We exploit conditionally random assignment of case files to judges in a Southern U.S. state to study the effects of same-gender judge assignment on juvenile justice decisions and long-run socioeconomic outcomes. Using a generalized differences-in-differences design, we find evidence of same-gender gaps in sentencing decisions: relative to males, female judges are 18% less likely to incarcerate female than male defendants. A novel rank-order test confirms these findings and indicates that the results are not driven by a small subset of judges. Own-gender judicial assignment, especially for Black children, has lasting effects on socioeconomic outcomes beyond the courtroom: educational attainment increases while adult criminal involvement and welfare use decrease. Further analyses suggest that these long-run improvements are not just driven by differential incarceration decisions.