Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
I present a model of conformism in social networks that incorporates both peer effects and self-selection. I show that conformism has positive social value and that social welfare can be bounded by network polarization and connectivity measures. I apply the model to empirical data on high school students' participation in extracurricular activities. I find that the local effect of conformism (i.e. the endogenous peer effect for a fixed network structure) ranges from 7.5% to 45%, depending on the number of peers that an individual has. Simulations show that the optimal policies of an inequality-averse policymaker change depending on a school's enrollment. Small schools should encourage shy students to interact more with other students, while large schools should focus on promoting role models within the school.