Purpose We examine the effect of an offender’s occupational status on criminal sentencing recommendations using a vignette experiment that crosses the offender’s occupational status (white-collar vs blue-or pink-collar) and the crime label, with one label (overcharging) associated with white-collar offenders and the other (robbery) associated with lower-status offenders. We expect negative and potent post-crime impressions of the offender and the crime to increase perceptions of criminality and, in turn, the recommended sentence. We term these negative and potent impressions “criminality scores.” Drawing on affect control theory (ACT) impression formation equations, we generate criminality scores for the offenders and the crimes in each condition and, using those scores as a guide, predict that white-collar offenders and offenders described as “robbing” will receive a higher recommended sentence. We also expect eight perceptual factors central to theories of judicial sentencing mediate these relationships. Methodology We test these hypotheses with a vignette experiment, administered to female university students, that varies a male offender’s occupation and the word used to describe his crime. Findings Consistent with our ACT-derived predictions, white-collar offenders and offenders described as robbing received a higher recommended sentence. But, contrary to predictions, only one perceptual factor, crime seriousness, mediated these effects, and the mediation was partial. Research Implications Our findings suggest the perpetrator’s post-crime appearance of negativity and power offer a valuable supplement to theories of judicial sentencing. Originality This study is the first to test the hypothesis that sentencing disparities may be due to the way the perpetrators’ sociodemographic attributes shape their post-crime appearance of negativity and power.