and the robust estimator of Lee. Each estimator is considered under bivariate normal, t, and chi-square error structures. The estimates indicate that the Heckman-Lee and Lee estimators do not provide an unequivocal mean square error improvement upon subsample ordinary least squares in small samples. While effectively controlling for selectivity bias, the two-stage estimators suffer a substantial loss of small-sample precision relative to subsample ordinary least squares. Copyright 1991 by MIT Press.

Litigants in civil lawsuits involving monetary damages often find an out-of-court settlement preferable to a trial. Most theoretical models of settlement choice employ an expected-utility-maximizing framework that emphasizes the importance of risk preferences, litigation costs, and the distribution of trial awards. An empirical model of settlement choice is used to examine whether the variables prominent in the theoretical literature are statistically useful in explaining the occurrence and monetary value of settlements. Estimates from a sample of civil filings provide new empirical evidence of how the legal system affects the behavior of litigants during the settlement process. Copyright 1989 by MIT Press.