When Disparities Compound

Why racial inequities in the justice system build from the very first decision

By: Marijana Kotlaja, Basil Soper, Sarah Jensen, and Shannon Magnuson

Imagine the criminal legal system as a stream flowing into a powerful river that leads to a waterfall. The moment someone enters, often through a police stop, arrest, or jail booking, they are caught in the moving current. At first, the flow might be slow, and there are points along the banks where a person can step back onto land: release before trial, diversion programs, supportive pretrial supervision, or early case dismissal. These moments are the system’s earliest chances to redirect someone back into the community and out of the water. And they are important. Research consistently shows that the longer someone stays detained before trial, or the more often they cycle back into jail during their case, the greater the likelihood of receiving a conviction and being incarcerated.

But those calm edges are not equally accessible to everyone. Small differences in early decisions: who is stopped, who is charged, who is released pretrial, and who is offered diversion, quietly shape the strength and direction of the current.

In our recent research report, supported by the John D. and Catherine T. MacArthur Foundation, which leads the Safety and Justice Challenge, a national initiative to reduce overreliance on and misuse of jails, we examined how these decisions accumulate across key milestones in the case process. What we found is that disparities rarely appear all at once. Instead, they build gradually as people move downstream through the system. As the current strengthens, options narrow and consequences become more severe, carrying some individuals much closer to the system’s most serious endpoint, like a river pulling toward a waterfall.

A Black man can’t access the calm edges of the riverbed and faces the risk of going over the waterfall. Illustration by: Franco Zacha

Key Milestones in Case Processing

The criminal legal system is not a single decision. It is a chain of decisions and key milestone moments. Every case moves through a sequence of decision points and each step is informed by the one before it and shapes what comes next.

Several key milestones act like currents in a stream, shaping how cases move through the criminal legal system. Charging is the first major decision point, where prosecutors determine what charges to bring based on probable cause from police. Jurisdictions handle this process differently where some use charging on information or pre-filing charging, meaning the initial charges presented to the court may not be the final ones. Even when fact patterns are similar, who gets charged with what can vary widely. Some people face multiple charges or more serious versions of an offense, while others face fewer or less severe allegations. The next milestone is pretrial release, typically decided at arraignment, when judges determine whether someone will remain in jail or return to the community while their case proceeds. Prosecutors often make recommendations, and defense attorneys (when present) argue for the least restrictive release possible. Yet roughly half of jurisdictions do not guarantee defense representation at this stage.

Additional milestones continue to shape the direction of a case. Diversion programs, usually administered by prosecutors, allow some individuals to complete requirements in exchange for dismissed or reduced charges, though eligibility is often limited to people with lower-level charges or minimal prior records. Most cases then move toward plea negotiations and conviction, since only about three percent of cases go to trial nationally. These agreements between prosecutors and defense attorneys determine charges and recommended outcomes before being presented to the court. The final milestone is sentencing, where judges determine the consequences of a conviction within the bounds of state or federal guidelines. Taken together, these moments show how cases move through a series of interconnected decisions, each shaping the options available at the next stage.

The front door matters

Our research focused on understanding how people experience these key milestones in three jurisdictions: San Francisco County, California, Orleans Parish, Louisiana, and Pima County, Arizona. Despite different geographic locations, political contexts, and populations, a key finding emerged across the sites: Black men and women were more likely to experience more punitive outcomes at nearly every stage of case processing.

Importantly, across multiple jurisdictions, Black individuals face significantly higher odds of being detained before trial, even when legal factors and demographics are taken into account.

In some jurisdictions, those odds are as much as 54 percent higher than for similarly situated individuals. Pretrial detention often destabilizes people’s lives. It disrupts employment, family responsibilities, and housing. It can also increase pressure to accept plea deals, even when the evidence is weak.

These disparities also play out differently across gender. Intersectional findings from San Francisco reveal a distinct pattern for Black women: they face 24 percent higher odds of experiencing the most restrictive pathway in the system: pretrial detention during case processing leading to conviction and incarceration.

Our research shows that Black individuals, particularly Black men, are overrepresented at some of the most consequential stages, including charge, pretrial detention, and incarceration, yet notunder represented at prosecutor-led diversion, one of the few streams where someone can exit the system permanently without a conviction. But when access to this pathway is uneven, the opportunity to leave the stream early is not equally available to everyone.

Importantly, this disparity is not only about who is offered diversion but it is also about who is eligible for it in the first place. Many diversion programs rely on eligibility criteria such as “no prior convictions” or limited conviction history. Yet those requirements can unintentionally exclude individuals who have been more heavily policed or more frequently arrested in the past. As a result, the very people most affected by earlier system contact are often structurally filtered out of one of the few points where a case can truly exit the system. That reality raises an important question for policymakers and practitioners: who are diversion programs actually reaching?

Outcomes are shaped long before arrest

Understanding racial disparities across case processing requires looking beyond the courtroom. Historical community disinvestment and intensive policing in neighborhoods shape who enters the system. Heavily policed neighborhoods experience more arrests, which can funnel individuals into jails early in life, keep them cycling in and out of jail, and make them ineligible for many of the opportunities that might otherwise slow or interrupt deeper system involvement. By the time someone reaches the courtroom, they may already be navigating a pathway shaped by structural inequities.

The compounding effect begins early, which is why intervening at the front end of case processing, particularly through thoughtful and accurate charging decisions, is so important. Initial charges often determine what options are available later in the process. They can shape whether someone is eligible for pre-arraignment release, the least restrictive forms of pretrial release such as release on recognizance, access to diversion programs, or even the sentencing ranges judges must follow.

Solutions are emerging

Although these patterns of racial disparity are deeply woven into the fabric of local criminal legal systems, it is important to recognize that they are not set in stone.

Recognizing the urgent need for reform, jurisdictions across the country are actively experimenting with new strategies designed to interrupt racial disparities. Rather than waiting for issues to fully manifest later in the justice process, these communities are seeking to intervene at the very beginning, exploring innovative approaches that could fundamentally reshape outcomes for countless individuals.

For example, in Harris County, Texas, a new collaborative approach has emerged. Instead of police officers unilaterally filing charges, they now consult with prosecutors before proceeding with certain cases, working together to carefully confirm that probable cause truly exists. This extra step introduces a layer of scrutiny and deliberation at the earliest stage, aiming to prevent cases lacking sufficient evidence from ever gaining traction within the system.

As a result of these consultations, many charges are declined before they ever enter the court system. The process becomes a critical filter, ensuring that only cases with a solid evidentiary basis move forward. This not only preserves valuable personnel and material resources across prosecutors, defense, and courtrooms but also protects individuals from unnecessary entanglement in the justice system, which can have lasting consequences even if charges are eventually dismissed.

Other jurisdictions are taking different but complementary steps to intervene earlier in case processing and keep more pathways open for people to exit the system. Many are encouraging law enforcement and prosecutors to establish clear probable cause before bringing charges and to avoid overcharging, a tactic that often pressures defendants into accepting plea deals regardless of their actual culpability. Several sites are also expanding pre-arraignment release, allowing more people to return to the community before their first court appearance. States and local jurisdictions are also experimenting with broader structural reforms. Illinois, for example, eliminated cash bail, replacing it with preventive detention hearings where judges determine whether someone should be held in jail based on the seriousness of the charges and concerns about community safety. Other jurisdictions are expanding eligibility for diversion programs, recognizing that strict criteria—such as requiring no prior convictions—often exclude the very individuals most affected by earlier system contact. In Lucas County, Ohio, defense attorneys are partnering with community-based organizations to connect people held in pretrial detention with services and programming, then using that engagement to advocate for pretrial release and, in some cases, case dismissals.

This shift represents a growing recognition that the seeds of fairness or inequity are often sown at the very beginning of the process. Together, these efforts focus on the same goal: creating more opportunities for people to step out of the current early, before prolonged detention or escalating charges push them further downstream.

Data systems remain a major barrier

Data plays a critical role in helping jurisdictions understand both the scale of racial disparities and where they emerge in case processing. Data can also provide insights into whether reforms are changing outcomes overtime. However, many jurisdictions still lack the comprehensive, high-quality data needed to fully understand where and how disparities arise. Without reliable information, it becomes nearly impossible to see the full picture or measure the true impact of reforms, no matter how effective they may be.

Even when good data exists, researchers and policymakers often struggle to link records across different stages or points within the justice system. Disconnected databases and incompatible systems create silos that obscure the connections between decisions made at arrest, detention, charging, and beyond. This fragmentation makes it difficult to trace the full journey of a case—or of an individual—through the system.

Without integrated data systems, the justice process remains opaque. It becomes exceedingly difficult to track how decisions at one stage may influence, amplify, or mitigate outcomes at another. This lack of connectivity undermines efforts to understand the whole system, making it challenging to implement reforms that address the root causes of disparity rather than just their symptoms.

Although better data alone cannot solve the complex problem of racial disparities, it serves as a crucial tool for pinpointing exactly where disparities emerge and for evaluating whether specific reforms are effective. With clearer insights, policymakers and advocates can target their efforts more precisely and hold systems accountable for meaningful progress.

What this research reinforces

Our research across these three communities confirms something communities have long understood: no single decision within the system is solely responsible for the racial disparities we see today. Rather, disparities emerge through a series of local policies and practices where each decision, however small, contributes incrementally to a web of outcomes that can reinforce inequality over time. Because these policies differ across jurisdictions, no single solution will fit every community. Yet, a clear pattern persists across places: Black and Brown communities are more likely to enter the system and face fewer exits once inside.

It is the system, with its interconnected processes and decision points, that ultimately generates and perpetuates disparities. But just as systems create these challenges, they also offer opportunities for systemic solutions.

By focusing reform efforts on high-impact decision points—charging practices, early pretrial release approaches, diversion eligibility—and by strengthening the underlying data systems that inform these decisions, jurisdictions can begin to interrupt the well-worn pathways where disparities tend to accumulate. These targeted interventions represent the first steps toward dismantling the structural forces that produce unequal outcomes.

None of these solutions are easy.

But if we are to achieve a justice system that is both fair and effective, the time to act is now. By reforming charging and initial pretrial release, rethinking decision-making at every stage, and building data systems that provide transparency and accountability, we can take meaningful steps towards equitable community safety. The most powerful reforms happen upstream where small shifts in policy and practice can change the direction of the current itself. When we reshape decisions at the front end of the system, we don’t just affect one moment in a case; we alter the flow downstream, reducing the force that carries people toward the waterfall of incarceration.