Prosecutors are the most powerful actors in the American criminal justice system. They decide who gets charged and with what offense. They determine whether to seek mandatory minimums. They set plea offers that the vast majority of defendants accept. They decide, effectively, who goes to prison and for how long. They exercise this authority with almost no external oversight, with limited discovery obligations that would expose their reasoning, and with almost no systematic data collection that would allow external analysis of their patterns.
This is not a controversial description. It is the consensus view among scholars who study the system. Former U.S. Attorney General Eric Holder said in 2013 that prosecutorial discretion had contributed to the overcrowding of federal prisons. The National Association of Criminal Defense Lawyers has documented in multiple reports that charging decisions and plea negotiation practices are the primary drivers of incarceration rates in most jurisdictions.
Consider how discretion operates at the charging stage. A person arrested for selling drugs can be charged under a range of statutes carrying widely different mandatory minimums. The same conduct, with the same evidence, can result in a two-year minimum or a ten-year minimum depending on which statute the prosecutor selects. The prosecutor's choice is largely unreviewable. Defense counsel cannot demand disclosure of how other defendants in similar cases were charged. Judges cannot second-guess charging decisions without extraordinary circumstances.
The plea bargaining process extends this discretion into an overwhelming majority of cases. Approximately 97 percent of federal convictions and 94 percent of state convictions result from guilty pleas, not trials. Plea negotiations are conducted between prosecutors and defense attorneys with no public record, no judicial oversight of the offer itself, and no requirement that the prosecutor offer consistent terms to similarly situated defendants. A defendant who happens to appear before a prosecutor carrying a heavy caseload on a bad day can receive a harsher offer than an identically situated defendant seen a week later.
The racial data on prosecutorial decisions is difficult to obtain precisely because of the opacity described above. Research that has penetrated that opacity finds consistent patterns. A 2021 study published in Nature Human Behaviour analyzed charging decisions in one large urban jurisdiction and found that Black defendants were 25 percent more likely to face mandatory minimum charges than white defendants arrested for the same conduct with similar prior records. When the authors sent case summaries to lay people and trained attorneys, they found no evidence that the conduct in Black defendants' cases was more serious. The disparity reflected prosecutorial choices.
Several major cities elected reform-minded prosecutors in the 2020s with mandates to change these patterns. The experience in Philadelphia under Larry Krasner, in San Francisco under Chesa Boudin (before he was recalled), and in Los Angeles under George Gascón produced genuine reductions in incarceration rates. They also produced intense political backlash, funded substantially by law enforcement associations and, in some cases, by the bail industry. The lesson is that prosecutorial reform is possible and that it generates fierce resistance from institutions that benefit from the existing pattern.
Transparency is the minimal precondition for any other reform. Prosecutors should be required to collect and publicly report charging and plea data disaggregated by race. That data should be available for independent analysis. Without it, the hidden engine of mass incarceration continues operating out of public view.