Originating in a period of rising crime rates and social foment and driven by punitive sentencing policy, the steep increase in incarceration in the United States was carried out with little regard for an objective evaluation of benefits or possible harms. This committee was charged with assessing the causes of the steep increase and the consequences that followed.
In this chapter, we first summarize the findings and state our conclusions from the review of the evidence presented in the preceding chapters. We next consider the implications of these findings for public policy. In so doing, we draw on the long-standing normative principles of jurisprudence and public policy that historically guided deliberations on the use of incarceration as a response to crime. Our findings and conclusions, supplemented by these normative principles, lead us to the main recommendation that federal and state policy makers should take steps to significantly reduce the rate of incarceration in the United States. We then make specific suggestions for reform in the areas of sentencing policy, prison policy, and social policy. The next section offers recommendations for further research. The final section presents concluding thoughts.
The U.S. rate of incarceration in 2007 was more than four and one-half times the rate in 1972 (Chapter 2 details these trends). By 2012, the prison
and jail population had grown to 2.23 million people, and the United States had by far the highest reported rate of incarceration in the world. Today, adult incarceration rates of the Western European democracies average around 100 per 100,000, and in the common law countries of Australia and Canada, the rates are only slightly higher. The U.S. rate in 2012 was seven times higher, at 707 per 100,000. At this level of penal confinement, the United States (accounting for about 5 percent of the world’s population) holds close to 25 percent of the global incarcerated population.
CONCLUSION: The growth in incarceration rates in the United States over the past 40 years is historically unprecedented and internationally unique.
The growth of incarceration rates, beginning in 1972, followed a tumultuous period of social and political change (see Chapter 4). From 1962 to 1972, the annual number of homicides had climbed from 8,530 to 18,670. Homicide was just one indicator of declining public safety, as the overall violent crime rate doubled in that same decade (Maguire, n.d., Table 3.106.2011). If rising crime were the only new social trend of the 1960s, the link between crime and incarceration might be clear-cut. But political activism and race relations also came to a boil. Civil rights action and conservative reaction produced a contentious and sometimes violent politics that blurred the line between protest and disorder. The civil rights acts themselves upended the racial order of the south and outlawed discrimination in labor and housing markets across the country. In short, the period of rising crime accompanied a period of intense political conflict and a transformation of U.S. race relations.
Cities also were transformed. Riotous unrest culminated in the Kerner Commission (1968) report that surveyed dozens of incidents of disorder in 23 cities. The Commission, struggling to untangle a complex mix of crime, racial inequality, and politics, famously concluded that the nation was moving to “two societies, one black, one white—separate and unequal.” Rising crime and disorder were accompanied by declining manufacturing sector employment in inner cities, classically described in William Julius Wilson’s (1987) The Truly Disadvantaged. In Wilson’s analysis, the outmigration of whites and working class blacks left behind pockets of concentrated disadvantage. These poor, racially segregated neighborhoods were characterized not just by high rates of crime but also by an array of other problems, including high rates of unemployment and widespread single parenthood. It was in these neighborhoods, decades later, where the effects of incarceration were felt most strongly.
Historic changes in politics, race relations, and urban life provided the context in which policy makers wrestled with the crime problem. Rising
crime rates gained a prominent place in national policy debates. Crime and race sometimes were conflated in political conversation. Public policy of the 1960s was moving in a liberal direction, through an expansion of social programs and stronger rights for criminal defendants and prisoners, but these measures did not appear to stem the rise in crime. The debates about crime unfolded in a setting where crime policy was mostly made at the state and local levels. Local elected officials—including state legislators who enacted sentencing policies and, in many places, judges and prosecutors who decided individual cases—were highly attuned to their constituents’ concerns about crime. Under these conditions, punishment policy moved in a more punitive direction.
Across all branches and levels of government, the policies governing criminal processing and sentencing were reformed to expand the use of incarceration. Prison time was increasingly required for lesser offenses. Time served was significantly increased for violent crimes and for repeat offenses. Drug crimes, particularly street dealing in urban areas, became policed and punished more severely (see Chapter 3). These changes in punishment policy—the enactment of mandatory sentence laws, long sentences for violence and repeat offenses, and intensified criminalization of drug-related activity—were the main and proximate drivers of the growth in incarceration.
CONCLUSION: The unprecedented rise in incarceration rates can be attributed to an increasingly punitive political climate surrounding criminal justice policy formed in a period of rising crime and rapid social change. This provided the context for a series of policy choices—across all branches and levels of government—that significantly increased sentence lengths, required prison time for minor offenses, and intensified punishment for drug crimes.
When evaluating criminal justice policies, researchers and policy makers may turn first to the effects on crime rates. Most studies conclude that rising incarceration rates reduced crime, but the evidence does not clearly show by how much. A number of studies also find that the crime-reducing effects of incarceration become smaller as the incarceration rate grows, although this may be reflecting the aging of prison populations. As with many rigorous assessments of large historical events, a high level of scientific certainty about the effects of increased incarceration rates is elusive. The relationships between incarceration, crime, sentencing policy, social inequality, and the dozens of other variables that describe the growth of incarceration are complex, variable across time and place, and mutually determining.
Because of the great scientific challenge of separating cause and effect from the surrounding array of social forces, the impact of increased incarceration on crime cannot be calculated precisely. There is only weak evidence that increased prison populations from the 1970s to the 2000s led to large aggregate reductions in crime rates (see Chapter 5).
CONCLUSION: The increase in incarceration may have caused a decrease in crime, but the magnitude is highly uncertain and the results of most studies suggest it was unlikely to have been large.
Although increasing prison admissions and increases in time served in prison both fueled incarceration rates, research has best illuminated the effects of time served. Long sentences are characteristic of the period of high incarceration rates, but research indicates it is the certainty of apprehension, not an increase in the duration of long sentences, that actively deters would-be offenders. The marked decline in offending with age also means that the incapacitation effect of long sentences is likely to be small (see Chapter 5).
CONCLUSION: The incremental deterrent effect of increases in lengthy prison sentences is modest at best. Because recidivism rates decline markedly with age, lengthy prison sentences, unless they specifically target very high-rate or extremely dangerous offenders, are an inefficient approach to preventing crime by incapacitation.
What are the effects of increased incarceration on prisoners and their families? The committee began to consider these effects by reviewing research on prison conditions and the health of the prison population. Increased rates of incarceration may have altered prison conditions in ways that are, on balance, harmful to some prisoners and undermine their chances of living a normal life when released. Although the rate of lethal violence in prison declined, increased rates of incarceration were accompanied by overcrowding, decreased opportunity for rehabilitative programs, and a growing burden on medical and mental health services. Psychological research shows that many aspects of prison life—material deprivations, restricted movement, the absence of personal privacy, and high levels of uncertainty and fear, for example,—are significant sources of stress that damage the emotional well-being of some of those incarcerated (see Chapter 6).
Medical and epidemiological research indicates that these stressors are focused on a population that carries a high burden of disease and experiences a high rate of mental illness. Incarceration is associated with overlapping afflictions of substance use, mental illness, and risk for infectious diseases (HIV, viral hepatitis, sexually transmitted diseases, and others).
People who have been inadequately treated while in prison and after release face higher risks of suicide, relapse to drug addiction, and drug-overdose death than the general public (see Chapter 7).
Outside of the prison, incarceration is strongly correlated with negative social and economic outcomes. The people who have been incarcerated have very low earnings, high rates of unemployment, and experience little earnings growth over the life course. Because of school failure, criminal involvement, mental health problems and related challenges, those who go to prison have very poor economic opportunities even before incarceration. These pre-existing traits make it difficult to precisely estimate the economic effects of incarceration. Still, the experience of incarceration may undermine the productivity and employment opportunities of those incarcerated. Controlled experiments further show that job seekers with criminal records face extreme reluctance from prospective employers, and criminal records can have lasting employment consequences (see Chapter 8).
Family instability in the lives of former prisoners and behavioral problems among their children is also well-established. Studies have documented the large increase in the number of children with incarcerated parents and have described the range of poor outcomes that may be associated with having a parent in prison. The evidence shows a strong relationship between a father’s incarceration and family hardship, including housing insecurity and behavioral problems in children, though it is difficult to draw causal inferences about that relationship. Studies that focus exclusively on incarcerated men have found that partners and children of male prisoners are particularly likely to experience adverse outcomes if the men were positively involved with their families prior to incarceration (see Chapter 9).
Beyond the research on individuals and families, the committee also explored the consequences of high incarceration rates for communities (see Chapter 10). The escalating rates of incarceration have been concentrated in poor and largely urban African American and Hispanic communities. As a result of the shift in penal policy toward greater use of prison, large number of residents of these communities, mostly poor men with little schooling, have experienced the cycle of arrest, detention, imprisonment, release, and supervision in the community, often followed by a return to prison for violating parole conditions or for a new crime. Given the challenge of drawing strong causal inferences, it is difficult to determine precisely the impact of this high concentration of the growth in incarceration on the levels of crime in these communities. Yet the evidence is clear that the large increase in incarceration has been concentrated in high-crime, disadvantaged minority communities and has transformed the character of life in poor urban neighborhoods.
The committee also reviewed evidence on the consequences of the growth of the prison population for civic and social life more broadly (see Chapter 11). High rates of incarceration are associated with lower levels of civic and political engagement among former prisoners and their families and friends. The quality of important demographic, political, and socioeconomic databases were compromised. High incarceration rates served as a gateway to a much larger system of stigmatization and legal, political, and social marginalization. The high cost of the penal system for state budgets also transformed incarceration into a major function of state government.
In summary, over the decades reviewed in this report, policy makers aggressively promoted measures that greatly increased incarceration rates. They adopted imprisonment as a key tool for crime control. Very long sentences became commonplace for repeat offenses and serious violence, and drug offenses were prosecuted more aggressively. Prison time was often mandated for offenses that previously were punished through community sanctions. Throughout this significant shift in penal policy, the possibility of negative social consequences was either not considered or disregarded. Nor did policy makers adequately consider the possibility that the crime reduction effects of the prison expansion might be modest. Yet the evidence reviewed by this committee indicates that the large increase in incarceration rates probably did not substantially reduce crime. At the same time, the available evidence suggests the prison expansion may have resulted in negative effects on prisoners, their families and the larger society. The committee recognizes that a great deal of scientific uncertainty characterizes scholarly efforts to assess these two effects. In carrying out its charge to assess the available evidence, and cognizant of this uncertainty, the committee concludes that, despite a profound change in penal policy in the U.S., large benefits failed to clearly materialize, and social harm may have resulted.
CONCLUSION: The change in penal policy over the past four decades may have had a wide range of unwanted social costs, and the magnitude of crime reduction benefits is highly uncertain.1
Race is a theme that runs through many of the chapters of this report. Racial disparities in incarceration have been observed since the relevant data were first available in the nineteenth century. Incarceration rates escalated rapidly for African Americans to levels six and seven times higher than those of whites, and reached extraordinary levels among young African American men with little schooling. Estimates indicate that by 2010, one-third of all African American male high school dropouts under age 40
1See Appendix A for a supplementary statement by Ricardo Hinojosa on this and other similar committee findings.
were in prison or jail, compared to an incarceration rate of 0.7 percent in the population as a whole (see Chapter 2). Much of the significance of the social and economic consequences of incarceration is rooted in the high absolute level of incarceration for minority groups and in the large racial and ethnic disparities in incarceration rates. Research on the spatial distribution of incarceration indicates that prisoners are overwhelmingly drawn from poor minority neighborhoods that also suffer from an array of other socioeconomic disadvantages. In the era of high incarceration rates, prison admission and return became commonplace in minority neighborhoods with high levels of crime, poverty, family instability, poor health, and residential segregation (see Chapter 10). Large racial disparities in incarceration focused any negative effects of incarceration disproportionately on African Americans, the poor in particular, and transformed their collective relationship to the state.
CONCLUSION: People who live in poor and minority communities have always had substantially higher rates of incarceration than other groups. As a consequence, the effects of harsh penal policies in the past 40 years have fallen most heavily on blacks and Hispanics, especially the poorest.
The findings and conclusions presented here do not easily lend themselves to a simple calculation of costs and benefits. The policies that produced very high rates of incarceration grew out of a historical period of rapid change and social conflict. By greatly expanding the use of penal confinement, the policies charted a new direction for the American criminal justice system. No other Western democracy went so far down this path. Through the 1990s and 2000s, crime rates fell significantly, but the evidence indicates it is unlikely that the rise in incarceration rates played a powerful role in this trend. Against weak evidence for large benefits, there is also the chance of significant social costs for individuals who are incarcerated, their families, and communities. The strong correlation of incarceration with unemployment, poverty, family disruption, poor health and drug addiction is very clear. Causality is harder to disentangle, but experiments and statistical adjustment point to the real possibility of negative social effects. These correlations and negative effects are concentrated almost entirely in poor, especially poor minority, communities. For policy and public life, the size of the effects of incarceration may be less important than the overwhelming evidence of the correlation between very high levels of incarceration, race, poverty, and the myriad of accompanying social problems.
The committee struggled with the meaning of these conclusions. Across the many perspectives and specializations represented by our members, we agreed that basic questions of justice cannot be answered by science alone, particularly in this context where the problem is complex, many different kinds of evidence—quantitative and qualitative—are relevant, and scientific certainty was often elusive. The decision to deprive another human being of his or her liberty is, at root, anchored in beliefs about the just relationship between the individual and society and the role of criminal sanctions in preserving the social compact. Thus, good justice policy is necessarily based on a combination of empirical research and explicit normative commitments.
CONCLUSION: In the domain of justice, empirical evidence by itself cannot point the way to policy, yet an explicit and transparent expression of normative principles has been notably missing as U.S. incarceration rates dramatically rose over the past four decades. Normative principles have deep roots in jurisprudence and theories of governance and are needed to supplement empirical evidence to guide future policy and research.
To frame the policy implications of the evidence presented in this report, we return to the normative principles first presented in Chapter 1 and elaborated in Chapter 12. The committee noted that, over the past 40 years, principles that would restrain the use of prison as a response to crime were given less weight in public discourse than the crime control mission for punishment. The principle of proportionality—that the sanction imposed for violation of the criminal law should be proportionate to the seriousness of the crime—is challenged by harsh sentences for minor offenses. The principle of parsimony—that the criminal sanction imposed for an offense should be sufficient but not greater than the punishment necessary to achieve sentencing goals—is inconsistent with overly long sentences. The principle of citizenship—the notion that the consequences of a prison sentence should not be so severe as to substantially weaken one’s status as a member of society—is tested by conditions of confinement that can be considered inhumane. Finally, the principle of social justice, as applied to prisons—that prisons should promote, not diminish, a fair distribution of resources, rights, and opportunities—is strained when incarceration becomes pervasive in poor and minority communities.
In weighing the scientific evidence on the causes and consequences of the high rates of incarceration in the United States, and then considering the implications of that evidence for public policy, the committee found it instructive to refer to the principles that govern the use of imprisonment for crime control and define the proper role of prison in a democratic society. The committee recognizes that a range of values might influence society’s
response to crime. The imposition of the criminal sanction is considered a validation of the social compact. The prevention and control of crime is also recognized as a long-established purpose of the criminal justice system. As is documented in Chapter 12, striking the appropriate balance between these competing values lies at the heart of the policy discourse in a democracy. Yet, Chapter 12 also shows that recent policy discussions have retreated from the principles that constrain the power of the state to punish, respect the human dignity of persons incarcerated, and are troubled by the intimate connection between prisons, racial inequality and poverty. As the committee considered the implications of its findings and conclusions, we affirmed the importance of reviving these principles and striking a new balance in the nation’s penal policies.
The growth of the prison population can be traced to policies expanding the use of imprisonment for felony convictions, imposing longer sentences on those committed to prison, and intensifying punishment for the sale and use of drugs. Proponents of those policies argued that more prison and longer sentences would reduce crime. The committee concluded that research indicates that the large increase in incarceration rates has not clearly yielded sizable reductions in crime. Furthermore, while the research does not clearly indicate the magnitude of the impact of incarceration on crime, there is strong evidence that increasing long sentences has promoted neither deterrence nor incapacitation.
The cost of expanding the penal system has been substantial. The financial costs are borne by taxpayers, who provide funding for local jails, state and federal prisons, and the operations of the larger criminal justice apparatus associated with institutions of incarceration. The opportunity cost is also considerable. Spending on prisons diverts resources from more effective public safety strategies, services for crime victims, or programs designed to help achieve effective reintegration of people who have been released from prisons. The burden of incarceration also falls on the millions of incarcerated individuals and, the evidence suggests, their families and neighborhoods. More broadly, as a consequence of the unprecedented rise in incarceration rates, the larger society bears the costs of an expanded share of the population that struggles with the stigma and negative effects of the prison experience.
To recognize the high cost of incarceration does not deny that, in specific cases, prison sentences are an appropriate societal response to the crimes committed. Nor does this assessment of the costs of prison overlook the fact that, in certain cases, incarceration will prevent crime. Similarly, incarceration has certainly improved life for some of those sent to prison.
Yet the weight of the scientific evidence on the consequences of high rates of incarceration, when viewed in light of the principles of proportionality, parsimony, citizenship, and social justice outlined above, suggests that too many people are in prison in the United States and that, overall, their sentences are too long.
The nation cannot yet accurately estimate the long-term consequences of imprisoning so many of its citizens. However, the current evidence is troubling and leads to our overarching policy recommendation.
RECOMMENDATION: Given the small crime prevention effects of long prison sentences and the possibly high financial, social, and human costs of incarceration, federal and state policy makers should revise current criminal justice policies to significantly reduce the rate of incarceration in the United States. In particular, they should reexamine policies regarding mandatory minimum sentences and long sentences. Policy makers should also take steps to improve the experience of incarcerated men and women and reduce unnecessary harm to their families and their communities.
Based on our analysis of the evidence, we urge policy makers at the state and national levels to reconsider policies in three distinct domains: (1) sentencing policy, (2) prison policy, and (3) social policy. Doing so will require political will. Just as the expansion of the penal system was driven by changes in policy, it must be reversed through policy choices. Most fundamentally, reversing course will require state and federal policy makers to significantly reform sentencing policy. More specifically, as discussed below, they should consider reforms to the current regime of long sentences, mandatory minimum sentences, and the overall enforcement strategies regarding drug laws. In addition, reversing course will require changes in the use of discretion available under current laws to police, prosecutors, parole decision makers, community corrections officials, and other actors in the criminal justice system. The development of new penal policies will depend, in turn, on a new public consensus that current policies have been, on balance, more harmful than effective and are inconsistent with U.S. history and notions of justice. Making this case to the public will require determined political leadership.
Before turning to our suggestions in the three policy domains, we note that, although our overarching recommendation involves issues of sentencing and prison policy, a broader menu of alternatives is necessarily implicated when reforming sentencing policies to reduce incarceration rates. To support the recommended change in policy direction, jurisdictions would likely have to review a range of allied programs, such as community-based alternatives to incarceration, probation and parole, prisoner reentry,
diversion from prosecution, and crime prevention initiatives. Correctional programs, such as cognitive-behavioral therapies aimed at changing criminal behavior, also are likely to be an important part of the needed change in direction (MacKenzie, 2006). Assessing the effectiveness of these programs is beyond this committee’s charge, but we note the importance of viewing the above recommendation in the context of this larger policy framework. We return to this issue in our discussion of research needs below.
The evidence we reviewed does not provide a roadmap for comprehensive sentencing reform. Just as research does not indicate precisely whether a sentence for a specific crime is too short or too long, it does not specify an optimal sentence reduction. Other values may also shape sentencing policy. For example, many sentencing reforms of recent years were intended to reduce racial disparities, and policy makers must be careful not to reverse any resulting gains in sentence proportionality. More important, and consistent with our emphasis on the importance of values and the need for political leadership, we recognize that the details of strategies for reducing incarceration levels will depend on a complex interplay between the public and policy makers. However, the evidence does identify some types of policies that drove the prison buildup; imposed sizable social, financial, and human costs; failed to produce commensurate public safety benefits; and were inconsistent with the normative principles articulated above. Three dimensions of sentencing policy are particularly appropriate for reexamination: long sentences, mandatory minimum sentences, and enforcement of drug laws.
The case for reducing long sentences is compelling. As this report has documented (see Chapters 2 and 3), the steady growth in incarceration rates has been significantly fuelled by longer prison sentences. A variety of statutory enactments have driven these results, including laws imposing truth-in-sentencing, life without parole, and three strikes enhancements. In addition to these changes in sentencing policy, states also have reduced or eliminated the use of discretionary parole release, increased the level of returns to prison for parole violations, reduced the use of “good time” provisions to accelerate release eligibility and cut back on the use of halfway houses, educational release programs and compassionate release options. These policy shifts significantly increased the average time served for a felony conviction.
Yet, as reviewed in this report, research shows that long sentences have little marginal effect on crime reduction through either deterrence or incapacitation. The deterrent value of long sentences is minimal, as the decision to commit a crime is more likely influenced by the certainty and swiftness of punishment than by the severity of the criminal sanction. Research on criminal careers shows that recidivism rates decline markedly with age. Prisoners serving long sentences necessarily age as they serve their time and their risk of re-offending declines over time. Accordingly, unless sentencing judges can specifically target very high-rate or extremely dangerous offenders, imposing long prison sentences is an inefficient way to prevent crime. Finally, the evidence is clear that long prison sentences incur substantial costs to state and federal budgets and will likely add significant future costs as the prison population ages.
Efforts to reduce incarceration rates by reducing long sentences could usefully follow the initiatives undertaken by the federal government and by many states. The U.S. Congress has curtailed the length of the sentence for crack cocaine offenses, and the U.S. Sentencing Commission has retroactively applied sentencing guideline changes for current prisoners (U.S. Sentencing Commission, 2013a). Between 2006 and 2011, 29 states shortened sentences with the aim of reducing prison populations. Some, including Michigan and Mississippi, modified truth-in-sentencing laws to accelerate parole eligibility. California, Indiana, and South Carolina scaled back their three strikes enhancements. Other states reduced sentence lengths by authorizing credits for “good time” that directly affected prison release and parole eligibility. Since 2001, these reforms have been implemented in at least 16 states (Arkansas, California, Colorado, Delaware, Georgia, Kansas, Louisiana, Mississippi, Nevada, North Carolina, Ohio, Pennsylvania, South Carolina, Vermont, Wisconsin, and Wyoming). Although the precise impact of these reforms on average sentence length has yet to be determined, it is reasonable to assume that they have contributed to the overall decline in incarceration rates among the states (Vera Institute of Justice, 2010).
If the policy reforms designed to reduce long prison sentences were prospective and applied only to new convictions, then prison populations would decline only slowly. More immediate effects could be obtained by re-examining and reforming the policies governing release from prison. For example, the state and federal governments could reconsider policies that abolished or restricted discretionary parole release, or curtailed the use of work release, educational release and half-way houses. They could follow the example of states that are considering the establishment or expansion of geriatric or medical parole, also called “compassionate release.” According to the National Conference of State Legislators, between 2000 and 2012, 29 states reformed their rules governing the medical release of inmates;
the changes frequently included compassionate release for the elderly and terminally ill (National Conference of State Legislators, 2012).
Finally, policy makers could implement reforms that would reduce the flow of individuals back to prison because their parole (or probation) has been revoked for technical violations of the conditions of their release. States that have pursued this reform strategy have substantially reduced the flow of people returning to prison. From 2001 through 2010, at least 9 states (Alabama, California, Colorado, Florida, Kentucky, Louisiana, South Carolina, South Dakota, and Washington) enacted these and related legislative measures (Vera Institute of Justice, 2010). Between 2007 and 2013, 13 states (Arkansas, Georgia, Hawaii, Kansas, Kentucky, Louisiana, Nevada, North Carolina, Pennsylvania, South Carolina, South Dakota, Texas, and West Virginia) authorized graduated responses for parole violations, providing front-line officers with a continuum of community-based sanctions to keep more parolees in the community while still maintaining accountability for violations (Pew Charitable Trusts, 2007).
These and other reforms offer a broad menu of policy options that together could significantly reduce the average length of stay in U.S. prisons. They are best combined with a more fundamental re-examination of overall sentence length under state and federal laws. As was noted in Chapter 2, the use of longer prison terms has been a critical driver of high rates of incarceration. A thorough inquiry into the value of longer sentences, including life sentences without the option of parole, resulting in the establishment of new sentencing parameters, could produce substantial reduction in the nation’s prison population in the long run. Cutting back the maximum sentence length for specified offenses would not yield savings in prison time until many years from now, but such a policy reform would be consistent with the normative values outlined in this report and would pose little risk to public safety.
Mandatory Minimum Sentences
Between 1975 and 1995, all 50 states and the U.S. Congress reduced the discretion available to sentencing judges by passing laws requiring imprisonment for a wide variety of offenses. Prior to these enactments, judges could impose noncustodial sanctions such as probation, restitution, or community service. As a result of these new mandatory minimum penalties, custodial sentences have increasingly been imposed for minor offenses. Mandatory minimum sentences were also enacted for drug offenses, murder, aggravated rape, felonies involving firearms and felonies committed by individuals with prior felony convictions. Over the decades covered by this report, mandatory minimums were the most frequently enacted sentencing law change in the U.S. (see Chapter 3).
The stated reason for these sentencing enactments was crime prevention. Policy makers asserted that requiring prison sentences for designated offenses would deter others from committing crimes. Yet the weight of evidence reviewed in this report is strong that such enactments have few, if any, deterrent effects. As is discussed in Chapter 5, three reports of panels convened by the National Research Council have reviewed the research literature on the deterrent effect of such laws and have concluded that the evidence is insufficient to justify the conclusion that these harsher punishments yield measurable public safety benefits. At the same time, there is substantial evidence in the research literature that the imposition of mandatory minimum sentences creates incentives for practitioners—police, prosecutors and judges—to circumvent these penalties.
A broad strategy for reducing the nation’s prison population would also entail review of mandatory minimum sentences in general. A number of states have undertaken such a review. Between 2001 and 2010, 12 states (Delaware, Georgia, Hawaii, Indiana, Louisiana, Michigan, Minnesota, Nevada, New Jersey, New York, Rhode Island, and South Carolina) relaxed their mandatory minimum sentence laws (Vera Institute of Justice, 2010). Statutory reform is not required to reach this result; changes in prosecutorial policy could also change the dynamics of sentencing. In recent instructions to U.S. attorneys, Attorney General Eric Holder has limited the use of mandatory minimums by federal prosecutors for some classes of drug cases (Holder, 2013a, 2013b).
The principles of proportionality and parsimony also call for a reexamination of penal policies mandating imprisonment for minor offenses. Allowing judges to exercise greater discretion in the imposition of a criminal sentence recognizes that any term of imprisonment is a severe sanction that must be imposed deliberately with clear reference to the facts of specific cases. The research also indicates that these reforms would reduce the practice of circumventing mandatory penalties. Finally, the evidence strongly suggests that reforms envisioned here would reduce the nation’s prison population without posing significant risks to public safety.
Enforcement of Drug Laws
The law enforcement strategy known as the war on drugs has been a significant driver of the increase in U.S. incarceration rates. Over the decades of the prison buildup, the incarceration rate for drug offenses increased tenfold—twice the rate for other crimes. Prison admissions for drug offenses grew rapidly, increasing from about 10,000 state prison commitments for drugs in 1980, to about 120,000 admissions by 1990, and peaking at 157,000 admissions in 2008 (see Chapter 5). Yet, as reported in a 2001 report of the National Research Council, these dramatic increases
in imprisonment for drug crimes did not clearly reduce drug use and were accompanied by a significant decline in drug prices from the 1980s to the 1990s. The evidence of high costs—particularly the high costs of incarceration—and of the apparently low effectiveness of the current drug enforcement strategy should compel a fresh look at alternatives. Furthermore, the disparate impact of the war on drugs on communities of color and the high rates of incarceration for drug offenses among African Americans and Hispanics make a reduction in drug-related incarceration an urgent priority.
Reducing incarceration rates requires reassessment of the nation’s war on drugs and the implementation of more effective responses. This reassessment should recognize that abuse of illegal drugs is both a health policy and a justice policy issue. Alternatives that rely more on health care measures might well reduce the social and economic costs of imprisonment and improve public health. A fresh look at drug policy should also confront the realities of current enforcement policies. Over the period of U.S. history covered in this report, the arrest rate for drugs increased substantially—from about 200 per 100,000 adults in 1980 to over 400 per 100,000 in 2009 (see Chapter 5). A more effective response that relied less on arrests would also reduce the reliance on prisons. One promising approach is the law enforcement intervention piloted in High Point, North Carolina. Reflecting principles of focused deterrence, this approach, since replicated widely across the U.S., has shown that a coordinated multiagency response to overt drug markets can effectively address their adverse effects with limited reliance on arrests and therefore reduced reliance on incarceration (Kennedy, 2011).
In addition to high levels of arrests, sentencing for drug offenses has also become more punitive. As mentioned above, reforms to limit mandatory minimum sentences and long sentences for drug offenses would reduce incarceration rates. Recent reductions in incarceration resulting from the reform of U.S. sentencing guidelines for crack cocaine offenses and of New York’s Rockefeller Drug Laws illustrate the potential benefits of reducing the use of incarceration for drug crimes.
Other strategies might be even more effective in addressing the underlying issue of drug use within the contours of the criminal justice system. A number of states and the federal government have taken steps to this end. For example, the development of drug treatment courts and prosecutorial diversion programs offer innovative possibilities that could reduce both drug use and incarceration rates. Recent innovative probation reforms, such as project HOPE (Hawaii Opportunity Probation with Enforcement), which mixes swift and certain sanctions with a regime of drug testing, represent promising efforts to treat problems of drug abuse without relying extensively on incarceration.
A full assessment of the evidence of effectiveness of these and other programmatic innovations is beyond the scope of this report. But what is clear is that reducing the nation’s reliance on incarceration will require a thorough and sustained fresh look at the current approach to drug use and drug crimes.
Other Sentencing Policy Considerations
Although the above measures do not exhaust the options for sentencing reform, we view reduced use of long sentences, review of mandatory minimum sentences, and a revised approach to drug law enforcement as three key main ways in which incarceration could be significantly reduced. Recent reform efforts also have addressed other phases of correctional supervision, notably community corrections. As was mentioned above, a shift in sentencing policy away from reliance on incarceration would necessarily require closer examination of the effectiveness of alternatives to incarceration, including the effectiveness of parole and probation supervision. Similarly, any well-conceived plan for reducing prison populations should consider the effectiveness of short-term and longer-term assistance to parolees. A 2008 National Research Council report on parole policies includes the recommendation that both in-prison and postrelease parole programs be redirected to providing a variety of supports to parolees and others released from prison at the time of release and suggests that no one should leave prison without an immediately available support program and a plan for life postrelease (National Research Council, 2008, p. 82).
Even if incarceration rates are significantly reduced, prisons will remain indispensable to the nation’s system of punishment. In parallel with our general recommendation to reduce the level of incarceration, we urge reduction of the potentially harmful effects of incarceration through reaffirmation of the principle of citizenship and recognition of the public character of penal institutions.
Reaffirmation of the Principle of Citizenship
The principle of citizenship requires that a person’s status as a member of a community not be fundamentally diminished by incarceration. In our view, respect for citizenship demands that punishment by incarceration not be so severe, or have such lasting negative consequences, that the person punished is forever excluded from full participation in mainstream society. Stated affirmatively, the principle of citizenship requires that prisons
operate to respect the autonomy and dignity of those incarcerated, consistent with the goal of administering safe and orderly institutions.
The principle of citizenship suggests a rigorous review of the conditions of confinement and of the legal disabilities and restrictions imposed on those who have been incarcerated. In particular, policies and practices that result in long periods of administrative segregation from the general population, deprivation of meaningful human contact, overcrowding, and unnecessarily high levels of custody all require rigorous review. Prison authorities and legislatures should consider reestablishing the commitment to programming and rehabilitation that was deemphasized during the period of rising incarceration. Conditions of confinement should be reviewed with the objective of increasing prisoners’ chances of reentering society with social relationships intact and better prepared to make a positive, productive transition. Review of these conditions and the policies that regulate them is compelling because, with rare exceptions, all those incarcerated in the nation’s prisons and jails will be released to return to their communities (Travis, 2005). The principle of citizenship also demands a broad review of the penalties and restrictions faced by the formerly incarcerated in their access to the social benefits, rights, and opportunities that might otherwise promote their successful reintegration following release from prison. In short, the state’s decision to deprive a person of liberty temporarily should not lead to permanently diminished citizenship.
The Prison as a Mainstay of Justice
Despite the nation’s great reliance on prisons, the public has few opportunities for a comprehensive and critical examination of the state of penal institutions and how they operate. Compared with other areas of social policy that require similar expenditures of billions of dollars, prisons in many states are subject to relatively little oversight. Through laws, such as the Prison Litigation Reform Act, the role of courts in reviewing conditions of confinement has been restricted (see Chapter 6). Many new prisons were sited in remote areas where they are not readily visible or accessible. The locations and forbidding design of many prisons stand as metaphors for this reality: prisons are far from the public mind and appear closed to public view.
The committee urges policy makers to elevate the public profile and transparency of prisons in recognition of their important role in U.S. society. The broad topics of concern might include the quality of life in prisons, public accountability for expenditures, designation of expected in-prison and postrelease outcomes for prisoners, standards for health and mental health care, limits on the use of administrative segregation, and access by researchers (see Chapter 6). Prison conditions and practices can
be improved over time through continuing outside scrutiny. Policy makers might also consider establishing or reinforcing independent monitoring and oversight of prisons, including independent commissions of the sort that operate in other Western nations.2 U.S. policy makers would benefit from discussions with their counterparts in some other nations where oversight of corrections policies and practices is more rigorous and systematic than is the case in most U.S. prison systems.
If incarceration rates are reduced, many people who would have been incarcerated will continue residing in their communities, often under community supervision. These are largely poor men and women with very low levels of schooling and poor employment histories, many of whom also have histories of substance abuse and mental illness. Their criminal responsibility is real, but embedded in a context of social and economic disadvantage. The close connections between crime, incarceration, and poverty have implications for reforms aimed at reducing high incarceration rates as well as those aimed at reducing criminal behaviors in the first place.
With fewer people in prison, there may be a greater need for social services in the community. It will be necessary to carefully assess available services to determine if there are sufficient quality services in accessible locations to meet the needs of otherwise imprisoned members of the community. Drug treatment, health care, employment, and housing will face especially strong demand. Sustainably reducing incarceration will depend in part on whether communities can meet the needs of those who would otherwise be locked up. If large numbers of intensely disadvantaged prime-age men and women are resituated in poor communities without appropriate social supports, the effects could be broadly harmful and could discredit decisions to reduce the use of incarceration.
Here, the historical example of the deinstitutionalization of the mentally ill offers a cautionary example. Deinstitutionalization, gradually unfolding through the 1950s and 1960s, was originally conceived to be buttressed by an array of community-based mental health services. Instead, state mental hospitals were shuttered, and policy makers were reluctant to
2In some U.S. states, independent oversight is provided by an ombudsman or inspector general. In the United Kingdom, Her Majesty’s Inspectorate of Prisons conducts announced and unannounced inspections assessing prisons against established standards for inmate safety, health care, respectful conduct among staff and inmates, programming, reentry, and administrative segregation (Her Majesty’s Inspectorate of Prisons, 2012). Similar independent oversight through the Council of Europe is provided by the European Committee for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment. An annotated bibliography of writings on independent prison oversight is provided by Deitch (2010).
support community-based programs. Homelessness and other hardships among the mentally ill resulted from the deficit of treatment and other services. Significant reductions in prison population without community planning risks similar problems. Responsibly reducing incarceration will require a parallel expansion of social services.
Policy research on released prisoners emphasizes the importance of employment, housing, and health services (e.g., Travis, 2005; Seiter and Kadela, 2003; Mead, 2011). Employment programs provide a variety of services, from job readiness training to subsidized work (see Chapter 8). Although evaluation research provides uneven evidence that labor market programs can boost employment and reduce recidivism, such programs often are intrinsically valuable when they provide income support and structure the time of program clients. There have been few evaluations of subsidized housing programs.3 However, housing insecurity is common among those at risk of incarceration (see Chapter 9), and like employment programs, housing support often meets serious needs of program clients. Evaluation research also indicates that recidivism can be significantly reduced when social opportunity programs, such as those providing employment, are combined with programs that address criminogenic behaviors (see MacKenzie, 2006, 2012).
The need for health services for released prisoners, including drug and mental health treatment, is similarly serious. The Patient Protection and Affordable Care Act (ACA) presents an unprecedented opportunity to extend health insurance coverage to this population. Improving the health of this and other disadvantaged populations will require continuity of health care from custody to community. Comprehensive screening, diagnosis, and treatment—particularly for infectious diseases such as HIV, hepatitis C virus, and sexually transmitted diseases and for mental illness and substance use disorders—would address broader public health and improve health for those at risk of incarceration. Improving health insurance coverage and medical care is especially important given the evidence on the effectiveness of substance abuse treatment (see Chapter 7). Recent meta-analyses have indicated that drug treatment is associated with reductions in both drug use and recidivism after release (Egli et al., 2009; Mitchell et al., 2012).
In many places, programs already address some of the needs of those diverted or released from incarceration. Prisoner reentry programs have been introduced in all 50 states and a number of local jurisdictions. In various locations, education and transitional employment programs, community health and substance abuse treatment, and community investment
3One notable evaluation is that of the ComALERT reentry program in Brooklyn, which includes a large residential population. It was found that the program delivered reductions in arrests and improvements in employment (Jacobs and Western, 2007).
and neighborhood capacity building have been implemented as part of a comprehensive approach to reducing reliance on incarceration (Pew Center on the States, 2010; Council of State Governments Justice Center, 2013). These developments have been spurred by federal initiatives under the Second Chance Act and the Justice Reinvestment Act. The exploration of social policy supports, in tandem with reduced incarceration, would reflect recognition that the growth in incarceration was in part a response to real social problems in poor communities for which comprehensive approaches are needed. Using policy tools such as the ACA combined with investments in employment, housing, and health care can also provide support to vulnerable populations at the earliest possible time before involvement with the criminal justice system begins.
As noted throughout this report, the committee encountered a variety of gaps in data and empirical research. Most generally, our review of the research revealed great variation in how incarceration is experienced. To a significant degree, this variation reflects broad differences in routines, management, and organizational culture across correctional facilities. These differences in prison conditions are reflected in data on levels of custody. A rough national measure of confinement conditions is also derived from statistics on overcrowding. But beyond these rudimentary indicators, detailed knowledge about the spectrum of conditions of prison life is sparse. Given the extent to which carceral policies in the United States have diverged from those of other affluent democracies in the past four decades, cross-national comparative studies could be expected to shed light on several of the research questions posed below. Across jurisdictions in the United States, great variation also is seen in penal codes and their application in the courts. Variation in incarceration rates has grown across states as the national incarceration rate has increased.
Looking forward, we see several key priorities for research. We begin with one overarching recommendation:
RECOMMENDATION: Given the prominent role played by prisons in U.S. society, the far-reaching impact of incarceration, and the need to develop policies that reduce reliance on imprisonment as a response to crime, public and private research institutions and statistical agencies should support a robust research and statistics program commensurate with the importance of these issues.
More specifically, we recommend support for research aimed at developing a better understanding of (1) the experience of being incarcerated
and its effects, (2) alternative sentencing policies, and (3) the impact of incarceration on communities. (A more detailed discussion is presented in Appendix C.)
Understanding the effects of conditions of confinement on those incarcerated and their chances for successful reentry after prison is important, yet there has been too little systematic research on these questions. Some studies have examined the effects of sentence length on employment and recidivism, and a large literature evaluates prison programs, but researchers know little about interstate variation in sentence lengths and prison conditions (National Research Council, 2012a). Knowledge also is inadequate about the effects on postprison life of overcrowding, victimization in prison, administrative segregation, long-term isolation, mental health treatment, staffing levels, custody levels, and staff training. Most research on social and economic effects treats prison as a black box, with little detailed study of what takes place inside and its potential effects. Because correctional facilities vary so greatly, mapping the differences across facilities would fill a first-order gap in knowledge with immediate policy significance.
A research agenda in this area could assist in the development of standards for conditions of confinement. A national statistical series would allow for cross-jurisdiction comparisons of the dimensions of the prison experience, including such variables as time served and sentence length by crime type, the quality and outcomes of different types of programming, the nature and extent of visitation, the number of prisoners held in different housing configurations, and responses to rule infractions. On the critical issue of health care and treatment of mental illness, a national database would allow for a better understanding of the health consequences of incarceration and the effectiveness of various health and mental health interventions. Another priority for future research is the collection of longitudinal data tracking individuals before and after their contact with the criminal justice system, including prison. Current research often cannot distinguish among the effects of criminal behavior, criminal conviction, and the experience of incarceration as they relate to such outcomes as recidivism, employment, and family life. The ability to make these distinctions is important both to the research community and to policy makers.
As the debate over sentencing policy continues to explore alternatives to incarceration, understanding the effectiveness of these alternatives is a key policy priority. Understanding the available options and assessing
their costs and benefits will require a broad research agenda. At its core, this agenda should include further research on the effects of incarceration on crime rates so that alternative interventions can be compared with the prison sentence. Priority should be given to investigating the magnitude of deterrence as a function of sentence length and to establishing whether other components of the certainty of punishment beyond the certainty of apprehension (such as the probability of imprisonment given conviction) are effective deterrents. Such studies should include estimates of the long-lagged effects on crime, through deterrence or other mechanisms, of specific sentencing policies. Another priority is the development of a comprehensive database that would allow for cross-state comparisons of postconviction sentencing practices over time, as has also been recommended by the National Research Council (2012a).
The research agenda should include an extensive portfolio of evaluations of various sentencing policies that do not involve incarceration so that policy makers can assess available options. This portfolio should include, but not be limited to, evaluations of programs explicitly designed to serve as alternatives to incarceration. These evaluations should be rigorous, be open to replication, and inform the development of standards of best practice to help policy makers invest in these programs instead of prisons. The evaluations should include assessment of the deterrent effects of these sentencing alternatives, as well as estimates of their cost-effectiveness. Similarly, rigorous evaluations should be conducted of in-prison programs designed to facilitate successful reentry and community-based programs committed to reintegration of formerly incarcerated men and women. This research agenda should also yield a better understanding of the impact of various impediments to reintegration, such as legal exclusions from certain employment sectors and restrictions on voting and public housing.
Throughout this report, we have reviewed strong evidence of the extreme concentration of incarceration in poor communities and in the poorest segments of the population. Much of the research on the effects of incarceration has focused on individual-level outcomes for formerly incarcerated individuals and sometimes their families. Yet because of the extreme social concentration of incarceration, the most important effects may be systemic, for groups and communities. If African American male high school dropouts have a high expectation of going to prison at some point in their lives, that expectation may change the behavior of all the men in the group, not just those actually going to prison. If a third of the young men in a poor community are incarcerated, skewing gender balance and disrupting family relations, incarceration may have community-level
effects that shape the social context of community residents, even if their families are not involved in the criminal justice system. Too little is known about these effects.
A rigorous program of research on communities, crime, and crime control (including incarceration) should include comparative qualitative studies of the communities from which the incarcerated come and to which they return; research that takes advantage of “natural experiments” that induce exogenous change in prison admissions or releases; longitudinal or life-course examination of individuals as they are arrested, convicted, and admitted to and released from prison; and the study of neighborhood-level relationships among crime, cumulative neighborhood disadvantage, and criminal justice processing.
Future research also should focus on estimating the aggregate effects of high rates of incarceration on family stability, poverty, economic well-being, and child well-being. As with micro-level research, causal inference is challenging because family stability, poverty, and economic well-being may themselves contribute to local incarceration rates. Also similar to micro-level research, changes in policy or criminal justice practice may induce exogenous variation in incarceration that might enable causal inferences. At the family level, studies should examine how the effect of a parent’s incarceration varies depending on living arrangements prior to incarceration and the quality of relationships with partners and children.
Our examination of the causes and consequences of high rates of incarceration in the United States, informed by a set of normative principles, leads us to conclude that the nation’s incarceration levels are unnecessarily high. We urge a systematic review of the nation’s current sentencing policies with one main goal: a significant reduction in U.S. rates of incarceration. We also urge that the nation take positive steps to treat all prisoners humanely and fairly and to provide prisons with appropriate resources. Finally, to complement a reduction in incarceration and ensure that it does not further disadvantage poor communities, we urge a review of social policies to address the needs for health care, housing, and employment of those who would otherwise be in custody under conditions of high incarceration rates.
The potential impact of the proposed reforms is great. If the share of discretionary funds now allocated to prisons and jails were reduced, savings would accrue to governments and could be used to support other public priorities. Rethinking the proper application of prison sentences could result in a better balance of responsibilities among prosecutors, courts, and legislators consistent with long-established principles of sentencing. A
focus on effective alternatives to incarceration and improved coordination between prison programs and community organizations would strengthen the capacity of the public and private sectors to support reintegration for those convicted of serious crimes. Lowering incarceration rates also would reduce the number of people damaged by imprisonment, limit harmful family separations, keep more workers in the labor market, and mitigate the stigma now associated with time in prison. Improving the quality of life in the nation’s prisons would likely contribute to better physical and mental health, enhance human capital, and improve family relationships.
More fundamentally, reducing the nation’s reliance on imprisonment as a response to crime, together with a parallel reduction in the collateral consequences of incarceration, would recognize appropriate limits on the power of the state, promote social inclusion and racial justice, and enhance the quality of citizenship for those who have been incarcerated. Based on our assessment of the research, we believe a reduction in the nation’s incarceration rates—if implemented with all the necessary policy supports—would achieve these benefits with little if any impact on public safety.
In this report, we have attempted to illuminate what Associate Supreme Court Justice Anthony Kennedy has called the “hidden world of punishment.” In a keynote speech to the American Bar Association in 2003, Justice Kennedy warned that if we look closely at America’s prisons, “we should be startled by what we see.” After reviewing the history of the American prison buildup, the costs of incarceration, and the human toll of imprisonment, Justice Kennedy concluded, “Our resources are misspent, our punishments too severe, our sentences too long.” He ended his speech by reminding his audience that “the more than 2 million inmates in the United States are human beings whose minds and spirits we must try to reach.” With these words, Justice Kennedy anticipated the conclusions of this committee.