Talks and presentations

Maximum likelihood estimation of social interaction effects with nonrandom group selection

August 01, 2004

Talk, Unpublished, Burnaby, BC Canada

This paper derives a maximum likelihood estimator for an econometric model of discrete choice with social interaction effects. Endogenous selection of reference group is addressed within the econometric model through the incorporation of a reduced form within-group correlation in both observed and unobserved characteristics. The estimator requires only standard survey data which provides information on a binary choice made by the respondent, a vector of the respondent’s background characteristics, and the average choice made by a large reference group (for example, school or census tract). Properties of the estimator are demonstrated analytically and through Monte Carlo experiments.

Peers and prices: Explaining the black-white youth smoking gap

August 10, 2001

Talk, Unpublished, Burnaby, BC Canada

In 1976, black and white teenagers in the United States were about equally likely to be cigarette smokers. By the early 1990’s, the smoking rate of black teenagers had dropped to one-third that of white teenagers. This paper analyzes the role of peers, prices, and other factors in explaining this divergence in behavior. I find that the dynamics of youth smoking can best be explained by the combination of rising prices in the 1980’s, a higher price elasticity for black teenagers, and the amplifying effects of social interactions (peer effects). In the process, I develop and implement several empirical tools for the analysis of the equilibrium implications of social interactions. In particular, I develop a procedure for determining whether peer influence is strong enough to produce multiple equilibria, and a procedure for estimating the ‘social multiplier’ associated with peer effects. I find that the multiple equilibria explanation is not empirically supported, but that the social multiplier effect is large enough to account for roughly half the difference in smoking rates.

Social interactions, thresholds, and unemployment in neighborhoods

December 20, 2000

Talk, Unpublished, Burnaby, BC Canada

This paper establishes a stylized fact about unemployment in U.S. urban neighborhoods: the relationship between average human capital and the unemployment rate in a given neighborhood is highly nonlinear. Specifically, the predicted unemployment rate for a neighborhood increases dramatically when the fraction of neighborhood residents with college degrees falls below twenty percent. This pattern appears across all of the major U.S. cities. I then use a simple model of neighborhood selection and neighborhood effects on employment to evaluate ‘epidemic’ and ‘sorting’ explanations for this stylized fact.