Caged Birds and Those That Hear Their Songs: Effects of Race and Sex in South Carolina Parole Hearings

Abstract

When most incarcerated persons go before the parole board, they hope that the decision whether to release them will be based on their institutional record; put differently, that the board will consider the use of opportunities available in prison, rehabilitation, and likelihood of success outside the carceral environment. However, numerous persons with excellent records and reentry plans are denied parole every year. Why? The actual variables that influence parole board decision making are often a mystery; parole rejections are left unexplained or opaque. Empirical research examining what drives parole outcomes is scarce, yet this research is necessary given the power the parole boards have in determining the actual amount of time served in prison. In this Article, we examined the influence of institutional variables (those related to a person's behavior while incarcerated) and noninstitutional variables on parole hearing outcomes in South Carolina. We predicted that institutional variables, such as the conviction of additional crimes during incarceration, would predict parole outcomes, but we also predicted that noninstitutional variables which may cue characteristics such as dangerousness (e.g., the nature of the offense), regardless of relevance to a person's rehabilitation, would also predict parole outcomes. We analyzed the outcomes of all (43,290) parole board hearings from 2006 to 2016 and examined the influence of variables such as a person's race, biological sex, age at the time of the first offense, time served, conviction of another offense while incarcerated, sex offender status, and number of felonies. Our results confirmed our hypotheses: although institutional variables, such as being convicted of another crime while incarcerated, influenced parole outcomes, several noninstitutional variables, particularly those which may cue dangerousness, were also significant. The most alarming results were those concerning race and biological sex. The parole board was significantly less likely to grant parole to incarcerated men compared to women and to Black people compared to white people. Further, there was a significant interaction between sex and race such that Black men were least likely to be granted parole, whereas white women were the most likely to be granted parole. In addition to the above results, the number of convictions and the severity of the crimes a person was convicted of were associated with significantly lower likelihood of being granted parole. Additional research highlighting the specific roles that noninstitutional variables should play in parole hearings is warranted, if only to root out undesirable effects on a critical aspect of the criminal justice system.