Decarcerating America Read online

  In United States v. Kupa, the Honorable John Gleeson of the U.S. District Court for the Eastern District of New York, described this practice and its effect on a particular defendant. Kupa, a Staten Island–born child of Albanian immigrants, had two prior convictions for selling marijuana before the drug arrest that brought him into Judge Gleeson’s court. Federal prosecutors offered Kupa a plea bargain that would have resulted in a sentence as low as seven years and ten months in prison, but Kupa refused to admit guilt. The government then filed a prior felony information, advising the court of the two marijuana convictions. Those two prior drug convictions triggered a provision under mandatory sentencing rules that would have required a sentence of life in prison without the possibility of parole—but only if Kupa went to trial. Five weeks later, Kupa finally pled guilty to a deal providing for a sentence of 140–175 months in prison—thirteen and a half years or more—telling Gleeson that he was doing so in order to avoid spending life in prison. “I want to plead guilty, Your Honor, before things get worse,” he said. Judge Gleeson sentenced Kupa to 132 months in prison, but wrote an opinion castigating the Department of Justice for using its prior felony information power to coerce plea bargains. Gleeson concluded, “If DOJ cannot exercise its power to invoke recidivist enhancements in drug trafficking cases less destructively and less brutally, it doesn’t deserve to have the power at all.”

  While plea bargains are essentially negotiated contracts between the parties, a negotiation from which judges are appropriately excluded, the acceptance of that plea is a judicial responsibility. In the light of Judge Gleeson’s powerful reasoning, judges should probe more deeply into the bargaining process to ensure that it was not improperly coercive. These inquiries can play an important role in drawing attention to unfair practices, though the circumstances of Kupa show the limits of a judge’s ability to alter the outcome: Judge Gleeson could have ruled that Kupa’s guilty plea was coerced and thrown it out, but that would have left the defendant facing trial, and potentially a life sentence.

  The Sentencing Process

  The judge has the responsibility of determining and pronouncing a sentence. The judge must resolve the conflicting views of acts involved and justify the decision to the defendant (and frequently his or her family), to the victims of the crime, and to the public. The methodology and rules of this wrenching and difficult process have drastically changed over time.

  In the 1970s, when one of us, Justice Robert Sweet, became a judge, the only legislative constraint was the range of penalties prescribed by Congress. There was no reporting on sentencing or information available about norms, except on a case-by-case basis. To establish consistency in sentencing even for an individual judge was a daunting task, and a comparative court-wide analysis was not possible. The judge had virtually untrammeled discretion. Appellate review was infrequent and limited to whether the sentence imposed was permissible under the law passed by Congress that criminalized the offense. The Supreme Court’s jurisprudence on the issue recognized that the trial-level judge who supervised the progress of the criminal case generally had a superior understanding of the crime, the offender, and the interests involved, and so the sentence imposed was “met with virtually unconditional deference on appeal,” in the words of Justice Harry Blackmun.

  In practice, however, this meant that sentences could vary based on the priorities and prejudices of the judge overseeing a given case. This situation was criticized by those who sought stricter enforcement for white collar crime, for instance, and those who felt that minorities were disproportionately punished. After several years of urging, with the support of both the Reagan administration and liberals including Senator Ted Kennedy, Congress passed the Sentencing Reform Act of 1984, which created the United States Sentencing Commission and empowered it to create guidelines that would “recommend to the sentencing judge an appropriate kind and range of sentence for a given category of offense committed by a given category of offender,” according to the Senate Judiciary Committee. The new guidelines system would require the judge to consider the history and characteristics of the offender, the nature and circumstances of the offense, and the purposes of sentencing before issuing a sentence within a prescribed range, which would be generated by mechanistic calculations. Judges across the country were required to go through the same steps in determining a sentence, and those steps were recorded for appellate review. (Congress did, however, provide that judges could impose sentences outside the guidelines in extraordinary cases.) Critics concluded that the guidelines impermissibly restricted the judge’s authority—in 1988 United States v. Alafriz, the author declared them unconstitutional because they deprived defendants of their right to be sentenced as individuals and because the regime gave appointed commissioners authority over sentencing, a function reserved for the judicial branch. However, the guidelines were upheld as constitutional by the Supreme Court in Mistretta v. United States (though Justice Antonin Scalia agreed with the separation of powers analysis in his dissent).

  As time went on, certain glaring inequities were recognized in the law of sentencing, particularly the disparity between the treatment of crack and powder cocaine. The two drugs share the same active ingredient and psychotropic effects, but the Anti–Drug Abuse Act of 1986, enacted during a period of public panic about drug abuse in general and crack cocaine in particular, required courts to treat each gram of crack cocaine—the inexpensive form primarily used in African American communities—as equivalent to 100 grams of powder cocaine, the form primarily used by white Americans, for sentencing purposes where severity of sentences was determined by the weight of the drugs involved. The effects of this disparity were particularly destructive when combined with the Anti-Drug Abuse Act’s mandatory minimum provisions, which required a five-year sentence for any defendant distributing 5 grams of crack cocaine (as opposed to 500 grams of powder cocaine), and a ten-year sentence for defendants convicted of distributing 50 grams of crack cocaine (as opposed to 5 kilograms of powder cocaine). As the Supreme Court noted in Kimbrough v. United States (discussed in further detail below), since cocaine tends to be imported into the country in powder form before being chemically converted into crack for distribution to individual users, the disparity resulted at times in street-level sellers of crack cocaine being punished more severely than the international traffickers who were responsible for the wholesale importation and distribution of the same drugs. Because the Anti–Drug Abuse Act was passed through Congress on an emergency basis, bypassing much of the normal legislative process, there is “no authoritative legislative history,” in the words of the Sentencing Commission, that would explain what empirical rationale, if any, Congress used in adopting the 100-to-1 disparity.

  Although the Sentencing Commission sent reports to Congress criticizing the unfairness of the cocaine sentencing regime, Congress took no action in response. The first hint of change came thanks to the action of a federal judge. In April 2005 (shortly after the Supreme Court’s seminal Booker decision, discussed further below), the Honorable Raymond A. Jackson of the Eastern District of Virginia refused to sentence a crack dealer named Derrick Kimbrough to the nineteen years required by the sentencing guidelines, reasoning that since Kimbrough would have received only a seven-and-a-half-year sentence for distributing the equivalent amount of powder cocaine, the case exemplified “the disproportionate and unjust effect that the crack cocaine guidelines have in sentencing.” The Kimbrough case made its way to the Supreme Court, where a 6–3 opinion written by Justice Ruth Bader Ginsburg upheld Judge Jackson’s departure, concluding that judges were empowered to deviate from the guidelines based solely on the recognition that the crack/powder disparity was unjust. Some amount of legislative relief finally arrived with the Fair Sentencing Act of 2010, which reduced the crack/powder cocaine disparity to 18-to-1, major progress from the regime that had existed before, but still far from the equivalent treatment that is warranted.

  Judge Jackson was not the only member of the
federal bench who strained against the rigidity of the sentencing guidelines and the inequities codified within them. Although the Sentencing Reform Act had made the guidelines effectively mandatory, judges were still authorized to issue non-guideline sentences for two reasons: either a motion from the government regarding the “substantial assistance” provided by a defendant who gave evidence to investigators, or a finding that there existed aggravating or mitigating circumstances that were not adequately taken into consideration by the Sentencing Commission. Although this latter category was intended to be used only in exceptional cases, federal judges began using it more frequently to correct guidelines outcomes that they perceived as unjust. Between 1991 and 2001, downward departures for reasons other than substantial assistance increased from 5.8 percent of sentences to 18.1 percent. The judiciary took a dimmer and dimmer view of the guidelines over this period; in a 1996 survey by the Federal Judicial Center, 73 percent of district judges and 69 percent of circuit judges felt that the mandatory guideline regime was unnecessary.

  Congress reacted to these acts of judicial independence by enacting the Feeney Amendment to the PROTECT Act of 2003, which was ostensibly geared toward defending children from sex offenders, requiring the Sentencing Commission to review every downward departure (thus reducing the number of areas where judges could argue that the Sentencing Commission had not considered key mitigating factors) and changing the standard for appellate review of fact-based sentencing decisions from one that deferred to the trial judge to one that allowed appellate judges to re-decide the issues from scratch. The amendment drew widespread condemnation from legal scholars and the federal bench. Chief Justice William Rehnquist referred to it as “an unwarranted and ill-considered effort to intimidate individual judges in the performance of their duties.”

  The landscape of sentencing law changed fundamentally with the Supreme Court’s 2005 decision in United States v. Booker. Booker essentially involved two half-opinions, each supported by a 5–4 majority of justices, that altered the Court’s previous divisions. In the first half of Booker, written by Justice John Paul Stevens with the support of Justices Scalia, David Souter, Clarence Thomas, and Ginsburg, the Court concluded that provisions of the guidelines that enhanced sentences based on findings of fact by a judge violated the Sixth Amendment right to trial by jury. In the second half, written by Justice Stephen Breyer with the support of Chief Justice Rehnquist and Justices Sandra Day O’Connor, Anthony Kennedy, and Ginsburg, the Court determined that the Sixth Amendment problem would be solved by striking down the statutory language that had made the guidelines mandatory.3

  The practical consequence of Booker rendering the guidelines advisory is that judges have largely returned to their previous ability to tailor sentences according to the circumstances of individual defendants. The guidelines remain important, however: each judge still conducts the mechanical guidelines analysis for every sentence, and the result plays an important anchoring function. Because the guidelines tell the judge and the parties what sanction a normal defendant would receive in the given circumstances, judges remain likely to impose a sentence within the guidelines range unless there is a reason not to do so.

  During all of this period, and up to the present day, the factors a judge must consider in imposing a sentence have been codified in a statute—Title 18, Section 3553(a) of the United States Code. The section requires judges to consider the nature of the offense, the characteristics of the defendant, the four fundamental purposes of criminal law (retribution, deterrence, public protection, and rehabilitation), the kinds of sentences available, compliance with the guidelines and Sentencing Commission policy statements, the status of co-defendants, and restitution to victims.

  The § 3553(a) factors are certainly broad enough to permit judges to consider the many problems of mass incarceration. The statute provides for the consideration of the nature and circumstances of both the offense and the defendant, as well as the need for the sentence imposed “to reflect the seriousness of the offense” and “to afford adequate deterrence to criminal conduct.” While these factors alone are sufficient to permit consideration of many of the negative impacts of drug criminalization and mass incarceration, Congress might well add to § 3553(a) a provision requiring the consideration of the comparative costs to taxpayers of incarceration or community supervision. Currently, the factors do not account for the financial burden that incarceration imposes on taxpayers, which the Administrative Office for the U.S. Courts estimates at $30,621 per inmate per year. Such a revision might well appeal to those concerned about both mass incarceration and those who are sensitive to the economics of the government.

  One arbitrary imposition on the judge’s sentencing discretion remains: the mandatory minimum sentences resulting from Congressional legislation. The damaging effect of mandatory minimums could be seen in the Kimbrough case where, despite a Supreme Court decision in his favor, Kimbrough was still sentenced to the statutory minimum of fifteen years for crack cocaine, twice what he would have received had he pled guilty to distributing the equivalent amount of powder cocaine. These mandatory minimums can be set aside by the prosecution, who can grant a letter outlining the cooperation of the defendant in assisting the prosecution. The provisions are primarily applicable to drug crimes, though other offenses are covered.

  Mandatory minimums and three-strike provisions are responsible to a substantial degree for the mass incarceration problem. Some amelioration of these provisions is presently under consideration in Congress at this writing, with no assured resolution. According to a report by the Sentencing Commission to Congress, as of 2010, 58.1 percent of federal prisoners were convicted of offenses carrying a mandatory minimum penalty, and 39.4 percent failed to qualify for any of the exceptions and were sentenced subject to the minimum. The data also show a significant racial disparity in the way mandatory minimums are applied, with black and Hispanic defendants subject to minimum penalties at a higher rate than non-Hispanic white defendants.

  Although progress has been made in curbing the most grievous aspects of the current sentencing regime, reforms should be adopted by courts and judges that would lead to the development of a common law of sentencing. By making the outcome and reasoning of every sentencing decision available to other sentencing judges, both individually and as part of aggregated sets of data, the decisions of federal judges across the country could come to play the anchoring role currently occupied by the guidelines. Such a system could well replace the guidelines’ current advisory form with judicial decisions, reflecting the reasoned judgment of a large sample of federal cases rather than the outcome of a set of mechanistic calculations.

  Former district court judge Nancy Gertner has addressed this subject in an article in the Harvard Law and Policy Review entitled “Supporting Advisory Guidelines.” Judge Gertner noted that while the guidelines were no longer mandatory, there was little background information available to judges seeking an alternative framework for sentencing. She suggested that every sentencing decision, whether in the form of a full opinion, a hearing transcript, or a standardized form, should be made searchable and available to other sentencing judges and to the public. Meanwhile, the Sentencing Commission should educate jurists not only on how to comply with the guidelines but also on alternative sentencing structures backed by social science research on best practices. Twenty years ago, during the mandatory guidelines era, one of us, Judge Sweet, recommended in the Fordham Law Review that all federal judges issue written sentencing opinions in each criminal case, which could then be digitized by the sentencing commission to produce a database showing statistical norms for each combination of offense and offender.

  Even without a national system, there remain small-scale changes that can help lead to a common law of sentencing. Openness and clarity are served by judges filing sentencing opinions outlining the considerations underlying the sentence in each case.4 If sentencing opinions were produced and made broadly available, norms could be devised within
districts and circuits, and a significant departure from those norms could constitute a challenge on appeal to the judge’s discretion. Currently, there is virtually no appellate review of sentencing except for an illegal sentence or improper guidelines calculation.

  Another option would be to make public the “statement of reasons” that is submitted to the Sentencing Commission in every case. In this brief, four-page form, the judge certifies whether there is a mandatory minimum sentence (or whether the defendant is somehow exempt), the results of the guidelines calculations, whether the judge is sentencing the defendant within the guidelines, and, if the judge chooses to sentence the defendant outside the guidelines range, what basis he or she gives for departing. Although the document is submitted to the Sentencing Commission and the aggregate data from these statements are the basis for much of the annual statistical report that the commission generates, the documents themselves are unfortunately not part of the public record.

  The rule of secrecy that keeps these important documents sealed is a judge-made one, enacted by the Judicial Conference of the United States in 2001, meaning that judicial action could also make the statement of reasons public.5 In fact, the District of Massachusetts has voted to make available all statements of reasons for sentences imposed in that state (outside of extraordinary cases), which has allowed for public scrutiny and statistical analysis of judicial reasoning. If this policy were enacted nationwide, norms presumably could be created from the aggregated data, effectively creating guidelines devised by those primarily responsible, the judges. Those norms and departures from them could form the basis for appellate review.