A Win for Judicial Sentencing Discretion in Armed Robbery Cases; Additional Reform Still Needed

Earlier this month, in Dean v. United States, the Supreme Court ruled that federal sentencing judges retain some discretion to soften the notoriously draconian sentencing scheme of 18 U.S.C. §924(c). The statute establishes a mandatory prison term when a defendant uses or possesses a firearm in connection with a violent or drug trafficking crime. Unlike most minimums, though, this one must be imposed to run consecutively with any other sentences imposed at the same time. Thus, for instance, a defendant convicted of both a robbery and possession of a firearm during the robbery must get at least five years on top of whatever sentence is ordered for the robbery.

But what if a judge—in light of all of the facts of the case and the circumstances of the defendant—decides that five years is a sufficient punishment for the crime? Could the judge impose a sentence of just one day on the robbery count, so that the total sentence does not exceed what is necessary? In other words, in sentencing for the robbery count, can the judge take into consideration what she will have to impose for the §924(c) count?

Yes, said the Supreme Court in Dean.  


In reaching this result, the Court relied primarily on another statute, 18 U.S.C. §3553(a)—the basic set of sentencing instructions that federal judges are supposed to follow in all cases. In essence, §3553(a) provides a list of broadly worded factors for judges to consider, and then tells them to select a sentence that is “sufficient, but not greater than necessary, to comply with” the various general purposes of punishment. The statute is properly regarded as one that bestows broad discretion on the sentencing judge. Implicitly, the Dean Court observed, this discretion includes the authority to consider the sentences that must be imposed for different counts in the same case.

Given this interpretation of §3553(a), the question became whether anything in the text of §924(c) overrides the more general statute. Without a clear statement of congressional intent to that effect, the Court concluded that sentencing judges are free to take into account the effect of §924(c) when selecting a sentence for a predicate offense.

Dean thus empowers judges to diminish the impact of §924(c) in cases in which rigid application of the minimum would result in a disproportionately severe sentence. But Dean does little to address the worst problem with §924(c)—a problem the Supreme Court itself created through a dubious interpretation of the statute in 1993.

That year, in Deal v. United States, the Court authorized “count-stacking” under §924(c). As a result, sentence lengths can add up quite dramatically for a defendant on a robbery spree. Each robbery supports its own robbery count and its own §925(c) minimum—all of which must run consecutively. The count-stacking is especially draconian because of an unusually harsh recidivism feature of §924(c): second and subsequent convictions trigger not the basic five-year minimum, but a twenty-five-year minimum.

Deal itself provides a good illustration. The defendant was charged in a single case with six bank robberies, with a separate §924(c) count for each one. Robbery number one triggered a five-year minimum, while robberies two through six each carried a minimum of twenty years, for a total of 105 years. (Congress increased the minimum to twenty-five years in 1998.) With no parole available in the federal system, Deal’s punishment was effectively a life sentence—an extreme and highly unusual sentence for a nonhomicide crime. By comparison, the average sentence in robbery cases in urban state courts is less than eight years.

Dean provides no meaningful help for the Deals of the world. Their stacked §924(c) sentences are so high that the sentences for the predicate offenses are of little practical significance.

To be sure, there is nothing unusual about recidivism enhancements. Many statutes authorize or require longer sentences for a second or third offense than for a first. What is unusual about §924(c), as interpreted by Deal, is that the recidivism enhancement kicks in with multiple convictions in a single case. Normally, recidivism enhancements require a fresh offense committed after the first conviction. This is in keeping with a central purpose of recidivism enhancements, which is to impose extra imprisonment based on an offender’s persistence in crime even after receiving punishment once before—a persistence that represents a heightened defiance of the legal system and signals a worrisome unresponsiveness to the law’s deterrent threats. By contrast, when multiple, similar crimes are prosecuted in a single case, the traditional sentencing approach is to impose reduced incremental punishment for each subsequent offense—an approach that was deliberately built into the federal sentencing guidelines.

Properly read, the text of §924(c) does not mandate count-stacking. As the Deal dissenters argued, there are perfectly reasonable interpretations of the statute that would bring it in line with the guidelines’ approach to recidivism.  If the Court is serious about its embrace of judicial sentencing discretion in Dean, then it would do well to reconsider the earlier holding in Deal.

Better still, Congress could pass the Sentencing Reform and Corrections Act, which would limit count-stacking and reduce the minimum for second and subsequent convictions from twenty-five to fifteen years. As introduced in the last Congress, the SCRA had broad bipartisan support and was approved by the Senate Judiciary Committee, but was never brought to the floor—an apparent victim of election-year politics. The election now behind us, Congress should take up mandatory minimum reform again.