Researchers explored if, how and when race factors in criminal sentencing by analysing more than 17,000 decisions from South Carolina in the US.
The study by researchers from University of Sheffield in the UK and University of Minnesota in the US highlights inequalities in incarceration rates and sentence lengths for minority offenders.
The research tested the "liberation hypothesis", which concerns how much flexibility judges have when sentencing, depending on the relevant case facts.
However, in more ambiguous contexts, judges are "liberated" from the constraints of extreme criminality.
Also Read
In these instances there is room for judges to exercise discretion, and the door is opened for extra-legal characteristics such as race to influence sentencing decisions.
The researchers chose to study data from South Carolina, where there are no sentencing guidelines and decision-makers have greater discretion when sentencing offenders.
Black people with lower levels of criminal history were more likely than white people to be jailed, with the likelihood of incarceration increasing by as much as 43 per cent for those with no past criminal history to ten per cent for those with moderate criminal history.
However, when offenders had a substantial criminal record, this had a constraining effect that neutralised the impact of race.
Black offenders of low severity crimes received slightly longer sentences than white offenders, but high severity black offenders received shorter average sentences than white offenders, researchers said.
"Of course, this is just part of the story, as contact with law enforcement is only the first stage of the criminal justice system," Hartman said.
"It is particularly concerning that this pattern of disparity appears to be affecting African American offenders with limited criminal histories or for less severe crimes," he said.
The study was published in the Journal of Quantitative Criminology.