Modified labeling theory theorizes that when people acquire a label, personally held views about that label gain relevance and exert negative effects. We assessed whether being arrested reduces self-esteem to different extents based on the degree to which individuals hold stigmatizing beliefs about people with arrest records. Adults living in the South Bronx, New York City (N = 532, 56% of whom had ever been arrested) indicated their level of agreement with statements about people with arrest records. We used exploratory factor analysis to identify categories of stigmatizing views, and calculated scores for thetwo following categories: "stereotype awareness" and "stereotype agreement." Self-esteem was assessed with the Rosenberg self-esteem scale. Using fitted linear regression models, we assessed interaction between arrest history and each stigma score, and calculated mean differences representing the association between arrest history and self-esteem score, for those with stigma scores one standard deviation (SD) below and above the mean. For each type of stigma, participants with stigma scores one SD below the mean had similar self-esteem scores, regardless of arrest history. However, among participants with stigma scores one SD above the mean, those who had experienced an arrest had lower self-esteem scores than those who had not (mean difference = - 2.07, 95% CI - 3.16, - 0.99 for "stereotype awareness"; mean difference = - 2.92, 95% CI - 4.05, - 1.79 for "stereotype agreement"). Being arrested affects self-esteem to a greater degree among persons who hold stigmatizing views about people with arrest records. These findings support a modified labeling theory of arrest-related stigma.