Though scholars recognize that peer-based risks for offending are especially robust, a handful of researchers have started to question whether the vulnerability to these risks varies across people in theoretically meaningful ways. For instance, drawing on theory and empirical research, there is reason to suspect that individuals of high and low morality are not vulnerable to deviant peer reinforcement, whereas those who exist in the “middle ground” of morality are. In this way, there may be an inverted “U” of susceptibility to deviant peer socialization according to the level of subjects' propensity for offending. The current study investigates this hypothesis using longitudinal data from the National Youth Survey. Peer reinforcement does not significantly influence the offending behavior of the high morality individuals, and is a consistent and significant predictor for medium morality offenders. For low morality offenders, however, the results are inconsistent across the models. The theoretical and methodological implications for future research on the peer-propensity interaction are discussed.