Using prior arrests instead of sentences drives racial plea disparities

Evidence

Prosecutorial overreliance on prior arrests history is another vehicle for exacerbating racial disparities in the justice system. Although nearly half of all stop-and-frisk arrests did not result in conviction in New York City (New York State Civil Rights Bureau, 2013), Manhattan prosecutors routinely relied on prior arrests in making plea offers on subsequent unrelated cases. According to the Plea Guidelines used by the New York County District Attorney’s Office, only defendants who have no history of prior arrests should be offered a plea to a lesser charge. Our research suggests that if these guidelines were based on prior sentences, as opposed to prior arrests, much of the disparities between Black and White and Hispanic and White defendants, would have disappeared, at least in misdemeanor cases (Kutateladze & Andiloro, 2014). — Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law, 571

Although nearly half of all stop-and-frisk arrests did not result in conviction in New York City (New York State Civil Rights Bureau, 2013), Manhattan prosecutors routinely relied on prior arrests in making plea offers on subsequent unrelated cases. According to the Plea Guidelines used by the New York County District Attorney’s Office, only defendants who have no history of prior arrests should be offered a plea to a lesser charge. Our research suggests that if these guidelines were based on prior sentences, as opposed to prior arrests, much of the disparities between Black and White and Hispanic and White defendants, would have disappeared, at least in misdemeanor cases (Kutateladze & Andiloro, 2014). — Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law, 571

Categorization

• QUESTION: Worth chasing: whether the arrest-vs-sentence distinction reflects a deeper pattern where ‘legally neutral’ data choices embed racial bias — a mechanism generalizable beyond plea bargaining.
• XREF: Connects to the ‘risk assessment instruments encode bias’ literature and to broader algorithmic-fairness debates about proxy variables.
• SEED: A clean causal finding: substituting prior-conviction for prior-arrest metrics would erase most Black-White plea disparities — a concrete policy lever for the bias essay.
• corroborated: 4 independent passages support this claim — Joseph Avery, Jim Sidanius, Joel Cooper, Gregory Davis, Amanda N., Bias in the Law: A Definitive Look at…, 571