Answer: To the extent that the problem is connected to long-standing, deep-rooted patterns of socioeconomic disadvantage, disparate incarceration defies easy resolution. Additionally, reformers must take care that reduced minority incarceration is not pursued at the cost of increased crime and violence in minority communities. While no good quick fixes seem available, it is possible that some progress might be achieved by raising awareness of implicit racial bias among those who work in the criminal justice system and by requiring policy makers to consider the racial impact of any proposed changes to criminal justice policies.

The Facts: Disparate incarceration means that the costs of mass incarceration—including the stigmatization and diminished life prospects experienced by individuals who have been incarcerated, the financial harm to families, and the psychological and developmental harms to children—are concentrated disproportionately in historically disadvantaged communities, thus perpetuating and compounding legacies of discrimination and marginalization. Disparate incarceration also seems to contribute to the distrust of police and the legal system that exists in some minority communities and that may diminish the effectiveness of law Page 201  |  Top of Articleenforcement efforts in those communities. Disparate incarceration thus seems a social problem worthy of policy makers’ attention.

The problem, however, defies easy resolution. As detailed in Q35, much of the phenomenon of disparate incarceration can be attributed to disparate offending. The interests of disadvantaged minority communities would hardly be well served by policy changes that had the effect of largely eliminating accountability for violent and other serious crimes in those communities. Reducing minority crime rates may need to be a central part of any strategy that aims to achieve large reductions in minority incarceration. Yet, after decades of research, policy makers still lack a “silver bullet”—proven, easily replicable methods to achieve large, sustained improvements in public safety in high-crime, socioeconomically disadvantaged neighborhoods. There seems only so much that policing can accomplish—especially if we do not wish to see arrest, conviction, and incarceration rates increase. If the goal is to reduce both crime and incarceration on a large scale, there may need to be substantial increases in social investment in disadvantaged communities in pursuit of broad improvements in areas such as education, public health, and employment.

Even to the extent that disparate incarceration results from unfair decisions within the criminal justice system, effective policy solutions still seem elusive. For one thing, as seen in Q35, disparities are not caused by decisions at just one stage of the criminal process, but rather arise in relatively modest ways throughout the system. For another, conscious, deliberate racial discrimination does not seem to be a primary driver of unwarranted disparities. A more significant problem may be implicit racial bias, which, by its very nature as a phenomenon of the unconscious, seems harder to address. Adding to the difficulty is the unfortunate correlation of race with various socioeconomic considerations that are traditionally—and arguably quite appropriately—a routine part of the decision making by police, prosecutors, judges, and corrections officials. For instance, the fact that a defendant has a job is typically regarded as favorable if the defendant is being considered for pretrial release, for admission into a diversion program as an alternative to conventional prosecution, or at sentencing for a community-based sanction in lieu of incarceration. However, since black unemployment rates are substantially higher than white, these seemingly race-neutral practices will tend to exacerbate disparate incarceration. Rooting out the consideration of all factors that have a disparate impact in all criminal justice decision making would be an extremely difficult project, and might well create worse problems than it solves, for example, if most diversion programs then proved unworkable and had to be discontinued.

The history of the federal sentencing guidelines, which were adopted in part to address racial disparities, provides a cautionary tale. Going into effect in 1987, the guidelines greatly reduced judicial sentencing discretion, discouraged the consideration of socioeconomic and related factors by the sentencing judge, and made sentencing decisions overwhelmingly dependent on just offense severity and criminal history ( O’Hear, 2017 , 99-100). By taking off the table many traditionally recognized mitigating circumstances, the guidelines contributed to a sharp increase in the severity of federal sentences in the 1980s and 1990s, and a related explosion in the size of the federal prison population. Yet, even at those costs, the guidelines proved unsuccessful at wringing racial disparities from the federal sentencing system. Indeed, the gap in average sentences between black and white defendants actually increased under the guidelines (109). The most important driver of this disparity was likely the guidelines’ particularly harsh treatment of crack cocaine offenses, which were disproportionately associated with black defendants (112). This provides a good example of the potential for facially race-neutral sentencing factors to generate racial disparities. In 2002, on the eve of major changes to the federal sentencing system that reduced the importance of the guidelines, the average sentence in crack cases (mostly involving blacks defendants) was 119 months, while the average sentence in powder cocaine cases (mostly involving white defendants) was only 78 months. There is no reason to think that the guidelines’ crack-powder disparity was intended to exacerbate racial disparities, but that was precisely the impact. Racial effects can creep into even the most aggressive efforts to control discretion and eliminate bias in the criminal justice system.

The unintended or incidental character of racial disparities in incarceration helps to explain why constitutional litigation has proven ineffective as a reform strategy. The Fourteenth Amendment to the U.S. Constitution requires “equal protection of the laws,” but the Supreme Court has interpreted this provision such that a policy’s disparate racial impact does not, in itself, make the policy unconstitutional. In order to have a policy overturned on equal protection grounds, a person must normally show that the policy is racially discriminatory on its face or otherwise reflects a conscious intent to discriminate. Thus, for instance, in Johnson v. California (2005 ), the Supreme Court supported a lawsuit challenging a California policy requiring the racial segregation of some prison inmates. However, in McKleskey v. Kemp (1987 ), the Supreme Court turned aside an equal protection challenge to the death penalty that was based on a statistical demonstration of racial disparities in its administration. In contrast to Johnson, where there was a policy that openly discriminated Page 203  |  Top of Articleon the basis of race, the disparate impact in McKleskey did not result from any policy that was discriminatory on its face; nor was there otherwise proof of intentional discrimination in the adoption or administration of the death penalty.

Arguably, the best response to racially disparate incarceration would be reforms designed in a general way to minimize the unnecessary incarceration of all defendants and to diminish the long-term stigma and collateral consequences of all convictions. In effect, such reforms would mean that racial disparities in incarceration simply matter less. One imagines, for instance, that there would be far less consternation over a prison population that is two-thirds people of color if the overall size of the population were 500,000, instead of the actual 1.5 million.

On the other hand, some incarceration-reducing reforms threaten to increase racial imbalances. Drug treatment courts (DTCs) provide a good illustration. As detailed in Q10, DTCs are found throughout the United States and are generally regarded as a successful innovation that helps to reduce participants’ drug use and recidivism rates. However, to the extent that DTCs serve to divert some offenders from prison, those who are white are more likely to benefit than those of color ( National Association of Criminal Defense Lawyers, 2009 , 42-43). This is because people of color are more likely to have a disqualifying criminal history that keeps them out of a DTC, and because, even if admitted, people of color are more likely to suffer from unemployment, poverty, low educational attainment, and other risk factors that diminish the likelihood of successful treatment. Unfortunately, failing out of a DTC often results in even more incarceration than would have been imposed through conventional case processing. In any event, DTCs throughout the country have struggled to maintain diverse “client” populations. All too often, they are perceived to be a pathway out of incarceration for white offenders only. No matter how well intentioned and successful on their own terms, incarceration-reducing reforms that disproportionately benefit white offenders will do little to reduce racial tensions related to mass incarceration, and may even prove counterproductive.

There are few serious reform proposals that get more directly at the problem of racial disparities in incarceration. Moreover, those reforms that have been adopted in a few jurisdictions in recent years are of uncertain effectiveness. Still, pending further research on their impact, they may merit consideration for replication elsewhere. First, a handful of states have adopted policies requiring the preparation of “racial impact statements” when changes to criminal laws are considered ( Ghandoosh, 2015 , 20). These are similar, for instance, to environmental impact statements, which Page 204  |  Top of Articlerequire policy makers to assess and consider the consequences of their decisions on the natural environment. Likewise, racial impact statements aim to ensure that policy makers understand the potential for proposed criminal justice policy changes to exacerbate racial disparities.

Second, police departments, courts, and other criminal justice agencies can work to mitigate implicit racial bias. Training programs have been developed that help people to better understand their unconscious biases, and some research indicates that such programs can have a positive impact ( Nellis, 2016 , 13). It may also help for agencies to work on recruiting and retaining more diverse workforces. Some evidence suggests, for instance, that black police officers may be less prone to racially biased attitudes than whites ( Ghandoosh, 2015 , 21).

Finally, two states have adopted (although one subsequently repealed) a “racial justice act,” which allows defendants who have been sentenced to death to challenge their sentences based on statistical demonstrations of disparities in the administration of the death penalty ( Spohn, 2017 , 183). In effect, these statutes overrule the Supreme Court’s decision in McKleskey v. Kemp (noted earlier). It is also possible to imagine more broadly targeted racial justice acts that would force consideration of racial disparities in sentencing outside the death context. For instance, there might be statistically based challenges to life sentences, too, or even more ambitiously to any prison sentence.