Vol. 99, No. 2

Sentencing Immigrants

Eric S. Fish*
Asylum, Criminal Law, Immigration Law, Refugees
Articles

Introduction

The most commonly charged federal felony is the crime of reentering the United States after deportation.¹ On average twenty thousand people have been prosecuted for it per year since 2010, making unlawful reentry cases about one-third of the federal criminal caseload.² And that number is about to dramatically increase. President Trump’s Department of Justice has announced a new policy of charging felony unlawful reentry in every available case, and it plans to reassign federal prosecutors working on drug and terrorism cases in order to prosecute more immigrants.³ Roughly 99% of the defendants in unlawful reentry cases are Latin American.⁴ The charge is, by definition, a nonviolent and victimless crime. It involves only entering or being found in the United States after a deportation. Yet defendants in these cases are systematically punished more harshly than comparable U.S. citizen defendants are in other cases. This happens in two ways.

First, the United States Sentencing Guidelines count reentry defendants’ criminal history against them twice. The Sentencing Guidelines are a complex set of rules that produce recommended sentences in federal criminal cases. While federal judges are not bound by the Guidelines, they follow the Guidelines about 90% of the time in unlawful reentry sentencings.⁵ In a standard criminal case, the Guidelines create a recommended sentence range by quantifying two variables: the defendant’s past convictions and facts about the current case. Past convictions create a “Criminal History Category” between one and six, while facts about the current crime create an “Offense Level” between one and forty-three.⁶ But in unlawful reentry cases, past convictions increase both the Criminal History Category and the Offense Level. Reentry defendants thus receive much larger sentence enhancements for their past crimes.

This double counting is worsened by another problem: the arbitrary way prior convictions are used to generate enhancements. The unlawful reentry guidelines make large swings in sentence length turn on obscure details of a person’s criminal history. This renders sentencing in these cases uniquely random and uncertain. Multiple years in prison turn on minutiae like the precise timing of a deportation order or probation violation, or whether a state uses a determinate or indeterminate sentencing system. People with similar criminal and immigration histories receive vastly different sentences based on such technical differences. And this problem is further compounded by the difficulty of finding and interpreting state conviction records. Figuring out the recommended sentence is like solving a high-stakes logic game. Defense lawyers are commonly unable to discern what length of sentence a defendant is looking at. And because the system pressures reentry defendants to plead guilty at the first opportunity, defendants often find out at sentencing that their punishment will be more severe than anticipated.

Second, undocumented immigrants serve a substantially larger portion of their sentences than do U.S. citizen defendants. The First Step Act, signed into law by President Trump in 2018, created a new system of “earned time” credits for federal prisoners.⁷ Under that law, a federal defendant who is willing to participate in anti-recidivism programs can get up to fifteen days off their prison sentence for every thirty days served.⁸ That means their time in prison can be reduced by up to one-third. In addition, defendants receive these credits even if they are not assigned to a program or no program is available. The only requirement is that they not refuse to participate. But deportable immigrants are prohibited from both participating in First Step Act programs and receiving custody reductions.⁹ This means that, by law, an undocumented prisoner will serve a significantly higher percentage of their sentence due to their immigration status.

To illustrate how this sentencing discrimination works in practice, consider two federal defendants: X and Y. Defendant X is charged with escaping from a federal prison, while defendant Y is charged with reentering the United States after deportation.¹⁰ Both X and Y have identical criminal histories: two prior felony convictions within the past fifteen years, each resulting in a two-year sentence. Under the Sentencing Guidelines, both X and Y will have a Criminal History Category of III due to the two prior felonies.¹¹ Assuming that there are no aggravating or mitigating facts (like using physical violence or returning to custody voluntarily), X will have an Offense Level of thirteen for escaping from prison.¹² Y will start with a lower Offense Level of eight for reentering the United States. But Y’s past convictions will also trigger enhancements of the Offense Level. If Y’s first deportation happened either before or after their two felony convictions, their total Offense Level will be sixteen.¹³ If their first deportation happened in between the two felony convictions, the Offense Level will be twenty-four.¹⁴ The Guidelines will recommend a sentence of eighteen to twenty-four months for X’s prison escape. And for Y’s reentry, they will recommend a sentence of either twenty-seven to thirty-three months or sixty-three to seventy-eight months (depending entirely on the timing of the first deportation). Unlawful reentry is the only federal crime for which criminal history is double counted like this. And once in prison, X will qualify for First Step Act programs that will let X spend up to one-third of their sentence out of custody or at a halfway house. Y, as a deported immigrant, cannot benefit from those programs.

As this example shows, these rules create two separate and unequal federal sentencing systems. Undocumented defendants, basically all of them Latin American, are treated more severely for their past crimes and are denied the lenient sentence reductions that U.S. citizens enjoy. And the enactment history of these policies reflects nativist animosity toward Latin American immigrants. The large recidivist enhancement for unlawful reentry was first created in 1991 with no study and little debate.¹⁵ It was enacted in response to Congress increasing the maximum sentence from two to fifteen years. That statutory change, in turn, was driven by racial panic in Florida over Haitian and Cuban refugees causing a supposed crime wave.¹⁶ The initial enhancement focused only on a small number of serious crimes, including murder and drug trafficking. But it evolved over time into a general enhancement for all past crimes. The First Step Act, in turn, was the first Trump Administration’s signature criminal justice reform initiative. Since the Trump Administration has been the most anti-immigration presidency in

recent memory, it is not surprising that the law excludes deportable immigrants from its earned time credit system.¹⁷

These sentencing rules explicitly discriminate against immigrants, and their history reflects racist and nativist antipathies. It might thus be logical to look to the Equal Protection Clause for a remedy, or to the federal statute mandating that the Sentencing Guidelines be “entirely neutral” as to race and national origin.¹⁸ But defendants trying to make such arguments have run into a problem—the Supreme Court’s antidiscrimination doctrines erect numerous obstacles to such equality claims.¹⁹ Federal laws that discriminate by immigration status are only subjected to “rational basis” review.²⁰ While these sentencing rules overwhelmingly burden Latin American defendants, the Supreme Court has disallowed racial discrimination claims that rely on disparate impact.²¹ And the Court adopts a strong presumption that government action is not motivated by racial animus.²²

But sentencing is not like judicial review. In the discretionary sentencing context, the judge is the legally designated decisionmaker. They are not reviewing the work of an agency or legislature. Federal judges are empowered to disregard the Guidelines and sentence a defendant anywhere within the statutory range.²³ The deference concerns that have hollowed out Equal Protection doctrine thus do not apply to discretionary sentencing decisions. In other contexts, scholars have identified circumstances in which executive branch agencies engage in “administrative constitutionalism”—incorporating constitutional norms into policy more robustly than formal legal doctrine requires.²⁴ Judges should also embrace a version of administrative constitutionalism at sentencing. Here I call this “sentencing constitutionalism.” Employing it, judges should reject sentencing rules that create racial and immigration-status-based hierarchies. And they should do so even if those rules would be upheld in the formal constitutional review context. Justice Kennedy gestured at something like this approach in Beckles v. United States.²⁵ In that case, the Supreme Court held that constitutional vagueness challenges do not apply to the Federal Sentencing Guidelines because they are advisory.²⁶ But Justice Kennedy wrote a concurrence arguing that quasi-constitutional vagueness arguments still have a place in the discretionary sentencing context.²⁷ Sentencing constitutionalism is also analogous to the idea of “imperfect” defenses in criminal cases. There are contexts in which a defendant cannot legally argue a defense like duress or self-defense, but the moral justification underlying that defense still clearly applies.²⁸ Judges will often reduce defendants’ sentences due to such imperfect defenses.²⁹ Sentencing constitutionalism operates the same way, but with constitutional equality claims in place of defenses.

Federal crack cocaine sentencing provides a powerful example of sentencing constitutionalism in practice, as well as a close analogue to the sentencing discrimination faced by immigrants. For sentencing purposes, the Guidelines initially treated one gram of crack cocaine the same as one hundred grams of powder cocaine.³⁰ This caused crack cocaine traffickers to be punished much more harshly than powder cocaine traffickers for the same volume of basically equivalent drugs. And this had a clear disparate impact by race because crack cocaine defendants were almost entirely African American.³¹ Much like the policy of double-counting immigrants’ criminal history, these crack sentencing rules were designed without formal deliberation or policy analysis,³² and were made in response to a congressional statute enacted during a racialized moral panic over crime.³³ The crack sentencing guidelines were ultimately upheld in direct constitutional challenges.³⁴ But many federal judges have exercised their authority to disregard the crack guidelines and adopt less discriminatory sentencing practices.³⁵ In doing so, these judges have adhered to a higher standard of antidiscrimination than formal doctrine requires. Judges should act similarly in the immigrant sentencing context. They should refuse to impose the discriminatory rules built into the Guidelines, and they should counteract the discrimination built into the First Step Act’s sentence reductions.³⁶

This Article has two goals: to describe how federal sentencing law discriminates against undocumented immigrants, and to propose a framework for judges to counteract that discrimination. It is organized into four Parts. Part I traces the history of federal immigrant sentencing over the last several decades, exploring how and why these unequal rules were adopted. Part II shows that these sentencing rules discriminate explicitly by immigration status, and in effect by race. Part III explains how the Guidelines’ double-enhancements for criminal history create a sentencing system where arbitrary details of a defendant’s criminal record dictate large (often multi-year) increases in their sentence. It also illustrates the problem with a real-life case. Part IV introduces the concept of sentencing constitutionalism. It argues that judges should exercise their inherent power over discretionary sentencing to counteract punishment rules that discriminate by race and immigration status. It also shows that federal judges have been doing precisely this in the context of crack cocaine sentences—varying downward to mitigate racial discrimination that has gone unredressed in formal legal challenges.

I. The Evolution of Undocumented Immigrant Sentencing

This Part explores how federal law crafted a separate sentencing system for undocumented immigrant defendants. It focuses on three key changes. First, in the late 1980s and early 1990s, three members of Congress from Florida—Lawton Chiles, Bob Graham, and Bill McCollum—successfully pushed to increase the maximum penalty for unlawful reentry from two to twenty years.³⁷ These legislators justified the increase by arguing that immigrants were causing a crime wave. They did so in reaction to a nativist panic in Florida over Cuban and Haitian immigration. Second, in response to this increase in the maximum penalty, the United States Sentencing Commission amended the unlawful reentry guidelines in 1991 to provide double sentence enhancements for prior convictions.³⁸ Those guidelines have changed significantly in the intervening years, most notably in 2001 and 2016. But they still punish unlawful reentry defendants more severely for past convictions than other federal defendants. Third, in 2018 Congress enacted the First Step Act with support from the Trump Administration.³⁹ Among other things, that law created a system of earned time credits that let federal prisoners reduce their custody time by up to a third. But, due to the Trump Administration’s antipathy toward Latin American immigrants, prisoners with deportation orders were excluded from earning such credits.

A. Florida’s Refugee Crisis and Reentry Sentences

Congress originally criminalized unlawful reentry in 1929, as part of the “Undesirable Aliens Act.”⁴⁰ The legislators who enacted it believed in scientific racism, and their goal was to limit immigration from Latin America in order to preserve the purity of the white race.⁴¹ The law established, among other things, that any noncitizen convicted of reentering the United States after deportation would face up to two years in prison.⁴² That unlawful reentry provision is now codified at 8 U.S.C. § 1326.

The sentences for unlawful reentry remained unchanged from 1929 until the late 1980s, when three members of Congress from Florida—Senator Lawton Chiles, Senator Bob Graham, and Representative Bill McCollum—sponsored a series of amendments that made the law harsher. To understand these legislators’ focus on punishing unlawful reentry more severely, one must appreciate Florida’s immigration politics in the 1980s and early 1990s.⁴³ During that period, large influxes of refugees from Cuba and Haiti sparked a racial panic.⁴⁴ In 1980, the Mariel boatlift brought nearly 125,000 Cubans and Haitians to Florida.⁴⁵ Thousands more Haitians arrived in Florida as refugees from the Duvalier regime, and later from the military government that overthrew President Aristide in a coup.⁴⁶ The presence of these immigrants caused a racist and nativist backlash among Floridians. The refugees were seen as carrying infectious diseases and creating a public health crisis.⁴⁷ They were also viewed as committing widespread crime.⁴⁸ Especially salient in Floridians’ minds was the specter of Latin American and Caribbean drug trafficking networks.⁴⁹ Florida’s politicians (including Chiles, Graham, and McCollum) stoked these nativist fears and responded to them with anti-immigrant measures.⁵⁰ The federal government incarcerated thousands of Cuban and Haitian immigrants in detention centers and jails.⁵¹ And Florida’s representatives in Congress sought, among other things, to increase the penalties for unlawful reentry.

The maximum sentence for unlawful reentry stayed at two years until 1988, when Senator Lawton Chiles proposed an amendment to increase it.⁵² Chiles’s amendment, which was attached to the Anti-Drug Abuse Act of 1988, increased the maximum penalty to five years for any defendant who was deported after a felony conviction.⁵³ It also increased the maximum to fifteen years for any defendant deported after an “aggravated felony” conviction (Senator Chiles’s original proposal was for a fifteen-year mandatory minimum sentence, but this was changed to an increased maximum in the final statute).⁵⁴ The term “aggravated felony” was introduced in the same legislation, as a way to designate crimes that will almost certainly result in deportation.⁵⁵ The term initially only applied to a small list of crimes—murder, drug trafficking, and firearms trafficking—but it was later expanded.⁵⁶ Along with the increase in reentry penalties, Chiles introduced a number of other amendments to make immigration enforcement more punitive. These included a provision increasing the penalties for immigrant smuggling, a provision making it a crime not to appear at deportation proceedings, and a provision denying bond in the immigration system for people with aggravated felony convictions.⁵⁷

In arguing for his amendments, Senator Chiles explicitly connected them to Florida’s panic over immigrant crime. In the committee hearing that he chaired presenting the proposals, he asserted in his opening statement that “an expansive drug syndicate established and managed by illegal aliens” was “so widespread and lucrative that they are attracting other aliens just to come into the illegal enterprise.”⁵⁸ He also claimed that “one of the largest and most widespread crack operations is run by illegal Haitians.”⁵⁹ Chiles elicited testimony from several Florida law enforcement witnesses, who testified that Haitian immigrants play a major role in the Florida drug trade, and that it is difficult to prosecute such immigrants due to them skipping bond or leaving the United States.⁶⁰ For example, one witness, an undocumented Haitian immigrant who was a police informant in Orlando, testified: “Many Haitians are brought into the United States illegally for the

sole purpose of dealing drugs and to recruit other Haitians for the drug business. These dealers are mixed with political refugees to hide their identity.”⁶¹

In Chiles’s speech introducing his proposed amendments on the Senate floor, he stated that “the number of illegal aliens who deal in crack in the Orlando area has increased 300 percent in the last year,” and that “these numbers are being duplicated throughout Florida.”⁶² He further warned that Florida was facing “expansive drug syndicates established and managed by illegal aliens.”⁶³ He claimed that “these syndicates operate many of the drug networks—crack, cocaine, and heroin—in Florida and throughout the United States.”⁶⁴ And also that “illegal alien felons . . . have no fear of or respect for our legal system.”⁶⁵ Chiles’s claims about the scope of immigrant drug activity were exaggerated and implausible.⁶⁶ Nonetheless, they were the major motivation behind his successful effort to increase reentry penalties. Chiles said the goals of the law were “creating a greater deterrent to alien drug traffickers who are considering illegal entry into the United States” and to “give law enforcement authorities a broader arena for prosecuting the drug offender.”⁶⁷

Several other legislative changes sponsored by Florida congressmen in the early 1990s also made § 1326 more punitive. In 1994, Representative McCollum successfully advocated an additional hike in the maximum sentences for unlawful reentry.⁶⁸ His amendment increased the maximum penalty to ten years for a defendant deported after a felony conviction, and to twenty years for a defendant deported after an aggravated felony conviction.⁶⁹ In addition, Representative McCollum and Senator Graham successfully sponsored several amendments that expanded the definition of “aggravated felony” to include a significantly longer list of crimes, many of them less serious.⁷⁰ Senator Graham also co-sponsored a provision that would have required the U.S. Sentencing Commission to increase unlawful reentry sentences for people deported after aggravated felony convictions.⁷¹ The proposal called on the Commission to “assign[] an offense level . . . that constitutes a meaningful deterrence to the commission of such offense.”⁷² Graham added this provision to Senate Bill 1711, titled “Implementing the President’s 1989 National Drug Control Strategy,” which passed in the Senate but did not ultimately become law.⁷³ A prosecutor testifying before a congressional subcommittee later described Graham’s proposal as “a ‘sense of the Congress’ recommendation to the Federal Sentencing Commission that would upgrade the penalty for an aggravated criminal felon who after conviction returns to the United States.”⁷⁴ Like Senator Chiles, Senator Graham and Representative McCollum were both focused on indulging Florida’s anti-refugee backlash.⁷⁵ They made unlawful reentry punishments harsher in response to the perceived immigrant crime wave in their home state. As the next Section will show, these three politicians’ efforts had enormous consequences for unlawful reentry sentencing.

B. Writing and Rewriting the Reentry Guidelines

In 1984, Congress enacted the Sentencing Reform Act and created the United States Sentencing Commission.⁷⁶ Three years later, the Commission published the first version of the United States Sentencing Guidelines.⁷⁷ The main purpose of these new Guidelines was to reduce sentencing disparities between judges.⁷⁸ They created a complex formula that takes facts about the defendant’s crime and past convictions as inputs, quantifies them into points,

and produces a sentencing range that judges were (until 2005) required to follow.⁷⁹

The Guidelines’ sentencing ranges are arrayed on a two-dimensional six by forty-three cell grid called the “Sentencing Table.”⁸⁰ The Sentencing Table’s Y-axis ranges from one to forty-three, and a defendant’s level is determined by facts about the current crime. Each crime is given a “base offense level” of a certain number of points, for example eight points for unlawful entry and thirteen points for escaping from prison.⁸¹ Additional points are then added (or subtracted) for “specific offense characteristics,” which are facts particular to the defendant’s crime. If someone escaping from prison uses or threatens to use force, for example, they get five additional points.⁸² The Sentencing Table’s X-axis provides the defendant’s “criminal history category,” which ranges from category I to category VI. Where a defendant lands on this dimension is determined by the number, severity, and recency of their past convictions.⁸³ The Guidelines thus take facts about the defendant’s current case and facts about their criminal history, plug those facts into two distinct formulas, and map the resulting numbers onto a two-dimensional grid to produce a sentence range.⁸⁴

When the Guidelines were first promulgated in 1987, a defendant convicted of unlawful reentry received a base offense level of six.⁸⁵ If the defendant had previously “unlawfully entered or remained in the United States,” they received two additional points as a specific offense characteristic.⁸⁶ The maximum offense level for unlawful reentry was thus eight. With eight offense level points, the sentence range varied from zero to six months (for those in criminal history category I) to eighteen to twenty-four months (for those in criminal history category VI).⁸⁷ According to the Sentencing Commission, this initial version of the reentry guideline was designed to reflect then-existing sentencing norms for immigration cases.⁸⁸ A year later, in 1988, the Commission wrote a new version of this guideline that simply had a base offense level of eight, with no specific offense characteristics.⁸⁹

In 1989, the reentry guideline was amended to double-count defendants’ criminal history. The Commission added a four-point enhancement applying to any defendant who “previously was deported after sustaining a conviction for a felony.”⁹⁰ This was the first time that the Commission created a specific offense enhancement for simply having a past conviction.⁹¹ This enhancement applied regardless of the prior conviction’s age—a felony from fifty years ago would still trigger the four-point enhancement.⁹² By contrast, the Guidelines’ standard criminal history score formula stops counting convictions after ten or fifteen years (depending on the prior sentence length).⁹³

The Commission enacted this enhancement without conducting any studies or hearings, and without providing any apparent explanation for the change.⁹⁴ The Commission’s commentary only notes that “this specific offense characteristic is in addition to, and not in lieu of, criminal history points added for the prior sentence.”⁹⁵ The enhancement closely tracks the language in Senator Chiles’s 1988 amendment to § 1326, which increased the maximum penalty to five years for any defendant “whose deportation was subsequent to a conviction for commission of a felony.”⁹⁶ The Guidelines’ commentary also mentions aggravated felonies, and suggests that “an upward departure may be warranted” in cases involving “a defendant previously deported after sustaining a conviction for an aggravated felony.”⁹⁷ The four-point enhancement in the 1989 Guidelines Manual is thus properly seen as a response to Senator Chiles’s amendment increasing the maximum sentences for defendants with prior felonies and aggravated felonies.⁹⁸

In 1991, the Commission added a much larger sixteen-point enhancement for any defendant who “previously was deported after a conviction for an aggravated felony.”⁹⁹ This was a very punitive change. For a reentry defendant in criminal history category I, it brought the sentencing range from zero to six months to fifty-one to sixty-three months.¹⁰⁰ For a defendant in category VI, it created a range of one hundred to one hundred twenty-five months. With this new enhancement, a single prior felony conviction for drug trafficking (or any other aggravated felony) would add multiple years to a defendant’s reentry sentence. This amendment was adopted a little over a year after Senator Graham’s proposal to mandate that the Commission increase penalties for reentry after an aggravated felony.¹⁰¹ While the Commission’s records do not reference Graham’s amendment, it was likely a motivating factor.¹⁰²

The Commission conducted hearings in 1991 on various proposed Guidelines amendments, including the enhancement for a prior aggravated felony.¹⁰³ At those hearings, a federal prosecutor named Joe Brown testified on behalf of the Department of Justice.¹⁰⁴ Brown recommended that the Commission adopt a twenty-point enhancement for defendants deported after committing an aggravated felony.¹⁰⁵ He connected this proposed enhancement to Senator Chiles’s amendment, noting that “[a]n increased penalty of this magnitude—two years to [fifteen] years—and limited to particularly defined offenses must, in our view, be reflected in the sentencing guidelines if the will of Congress is to be effectuated.”¹⁰⁶ Brown also emphasized the need to deter immigrant drug dealers, claiming that “[i]n the ordinary case, an alien drug dealer who illegally returns to the United States to practice his trade will continue this pattern of conduct until there is a substantial disincentive to do so.”¹⁰⁷ Brown was the only witness to discuss this enhancement at the hearings.¹⁰⁸ His analysis of it was perfunctory, taking up only two pages of the 171-page hearing transcript. By way of comparison, the Commission was simultaneously considering proposals for its corporate crime guidelines.¹⁰⁹ For those it held a public hearing with fourteen witnesses (including a former attorney general and two assistant attorneys general) that generated a 267-page transcript.¹¹⁰

The Commission conducted no policy analysis to justify the “aggravated felony” enhancement to the reentry guideline.¹¹¹ The Commission’s own staff declined to endorse the enhancement due to a lack of data.¹¹² No defense lawyers testified at the public hearing on the merits of the aggravated felony enhancement.¹¹³ Numerous federal prosecutors and law enforcement agents wrote to the Commission urging it to increase the penalties for reentry, with many of them citing Chiles’s law as a justification.¹¹⁴ The Sentencing Commission also interviewed practitioners in California and Texas, including INS agents, prosecutors, defense lawyers, and judges, who provided different perspectives on whether illegal reentry should be punished more harshly.¹¹⁵ Ultimately, aside from referencing Mr. Brown’s testimony, the Commission did not explain or justify the aggravated felony enhancement when it was adopted. The Commission enacted Mr. Brown’s proposal, but with sixteen points rather than twenty.¹¹⁶

These enhancements made the federal justice system much harsher for undocumented immigrants. They caused the average time in prison for unlawful reentry to more than double from 1991 to 2000, going from fifteen months to over thirty.¹¹⁷ The enhancements also helped the government bring many more reentry prosecutions. In 1991 only around 2,000 reentry cases were prosecuted in the federal courts, but by the year 2000 there were nearly 10,000 cases.¹¹⁸ The prior conviction enhancements facilitated this increase by giving federal prosecutors much more punishment leverage to compel plea bargains. And the federal prosecutors who lobbied the Commission for higher reentry guidelines in 1991 were keenly aware that higher guidelines meant more and faster guilty pleas.¹¹⁹ The increased reentry punishments empowered prosecutors to create a fast-paced plea system called “Fast Track.”¹²⁰ Under the Fast Track program, reentry defendants who qualified for an enhancement were offered a deal for a charge with a two-year maximum sentence.¹²¹ This deal required them to plead guilty at their first court appearance, waive the right to appeal, and agree to a specific sentence followed by immediate deportation.¹²² The stipulated sentence was usually two years, but prosecutors sometimes required sentences of two-and-a-half or four years.¹²³ If a defendant rejected this deal and had a prior felony or aggravated felony, then they faced a higher sentence under the mandatory Guidelines.¹²⁴ Fast Track thus used punishment leverage provided by the prior conviction enhancements to create a hyper-efficient plea process. The program was started in San Diego, but it spread to other border districts and allowed the massive growth of reentry prosecutions throughout the 1990s.¹²⁵

The Commission changed the reentry guideline in 2001 to create a more graduated system of prior conviction enhancements.¹²⁶ This change was motivated by concern over the proliferation of aggravated felonies. Congress repeatedly expanded the definition of “aggravated felony” over the course of the 1990s.¹²⁷ This caused a much larger number of crimes to trigger the sixteen-level enhancement. Initially, the enhancement had been narrowly focused on immigrants who committed murder, drug trafficking, and weapons trafficking.¹²⁸ But by 2001 the term “aggravated felony” applied to dozens of crimes, including fraud, forgery, burglary, immigrant smuggling, failing to appear for a sentence, crimes designated as “crimes of violence,” and many more.¹²⁹ The breadth of this list, combined with the severity of the sixteen-level enhancement, created some palpably unjust case outcomes. Defendants were regularly given much more severe sentences for reentry than they got for the original aggravated felony.¹³⁰ And a prior murder triggered the same sixteen-level enhancement as a prior forgery.

Federal judges and defense attorneys criticized the Guidelines for producing such harsh sentences.¹³¹ Judges also departed downward in about forty percent of cases, which the Commission characterized as judges “addressing this problem on an ad hoc basis.”¹³² The Commission rewrote the reentry guideline in 2001 in order to address these criticisms, and also to reduce the number of downward departures.¹³³ It created a tiered system of enhancements based on conviction type. Under the new guideline, different types of convictions would trigger enhancements of sixteen, twelve, eight, or four levels.¹³⁴ For example, the sixteen-level enhancement was limited to firearms offenses, drug trafficking, crimes of violence, immigrant smuggling offenses, and a few other enumerated crimes.¹³⁵ Less serious drug trafficking convictions (those with a sentence of thirteen months or less) triggered a twelve-level enhancement.¹³⁶ The remaining aggravated felonies resulted in only an eight-level enhancement.¹³⁷ And other felonies still caused a four-level enhancement.¹³⁸ The Commission also deleted an application note that permitted downward departures if an enhancement was based on a less-serious aggravated felony.¹³⁹

While the 2001 amendment made the reentry guideline somewhat less harsh, it still produced clear injustices. Relatively minor nonviolent crimes, like immigrant smuggling or low-level drug trafficking, triggered large twelve- or sixteen-point enhancements. And there were still no time limits, so crimes from several decades ago would count for the enhancements. The federal defense bar criticized these features of the guideline in letters and testimony before the Commission.¹⁴⁰ And in 2005, the Supreme Court declared in United States v. Booker that the Sentencing Guidelines were no longer mandatory.¹⁴¹ Federal judges could thus sentence defendants outside of the Guidelines range, through what became known as Booker variances. Judges took advantage of this new power to vary downward in illegal reentry cases in which the Guidelines were especially harsh.¹⁴² In one notable case, for example, the Ninth Circuit reversed a within-Guidelines fifty-two-month sentence because it was based on a prior conviction from twenty-five years ago.¹⁴³ From 2012 to 2016, about twenty-five percent of twelve-level and thirty-five percent of sixteen-level enhancement cases received below-Guideline sentences.¹⁴⁴ The Commission characterized this as an unusually high rate.¹⁴⁵

Federal judges and the Department of Justice also criticized the post-2001 reentry guideline for requiring use of the “categorical approach.”¹⁴⁶ Under that version of the guideline, the key question was whether a defendant’s prior conviction matched one of the enumerated crime categories triggering an enhancement. These categories included “crime of violence,” “firearms offense,” “drug trafficking offense,” “aggravated felony,” and more.¹⁴⁷ The categorical approach requires lawyers and judges to look at the statutory elements of a conviction, compare them to the generic federal version of the enumerated crime category, and determine whether the two match.¹⁴⁸ Many judges dislike the categorical approach because it raises difficult legal questions, creates additional litigation, and produces non-uniform results across different states.¹⁴⁹ And federal prosecutors repeatedly complained about the categorical approach at public hearings before the Commission, arguing that it produces arbitrary results and unnecessary litigation.¹⁵⁰

In 2016, the Commission rewrote the reentry guideline to address these criticisms.¹⁵¹ It made four major changes. First, the Commission keyed enhancements to the sentence imposed for the prior crime rather than the category of crime. This cut out the categorical approach. Under the new system, defendants receive the largest of a ten-, eight-, six-, or four-level enhancement depending on the length of the sentence imposed for a prior felony.¹⁵² In designing these enhancements, the Commission tried to approximate the enhancements previously imposed under the 2001 Guidelines for various kinds of prior convictions.¹⁵³ Second, the Commission added a new, second enhancement for a felony committed after the first deportation.¹⁵⁴ Prior versions of the guideline had only enhanced reentry sentences for felonies committed before the defendant was deported. This reflected Senator Chiles’s amendment, which only increased the maximum sentence for defendants whose removal was “subsequent to” a felony (or aggravated felony) conviction.¹⁵⁵ The Commission reasoned that it was arbitrary to enhance a sentence for pre-deportation crimes but not post-deportation crimes, so it added a second enhancement covering the latter.¹⁵⁶ Third, the Commission introduced time limits for the criminal history enhancements.¹⁵⁷ These time limits are the same as those for the normal criminal history score calculation—sentences of thirteen months or less count until ten years after imposition, and sentences of over thirteen months count until fifteen years after release from custody.¹⁵⁸ Fourth, the Commission added a new four-level enhancement for having been previously convicted of illegal reentry.¹⁵⁹

This new version of the reentry guideline mirrors the Sentencing Guidelines’ general criminal history calculation,¹⁶⁰ adding up to two additional (often much larger) prior conviction enhancements on top of it.¹⁶¹ This means reentry defendants now have their criminal history counted against them twice in two separate formulas, both based on prior sentence length. Other federal defendants are subjected to only one formula. The 2016 changes significantly lowered the amount of sentencing litigation in reentry cases, because they ended the use of the categorical approach.¹⁶² They also somewhat lowered the average reentry sentence.¹⁶³ But some defendants—namely those with a felony both before and after their first deportation—receive much higher sentences under the new guideline.¹⁶⁴ These changes did not seem to affect judges’ overall rate of downward variance, which has remained around eleven percent.¹⁶⁵ That is, notably, the lowest variance rate for all major categories of federal crime.¹⁶⁶

In sum, the reentry guideline has transformed repeatedly over three decades. The 1991 amendment reflected Senator Chiles’s (and Florida’s) panic over a purported wave of Haitian drug traffickers. It added sixteen levels for defendants deported after murder, drug trafficking, or weapons trafficking convictions. As Congress expanded the definition of “aggravated felony” over the 1990s to include a long list of crimes, the sixteen-level enhancement became obviously unjust in many cases. But rather than pare the guideline down, the Commission expanded it in 2001 to create graduated enhancements for many types of prior crime. Judicial, prosecutorial, and defense bar dissatisfaction with that version of the guideline caused the Commission to amend it again in 2016. The new version focuses on the length of prior sentences and adds an additional enhancement for post-deportation convictions. The reentry guideline has thus, in a process of preservation through transformation, morphed into a general criminal history enhancement that applies only to undocumented immigrants.¹⁶⁷ The Commission has yet to conduct any studies or publish any reports explaining why reentry defendants should be doubly punished for past crimes.¹⁶⁸

C. The First Step Act and Trump-Era Immigration Politics

Until 1984, federal prisoners were eligible for parole.¹⁶⁹ They could be released at the discretion of the United States Parole Board after serving at least one-third of their sentence. Under this system, the average federal prisoner served only forty-five percent of their sentence.¹⁷⁰ But Congress abolished federal parole in 1984, through the same law that created the Sentencing Commission.¹⁷¹ Congress replaced parole with a much less generous system of “good time” credits. Under this system, prisoners can now earn up to fifty-four days off for each year in prison.¹⁷² This means that, barring any other reductions, they must spend at least eighty-five percent of their sentences in prison.

The end of parole thus required federal prisoners to serve a much higher portion of their sentences. That did not change until 2018, when Congress enacted the First Step Act.¹⁷³ The First Step Act created a new system of “earned time” credits that further reduce a prisoner’s time in custody. Under this system, federal inmates are evaluated by a prison official to determine what kinds of in-custody programs they should participate in.¹⁷⁴ Such programs include, for example, prison work assignments, substance abuse treatment, psychological treatment, job training, and formal schooling.¹⁷⁵ As long as an eligible prisoner does not refuse to participate in these programs, they will receive ten days off of their prison sentence for every thirty days in custody.¹⁷⁶ Prisoners designated as “low” or “minimum” recidivism risks (fifty-five percent of federal prisoners) receive an additional five days off for every thirty days in custody.¹⁷⁷ “Low” and “minimum” risk prisoners also have their first year of earned time credit count toward early release from custody.¹⁷⁸ All other credit counts toward release into either home confinement or a halfway house.¹⁷⁹ Since halfway houses are designed for shorter stays to transition people out of prison, most of that time will go toward home confinement.¹⁸⁰ And prisoners can also receive the standard fifty-four days per year of good time credit on top of their earned time credit.¹⁸¹ Someone who earns all these credits will spend only about fifty-six percent of their sentence in a federal prison.¹⁸² The First Step Act has thus, at least for some, brought the federal system back to the old parole model where only about half of a sentence was served in prison.

But not all federal prisoners benefit from the First Step Act’s earned time credit system. Unlawful reentry defendants with a prior felony conviction are explicitly excluded.¹⁸³ So are all immigrants with a prior order of deportation.¹⁸⁴ And if a deportable immigrant who has not yet been ordered deported seeks to collect earned time credit, the law requires that they be put into deportation proceedings “as early as practicable” during their incarceration.¹⁸⁵ Thus, previously deported defendants (including all unlawful reentry defendants) cannot collect earned time credit, and deportable immigrants who try to collect it will swiftly be ordered deported.¹⁸⁶ The First Step Act also contains a list of forty-seven other criminal charges that trigger exclusion from earned time credits.¹⁸⁷ These are mostly violent crimes, and all of them are much more serious crimes than unlawful reentry. They include, for example, convictions related to explosives, drive-by shootings, arson, child pornography production, terrorism, torture, weapons of mass destruction, sex trafficking, and being a drug kingpin.¹⁸⁸ Defendants convicted of these crimes, as well as undocumented immigrants, only have access to good time credits. They therefore must serve at least eighty-five percent of their sentences.¹⁸⁹

The original version of the First Step Act allowed undocumented immigrants to collect earned time credit.¹⁹⁰ The bill passed by the House of Representatives in May 2018 did contain a list of conviction-based exclusions, including one for unlawful reentry with a prior felony, but not the broader immigration status exclusion.¹⁹¹ That was added when the bill went to the Senate. Contemporaneous reporting indicates that a bipartisan group of senators agreed to an amended version of the First Step Act in November 2018.¹⁹² This new version included the provision excluding all deportable immigrants from earned time credits.¹⁹³ The senators made that and other changes in negotiations with the White House, especially President Trump’s son-in-law Jared Kushner who played a significant role pushing for the law.¹⁹⁴ The senators’ goal with these changes was to persuade Senate Republicans to support the First Step Act, and ultimately to convince President Trump to sign it into law.¹⁹⁵ Including deportable immigrants was thus a nonstarter. Shon Hopwood, who worked directly with the Trump administration on the law, observed that “providing federal tax money to pay for rehabilitation programs and early release for those who will be deported had no chance of passing in the current Congress.”¹⁹⁶

These negotiations coincided with the first Trump Administration’s most intense period of hostility to Latin American immigration. In April 2018, shortly before the House passed its version of the First Step Act, President Trump made a series of posts on Twitter warning about immigrant caravans from Latin America. He wrote in one post, for example, that “Honduras, Mexico and many other countries that the U.S. is very generous to, sends many of their people to our country through our WEAK IMMIGRATION POLICIES. Caravans are heading here. Must pass tough laws and build the WALL.”¹⁹⁷ Trump also gave a speech claiming that immigrants in these caravans were committing widespread rapes.¹⁹⁸ Shortly after Trump’s posts, Attorney General Jeff Sessions announced a “zero tolerance” policy toward undocumented immigration.¹⁹⁹ This new policy caused an enormous increase in prosecutions for unlawful entry and unlawful reentry.²⁰⁰ That increase continued through 2018 and 2019.²⁰¹ The Trump Administration also, as part of its 2018 “zero tolerance” policy, systematically separated children from their parents at the border so that the parents could be prosecuted.²⁰² This family separation policy sparked a nationwide protest movement against the administration’s immigration policies.²⁰³ And in November 2018, while the First Step Act compromise was being negotiated, President Trump issued an executive order categorically denying asylum to immigrants who entered unlawfully.²⁰⁴ In signing the order Trump declared, “The continuing and threatened mass migration of aliens with no basis for admission into the United States through our southern border has precipitated a crisis and undermines the integrity of our borders.”²⁰⁵ This larger context of anti-immigrant policy and rhetoric explains why the Senate, in negotiations with the White House, excluded deportable immigrants from the new earned time credit system.

Immigrants’ rights groups and some Democratic House members opposed the First Step Act’s exclusion of undocumented immigrants. A coalition of activist groups including the NAACP, the ACLU, and the National Immigrant Justice Center raised concerns about the law in letters and press releases.²⁰⁶ This coalition cited, among other problems, that the law “excludes from its reforms most undocumented immigrants,” that it thereby “further criminalizes migration,” and that these exclusions “could also have a disparate impact on racial minorities.”²⁰⁷ A coalition of immigration-related nonprofits expressed concern that the law “sets the precedent that immigrants are outside the scope of people who should benefit from criminal justice reform.”²⁰⁸ And in the House debates concerning the First Step Act, several Democratic Representatives highlighted its discriminatory impact on immigrants. Representative Pramila Jayapal noted that she was “very concerned about language in the bill that excludes immigrants from being eligible for time credits.”²⁰⁹ And Representative Jerrold Nadler observed that “the new incentive system for pre-release custody credits could exacerbate racial biases,” because it “excludes large categories of inmates based on convictions for various offenses and on immigration status.”²¹⁰

The First Step Act enacted a generous new earned time credit system that, for those who qualify, can reduce a sentence by up to one-third. But the law also categorically excluded undocumented immigrants, keeping them trapped in a much more punitive system allowing only good time credits. That exclusion was crafted during the first Trump Administration’s most intense period of hostility toward Latin American immigration. And activists and legislators at the time pointed out its discriminatory consequences.

II. The Discrimination Problem

These policies create two separate and unequal federal sentencing systems: one system for U.S. citizens and another system for deported immigrants. In the citizens’ system, criminal history is counted once; in the immigrants’ system, it is counted twice. In the citizens’ system you can serve up to forty-five percent of your sentence out of prison, in the immigrants’ system you can serve only fifteen percent. This separate system for immigrants is not small. Since 2010, about one-third of all federal cases have been unlawful reentry prosecutions.²¹¹ And over ninety-nine percent of the defendants in those cases have been Latin American.²¹² Federal sentencing law thus creates a caste structure in which undocumented immigrants from Latin America are relegated to a punishment underclass. This Part makes the straightforward observation that this is discriminatory. It does so by comparing these two sentencing systems in detail. It shows that federal sentencing law discriminates facially by immigration status, and in effect by race and ethnicity.²¹³ The more complicated issue of constitutionality is explored in Part IV.

A. Double Counting Criminal History

The unlawful reentry guideline stands alone in basing a defendant’s offense level calculation entirely on their past crimes.²¹⁴ Other federal sentencing guidelines focus instead on the facts of the current crime. For drug trafficking, what matters is the amount and type of drugs.²¹⁵ For fraud, it is the amount of money lost or stolen.²¹⁶ For child pornography distribution, it is the number of images and the ages of the victims, among other factors.²¹⁷ But for unlawful reentry, facts about the current crime are not factored in.²¹⁸ The only thing that matters is the defendant’s criminal history. The reentry guideline enhances a defendant’s sentence once along the “criminal history category” axis, and a second time along the “specific offense level” axis. Past crimes thus count against a reentry defendant twice in two separate formulas. For nearly all other federal defendants, they count only once.²¹⁹

There are a few other places in the Guidelines where a past conviction triggers an increase in the offense level. A couple of guidelines have modest enhancements for defendants who committed the same kind of crime previously.²²⁰ For example, the guidelines for immigrant smuggling, passport fraud, and immigration fraud all contain a two-level enhancement for having been previously convicted of a “felony immigration and nationalization offense.”²²¹ The guideline for adulterating or misbranding food products contains a four-level enhancement for a prior conviction for the same offense.²²² And the guideline for domestic violence contains a two-level enhancement for a pattern of “stalking, threatening, harassing, assaulting the same victim,” which includes prior convictions.²²³ In addition, the guideline for being a felon in possession of a firearm ties the offense level to the nature and number of prior felony convictions.²²⁴ And the drug trafficking guideline increases the offense level if both (1) the drugs caused death or serious bodily injury and (2) the defendant has a prior felony drug offense.²²⁵

These other criminal history enhancements are much narrower than the reentry enhancements.²²⁶ They only apply to specific kinds of prior crimes, while the reentry enhancement applies to any prior felony. These other enhancements also have a much closer nexus to the defendant’s current crime. Most of them involve having committed the same kind of crime previously. For the felon-in-possession enhancements, the prior felony is itself an element of the new crime.²²⁷ And the drug trafficking criminal history enhancements only apply if the drugs caused death or serious bodily injury.²²⁸ In these circumstances, the prior crimes are directly relevant to the defendant’s conduct in the new case. By contrast, the enhancements in the unlawful reentry guideline apply to all criminal history in general, not just related crimes.

The reentry guideline’s criminal history enhancements are also much larger than other comparable enhancements. A single prior felony will get a reentry defendant between four and ten additional offense level points, depending on the sentence length.²²⁹ And this can happen up to twice, once for a felony before the first deportation and once for a felony after it.²³⁰ Nowhere else in the Guidelines can a defendant get such a high increase for just a single prior conviction. It is thus actually something of an understatement to say that reentry defendants have their convictions counted twice, because the extra enhancement is potentially much more severe than the Guidelines’ normal criminal history calculation. A prior conviction with a five-year sentence, for example, gives a reentry defendant both a one-level increase in their criminal history category and a ten-point increase in their offense level.²³¹ This takes a reentry defendant from a zero to six month range to a thirty to thirty-seven month range.²³² Other federal defendants starting from the same baseline would only see the one-level criminal history category increase, going from zero to six months to four to ten months.²³³ Reentry defendants can thus receive multiple additional years in prison due to past crimes that would net other defendants only a few extra months.

These enhancements apply in a significant percentage of reentry cases. According to the U.S. Sentencing Commission’s 2023 data, 43.6% of unlawful reentry defendants received a prior conviction enhancement.²³⁴ The average sentence for unlawful reentry was twelve months in 2023, and since 2012 it has fluctuated between nineteen months and eight months.²³⁵ But reentry defendants with prior convictions receive much longer sentences due to criminal history enhancements. According to the Commission, about 15% of reentry defendants since 2021 received sentences of at least two years, and 1% received sentences of at least five years.²³⁶ At the top of this distribution, hundreds of reentry defendants have gotten more than five years, and some even more than a decade in prison.²³⁷

The reentry guideline is also the only federal sentencing guideline that applies exclusively to noncitizens. It is used for four crimes: 8 U.S.C. § 1326 (unlawful reentry), 8 U.S.C. § 1325(a) (felony unlawful entry), 8 U.S.C. § 1253 (failure to depart after deportation order), and 8 U.S.C. § 1185(a)(1) (violating prescribed entry and departure regulations).²³⁸ By their explicit terms, these four provisions only apply to noncitizens.²³⁹ There are several other federal entry crimes that do apply to U.S. citizens, including passport fraud, entering the country without inspection, and entering the country through forgery or false statements.²⁴⁰ But those crimes are not covered by the reentry guideline, and do not receive its extra criminal history enhancement.²⁴¹ So, for example, a U.S. citizen fugitive who tries to reenter the country with a fake passport will not receive a double enhancement. And arguably comparable crimes that punish illicit presence, such as escaping from federal prison, failing to appear in court, failing to register as a sex offender, and trespassing on federal property, also receive no double enhancement.²⁴² Only counting criminal history twice in the reentry guideline thus constitutes explicit discrimination by immigration status. And because the reentry guideline is by far the most racially skewed, with over ninety-nine percent of defendants being of Latin American ancestry, doing so also effectively discriminates by race.²⁴³

The main rationale used to justify these criminal history enhancements is that they deter immigrants with prior convictions from reentering the United States. That was the rationale Senator Graham and Sentencing Commission witness Joe Brown invoked in 1991 when arguing for extra recidivist enhancements.²⁴⁴ It is also the primary rationale that federal judges have cited in upholding the enhancements.²⁴⁵ Though, notably, the Sentencing Commission itself has never announced an official justification for the enhancements.²⁴⁶ The logic of a deterrence rationale is simple—by imposing higher sentences, the reentry enhancements give immigrants with past convictions a stronger disincentive to return. But the Commission has not conducted any study or other policy analysis to explain why the normal criminal history enhancement inadequately deters.²⁴⁷ Rather, the story of the reentry guideline is one of politics and path dependence. As discussed above, the initial criminal history enhancement was a response to Senator Chiles’s 1988 statutory sentence increase.²⁴⁸ And that increase was, in turn, motivated by Floridians’ fear of Haitian drug dealers supposedly reentering the country to commit crimes.²⁴⁹ The Commission set the original enhancement at sixteen points for no recorded reason, and then revised it in 2001 and 2016 to accommodate judges’ and prosecutors’ complaints.²⁵⁰ But the size and nature of the reentry enhancements have never been connected to any empirical data about deterrence, or about reentry defendants committing additional crimes. Indeed, it is unlikely that first-time reentry defendants have any idea their Guidelines calculation will be enhanced so severely by their past criminal record.²⁵¹

These double enhancements also impose a great deal of suffering on defendants and their families. Many reentry defendants have deep ties to the United States. According to Commission data from 2013, 53.1% of reentry defendants first entered the United States when they were minors,²⁵²and 49.5% of reentry defendants have children living in the United States.²⁵³ And 74.5% had worked in the United States for more than a year before their arrest.²⁵⁴ People with criminal history in the United States are also more likely to have lived in the United States for an extended period.²⁵⁵ Thus, by focusing on prior convictions, the reentry guidelines direct more punishment at people with strong personal and family ties to this country. The prototypical target of the reentry enhancements is not a foreign drug dealer returning to commit more crimes. It is someone who grew up in the United States, has a family here, and wants to return to them. Many such immigrants spend large portions of their lives in federal prison because they repeatedly try to return to their spouses, their children, and the country they grew up in.²⁵⁶

B. Doing More and Harder Time

The First Step Act also discriminates against undocumented immigrants. It creates two different classes of prisoner: a majority who can collect earned time credits, and a minority who cannot.²⁵⁷ U.S. citizens are in the former category, excepting those who are in prison for a disqualifying crime like torture, child pornography production, or terrorism.²⁵⁸ Deportable immigrants, including all reentry defendants, are in the latter category.²⁵⁹ Prisoners who qualify for earned time credits can get up to fifteen days off their sentence for every thirty days served.²⁶⁰ That means they can serve up to one-third of their sentence in a halfway house, home confinement, or released from custody altogether. Prisoners who do not qualify for earned time credits only get the standard fifty-four days of good time credit per year. And the exclusion of deportable immigrants is also racially discriminatory, since nearly all deportable immigrants in federal prison are Latin American.²⁶¹

To defend this discrimination, one might argue that it makes sense to deny limited rehabilitation resources to prisoners who will be deported. Perhaps the federal government has an interest in limiting programs like drug treatment and job training to U.S. citizens. But this argument ignores a key fact about the earned time credit system: a prisoner can collect custody credits even while they are not participating in anti-recidivism programs. As long as a prisoner does not refuse to participate in programs assigned to them, they remain in earning status.²⁶² And this happens even if programs are unavailable, or if the Bureau of Prisons decides the prisoner does not need programs. For example, prisoners assigned to programs that lack openings still earn time credits while waiting for a spot.²⁶³ Prisoners also earn time credits for working jobs within the prison, like cleaning laundry or working in the kitchen.²⁶⁴ Such work assignments are standard in federal prison, and all able-bodied inmates (including deportable immigrants) are required to perform them.²⁶⁵ Many U.S. citizen prisoners receive earned time credit under the First Step Act for performing these prison jobs.²⁶⁶ But deportable immigrants do not receive that credit. The federal government has thus created a system where two prisoners—one a U.S. citizen and the other an

immigrant—will work alongside each other at the same job in the same prison kitchen, with the citizen getting a one-third sentence reduction for their labor and the immigrant getting nothing.

The federal prison system discriminates against deportable immigrants in numerous other ways as well. In addition to First Step Act credits, a federal prisoner can spend ten percent of their sentence (up to six months) in home confinement.²⁶⁷ Deportable immigrants, however, cannot get home confinement.²⁶⁸ They are excluded from nearly all in-prison treatment and rehabilitation programs.²⁶⁹ This includes the Residential Drug Abuse Program, a 500-hour treatment program that gives a one-year sentence reduction upon completion.²⁷⁰ Deportable immigrants cannot receive a “minimum security” designation by the Bureau of Prisons, meaning that they cannot be housed in the least restrictive prison facilities.²⁷¹ Unlike other prisoners, they cannot serve part of their sentences in a halfway house.²⁷² While the Bureau of Prisons is required to try to house inmates within 500 miles of their families, it does not do so for deportable immigrants.²⁷³ And the federal prison system houses most deportable immigrants in segregated facilities that have fewer amenities and less security than normal prisons, feature little in-custody programming, and are run by private corporations.²⁷⁴ In conjunction with the earned time credit exclusion, these policies create a separate and unequal prison system for undocumented prisoners.

III. The Arbitrariness Problem

The reentry guideline is also quite illogical in practice. It distributes prison time through a convoluted formula that is tangential to the severity of a defendant’s criminal history. From the defendant’s perspective, it feels like a twisted lottery—prison time is distributed at random based on what numbers turn up in what order. This Part explores several features of the guideline that cause it to treat defendants so arbitrarily. The basic problem is that up to two large (commonly multi-year) sentence enhancements are imposed based on superficial aspects of a defendant’s record. First, the guideline places a puzzling emphasis on the order of convictions vis-à-vis the first deportation. If all convictions occurred either before or after the first deportation, then there is only one enhancement. But if two convictions straddle the first deportation, there are two enhancements. Thus, multiple years in prison often turn on the order of events. Second, the guideline only looks at nominal sentences, not actual time served. The enhancements are thus inflated by state sentencing policies, like indeterminate parole and automatic good time credit, that make the nominal sentence much longer than the real one. Third, violations of criminal supervision (e.g., probation) artificially inflate these past sentences and cause very old convictions to still trigger enhancements. This is a common problem for reentry cases because deportation itself often results in probation violations. Fourth, the difficulty of obtaining conviction and deportation records, combined with the fast pace of guilty pleas in these cases, leaves many defendants in the dark. They often plead guilty thinking their prison term will be brief, only to find out that it will be much longer. This Part explains these sources of arbitrariness, showing how they exacerbate the guideline’s anti-immigrant discrimination. It also illustrates them with a real-world case example.

A. The Criminal History Lottery

Timing the Deportation

Before 2016, the reentry guideline only gave an enhancement for past convictions that were followed by a deportation.²⁷⁵ This mirrored Chiles’s amendment to § 1326, which increased the maximum punishment if a defendant was deported “subsequent to” a felony or aggravated felony.²⁷⁶ Convictions occurring after a defendant’s most recent deportation did not trigger an enhancement. The criminal history enhancement was thus a collateral consequence of being deported after a conviction, and not of a conviction alone. This reflected the Commission’s apparent purpose of deterring immigrants deported after felony convictions from returning to the United States.²⁷⁷

When the Sentencing Commission rewrote the reentry guideline in 2016, it expressed concern that just focusing on pre-deportation convictions was arbitrary.²⁷⁸ Take two defendants who committed the exact same prior crime, one before being deported and the other after being deported. Following the Commission’s logic, it made little sense to give the former defendant multiple more years in prison than the latter.²⁷⁹ But rather than simply removing the requirement that a conviction occur before a deportation, the Commission added a second enhancement.²⁸⁰ Under the new version of the guideline, a defendant can get up to two enhancements of up to ten specific offense points. The first enhancement is for a conviction occurring before the defendant’s first deportation, and the second is for a conviction occurring after their first deportation.²⁸¹ This solved the arbitrariness problem, by the Commission’s reasoning, because it treated defendants with pre- and post-deportation felonies equally.

However, the 2016 amendment added a whole new dimension of arbitrariness. Under this new system, the order of convictions vis-à-vis the first deportation is enormously important. And that fact bears little relationship to the actual severity of someone’s prior record. Consider the following four defendants, who suffer deportations and felony convictions in the following order:

Defendant A: Felony, Deportation, Felony: two enhancements (one under Guidelines section 2L1.2(b)(2), another under (b)(3))
Defendant B: Felony, Felony, Felony, Felony, Deportation: one enhancement (under (b)(2))
Defendant C: Deportation, Felony, Felony, Felony, Felony: one enhancement (under (b)(3))
Defendant D: Deportation, Felony, Deportation, Felony, Deportation, Felony, Deportation, Felony: one enhancement (under (b)(3))

Defendant A has the least serious criminal record, with just two felonies. But only Defendant A will receive two enhancements under the reentry guideline. Every other defendant gets just one enhancement, despite having more convictions than A and as many or more deportations than A. Why is Defendant A treated much more harshly? There is no logical explanation for distributing prison time this way. If the Commission’s goal is to punish people more severely for returning to the United States and committing additional crimes, it should not focus just on the first deportation. For example, Defendant D above receives no second enhancement despite being convicted of crimes both before and after their deportations. According to a study by the Sentencing Commission, in 2013 the median number of prior deportations for reentry defendants was 2, and the mean was 3.2.²⁸² According to the same study, the median number of past criminal convictions was 3 and the mean was 4.4.²⁸³ Conditioning sentence severity on the precise sequence of these events makes little sense.²⁸⁴ And from the defendant’s perspective, it seems quite random. Due to the size of these enhancements (between four and ten offense points), defendants who receive two of them are commonly looking at five to ten years in prison, sometimes even more.²⁸⁵ For example, assuming that all the felonies in the above illustration carried two-year sentences, Defendant A would have a recommended sentence range of sixty-three to seventy-eight months.²⁸⁶ Without the second enhancement, Defendant A’s range would be only twenty-seven to thirty-three months.²⁸⁷ And Defendants B, C, and D would face a range of forty-one to fifty-one months, despite having much more extensive felony records than A.²⁸⁸

Prosecutors can also exercise unilateral discretion over whether a post-deportation enhancement applies. In situations where a previously deported person is charged with a state crime, the federal prosecutor has a choice of when to bring them to federal court on the reentry case.²⁸⁹ If the prosecutor waits for the state case to end, and it results in a conviction and sentence, the defendant will then receive a sentencing enhancement in their federal reentry case.²⁹⁰ But if the prosecutor instead brings the reentry case first, there is no extra enhancement. The reentry guideline thus empowers prosecutors to manipulate a defendant’s sentence. And given the potential size of the enhancement, this sometimes means doubling or tripling a defendant’s prison term.²⁹¹ Different prosecutors’ offices have different policies on this question, with some prosecuting the reentry case right away and others waiting.²⁹² That adds yet another dimension of randomness to reentry sentencing.²⁹³

Misinterpreting State Sentences

Before 2016, the reentry guidelines’ prior conviction enhancements were keyed to substantive crime categories. These included “crime of violence,” “drug trafficking offense,” “aggravated felony,” and more.²⁹⁴ This caused much litigation using the categorical approach, with lawyers arguing over which state crimes did or did not qualify for which enhancements.²⁹⁵ Responding to judges’ and prosecutors’ complaints about this litigation, the Commission changed the guideline in 2016 to focus instead on past sentences. Under this new version of the guideline, the length of a sentence is treated as a proxy for the seriousness of the underlying crime.²⁹⁶ If the “sentence imposed” for a prior crime was five years or more, it is worth ten points; if it was at least two years and less than five, it is worth eight points; if it was more than thirteen months and less than two years, it is worth six points; and all other felonies are worth four points.²⁹⁷ This new system has reduced litigation over reentry sentences.²⁹⁸ But it interprets state sentences in an incredibly unsophisticated way. In doing so, it creates arbitrary differences in punishment, and artificially inflates many enhancements, based on variations in state sentencing procedures.

The current guideline looks at the nominal sentence imposed by a court, not at the amount of time a defendant served in custody.²⁹⁹ But the relationship between those two numbers varies widely between (and within) state systems. Some states require a defendant to serve all or nearly all of their sentence. Others set the sticker price of a sentence much higher than the time actually served. Consider some examples. In California, by default defendants serve only half of the announced prison sentence (and, under the recent “realignment” system, some defendants serve much less than that).³⁰⁰ Arizona and Florida, by contrast, are truth-in-sentencing states, and require all defendants to serve at least eighty-five percent of the announced sentence.³⁰¹ In Minnesota, felony defendants are released on supervision after serving two-thirds of their sentences.³⁰² And discretionary parole systems (which most states have) introduce further complications.³⁰³ In Utah, third-degree felonies are sentenced to an indeterminate term of zero to five years, and second-degree felonies get one to fifteen years.³⁰⁴ In Texas, defendants are eligible for parole after serving twenty-five percent of their sentence.³⁰⁵ But for all of these different state sentencing systems, the reentry guideline treats the maximum possible sentence as the “real” sentence. It does not matter if everyone in the courtroom—the judge, the lawyers, the defendant—understood that only a fraction of that time would be spent in prison. It does not matter if the sentence was an indeterminate term of zero to five years, and the defendant was paroled after just two months. The reentry guideline is deliberately ignorant of these facts. The sticker price is the sentence.

The 2016 amendment thus replaced one form of arbitrariness with another. Under the categorical approach regime, seemingly irrelevant differences in the elements of state crimes determined whether a prior conviction counted for an enhancement.³⁰⁶ Under the nominal sentence length regime, seemingly irrelevant differences in state sentencing procedures determine an enhancement’s size.³⁰⁷ The new system eases the burden of litigation on judges and prosecutors, but it maintains the burden of arbitrary treatment on defendants. One problem is that people with substantively similar prior sentences will get very different enhancements due to differences in state procedures. Another, related problem is that the nominal sentence often significantly overstates the seriousness of the underlying crime.³⁰⁸ For example, in Utah’s discretionary parole system you can serve just a few days in custody on a sentence that the Guidelines will count as five years long.³⁰⁹ But “five years” is a terrible proxy for the severity of the actual crime in such a case. And it seems seriously unjust to get a ten-point enhancement for conduct that only merited a short stay in jail. A system based on time actually served would be far more rational. Such an approach has precedent in the federal system: federal drug crimes use a time-served model for determining recidivist sentence enhancements.³¹⁰ But, for the sake of administrative convenience, the Sentencing Commission rejected that approach when it amended the reentry guidelines.³¹¹

Supervision Complications

Criminal supervision introduces further problems. Supervision sentences are very common in the United States: about 3.7 million people are currently on some form of probation or parole.³¹² When someone is sentenced to supervision, they can be put in custody for violating certain rules. This can happen if they do things like refuse to meet with a probation officer, fail a drug test, or commit a new crime.³¹³ Under the Guidelines, any sentence for a violation of criminal supervision gets added to the sentence for the original crime.³¹⁴ So if a person is initially sentenced to zero days in prison plus probation, and then gets a 219-day sentence for violating probation conditions, the Guidelines treat that as a 219-day sentence for the original crime.³¹⁵ In the context of the reentry guideline, this means supervision violations enlarge the prior conviction enhancements by adding to the measured lengths of past sentences.³¹⁶

Reentry defendants are especially susceptible to supervision violations, because deportation itself often causes them.³¹⁷ One standard supervision condition is that the supervisee must report in person to a probation or parole office.³¹⁸ But when someone is deported after release from jail, they cannot go meet their probation officer. This can trigger a violation report and an arrest warrant.³¹⁹ In addition, many state courts impose immigration-related conditions on undocumented supervisees, such as prohibitions on returning to the United States without legal permission or requirements to report to a probation officer upon return.³²⁰ When defendants are punished for violating immigration-based conditions, it also inflates their sentence enhancement in any later reentry case. This creates yet more double punishment—the state gives you a probation violation for returning to the United States, then the federal system treats that violation as evidence of separate bad conduct meriting an increased sentence for returning to the United States.³²¹

The Guidelines’ treatment of supervision violations also reproduces the problems discussed in the above Sections. Consider a person sentenced to probation and then deported for the first time. If that person returns to the United States and is prosecuted for a new crime, they will also get a probation violation.³²² If the sentence for that probation violation runs concurrent to the sentence on the new crime, then a single term in prison will count for two separate enhancements.³²³ Further, a supervision violation can also result in a nominal sentence that is higher than the real sentence. For example, in California the half-time rule applies to probation violations, and in Utah a violation sentence can include an indeterminate prison term with parole.³²⁴ In such cases the violation is counted as the maximum possible custody time rather than the actual time spent in prison.³²⁵ Consider the case of United States v. Gomez-Colin.³²⁶ The defendant was initially sentenced to probation on a Georgia felony conviction and then deported.³²⁷ He later returned to the United States and received a state probation violation for reentering the country unlawfully.³²⁸ This resulted in a five-year sentence that was set aside by the state court (meaning he did not serve it).³²⁹ But the federal court treated it as a full five-year sentence under the reentry guideline, giving him a ten-point enhancement.³³⁰ The guideline’s treatment of violation sentences thus compounds with its treatment of indeterminate sentences to inflate the reentry enhancements even further.

Lastly, supervision violations also cause very old cases to trigger sentencing enhancements. For any sentence of over thirteen months, the reentry guideline only imposes an enhancement if the defendant was in custody within fifteen years of their arrest for reentry.³³¹ So if someone served a two-year sentence but was released from custody more than fifteen years ago, that prior sentence does not count. However, supervision violations extend this timeline forward.³³² Thus, if someone served a two-year prison sentence and was released on parole thirty years ago, but they served time on a later parole violation within the last fifteen years, the entire conviction still counts under the reentry guideline.³³³ This is a common problem for reentry defendants, because deportation itself triggers supervision warrants.³³⁴ If a supervisee fails to report because they were deported, the court can issue a probation warrant that indefinitely pauses the clock on the term of probation.³³⁵ Then, if the supervisee is arrested and jailed many years later on the warrant, the supervision violation causes the underlying crime to count for a Guidelines enhancement. This is true regardless of how long ago the underlying conviction occurred.³³⁶ Reentry defendants are thus punished anew for decades-old crimes.

Slow Court Records, Fast Guilty Pleas

Given all these complications, it is often quite difficult to figure out a reentry defendant’s Guidelines range. First you need to get their conviction records, their supervision revocation records, and their deportation records. This commonly means requesting documents from multiple state and federal government bodies. Ordering these records is often a complicated matter.³³⁷ Many county courthouses only use paper documents, and many require records requests to be made in person or by mail.³³⁸ And the federal government takes months to provide immigrants with their deportation records.³³⁹ Indeed, a large portion of federal immigration files are kept in a defunct limestone mine in Missouri, and requested documents must be physically located there before they are provided.³⁴⁰ Once you have all of these documents, you must piece together the defendant’s deportation and criminal history, and put it in chronological order to figure out the recommended sentence. For the defendant, this exercise is a high-stakes logic game. It produces a longer or shorter sentence based on the specific features (and sequence) of their prior record.

Unfortunately, this logic game has a short time limit. Unlawful reentry prosecutions proceed very quickly. Over ninety-seven percent of them end in guilty pleas.³⁴¹ About one-quarter of those pleas are obtained through a prosecutor-controlled program called “Fast Track.”³⁴² The Fast Track program provides the defendant with a somewhat lower Guidelines range (usually a two- or four-point reduction) in exchange for (1) not filing any motions or appeals, and (2) pleading guilty within thirty days of their arrest.³⁴³ Reentry cases in the Fast Track program thus resolve very quickly.³⁴⁴ Even outside the Fast Track program, reentry cases tend to end in much faster guilty pleas than other federal crimes.³⁴⁵ Most reentry prosecutions are brought in high-volume border districts, where they comprise the majority of cases.³⁴⁶ Judges, defense lawyers, and prosecutors in those districts expect reentry cases to be routine guilty pleas and process them accordingly.³⁴⁷

This means defendants are pressured to plead guilty without having access to their full deportation and conviction records. The prosecutor usually provides a RAP sheet with basic prior conviction information from the FBI’s criminal records database, but such RAP sheets are incomplete. They often lack information about the dates of past convictions, the sentences, and any probation or parole violations.³⁴⁸ That information must be obtained from state agencies, which is a burdensome process. And while complete deportation records should theoretically be provided in discovery, reentry defendants who plead guilty have no right to pre-plea disclosure of those records.³⁴⁹ The defendant’s full deportation and criminal history is thus often unknown until a court-appointed probation officer obtains the records while writing a presentence report. Presentence reports are prepared after the conviction and take several months to complete.³⁵⁰ When provided to the defendant before sentencing, they often contain unpleasant surprises. If the probation officer finds a deportation order or supervision violation that the defense lawyer missed, multiple years can be added to the expected sentence.³⁵¹ Defendants thus often plead guilty thinking their prison term will be relatively short, only to find out right before the sentencing hearing that it will be much longer. The rapidity of guilty pleas in these cases, combined with the difficulty of obtaining records quickly, keeps defendants in the dark.

B. The Case of Mr. R

The reentry guideline’s criminal history enhancements thus create numerous problems. They produce arbitrary and illogical punishments, inaccurately measure prior sentence lengths, and make it difficult to ascertain a defendant’s sentencing range before the guilty plea. It is helpful to illustrate these problems with a concrete case. To that end, here is the story of Mr. R, a reentry defendant who was sentenced to ninety-six months in prison in 2014.³⁵² The sentencing in his case illustrates several of the reentry guideline’s absurdities.

Mr. R was born in a Latin American country. He moved to the United States with his father at the age of eight. In his late teens and early twenties, he collected a series of criminal convictions in California. In June 1995, when Mr. R was eighteen years old, he was convicted of stealing a car.³⁵³ He was sentenced to 365 days in jail and three years of probation. Shortly after his release, in September 1995, he was convicted of possessing cocaine with intent to sell.³⁵⁴ He was sentenced to three years in prison for the drug conviction and given a probation violation on the auto theft case with a concurrent three-year sentence. After his release from custody, he was deported from the United States for the first time in November 1997. He also received two separate three-year terms of parole in California, one for each of the 1995 convictions.³⁵⁵

Mr. R then reentered the United States and returned to California. He was convicted of robbery in 1998.³⁵⁶ For that conviction he received a state sentence of fifteen years in prison. He also received two concurrent parole violations in each of the 1995 cases. He remained in prison in California on this sentence until 2012, when he was released from custody and deported again. He then reentered the United States and was prosecuted federally for unlawful reentry. Under the pre-2016 Guidelines, he received a sentence of fifty-seven months in prison because of his prior robbery conviction.³⁵⁷

Fast forward to 2018. Mr. R is arrested again for entering the United States and charged with unlawful reentry. He is now in his early forties, and his most recent state conviction is from 1998. Mr. R’s court-appointed defense lawyer knows that the 1998 robbery conviction will trigger a ten-point enhancement, because Mr. R was released from custody within the last fifteen years.³⁵⁸ But the lawyer does not think either of the 1995 convictions will count, because they are too old. This means Mr. R will receive just one prior conviction enhancement (for a post-first-deportation felony).³⁵⁹ The lawyer tells Mr. R that he is looking at a Guidelines range of forty-six to fifty-seven months, and Mr. R pleads guilty less than two months after his arrest.³⁶⁰

Between the guilty plea and sentencing, a court-appointed probation officer orders Mr. R’s criminal case records from California to prepare a presentence report. These records reveal that the two 1995 convictions produced parole violations that ran concurrent with the 1998 robbery sentence. The probation officer concludes that those 1995 convictions should therefore be counted in Mr. R’s 2018 Guidelines calculation, because he was in custody within the relevant timeframe. This has two consequences for the sentence. First, it gives Mr. R six more criminal history points, putting him in the Guidelines’ highest criminal history category. Second, it gives Mr. R a large new pre-first-deportation enhancement under the reentry guideline. The 1995 convictions occurred before Mr. R’s first deportation, so they trigger a separate enhancement from the 1998 sentence. And even though the parole violation ran concurrent with the new robbery sentence, it is added to the 1995 sentence and thus the pre-deportation sentence enhancement.³⁶¹ Based on this calculation, the probation officer concludes that Mr. R’s Guidelines range is 130 to 162 months.

The defense lawyer makes no objections to the probation officer’s conclusions. At the sentencing hearing, the defense lawyer argues that these convictions are quite old and the product of Mr. R’s troubled youth, while today Mr. R is a deeply religious man who works as a Christian pastor. Mr. R complains to the judge that his defense lawyer has not communicated with him about the surprise increase in his sentence. The judge proceeds with sentencing anyway and gives Mr. R ninety-six months in prison.

Subsequent litigation revealed that this Guidelines range was in fact wrong. The 1995 convictions should not have counted for the 2018 Guidelines calculation, because Mr. R’s parole violation sentences were discharged in September 2000. California law provides that a person cannot be kept in custody longer than four years on a parole violation for a noncapital felony.³⁶² This means Mr. R was not in custody on those cases within fifteen years of 2018. But his discharge from parole could only be discovered by looking at prison administrative records kept by the California Department of Corrections. Mr. R’s discharge was noted in pencil on a paper log in his prison file, which could only be obtained through a written request.³⁶³ Because of this error, Mr. R should have been in a lower criminal history category, and he should not have received a pre-deportation felony enhancement. Mr. R’s actual Guidelines range should have been forty-six to fifty-seven months.

This case illustrates several absurdities about unlawful reentry sentencing. First is the strange focus on the timing of the first deportation. Mr. R was deported from the United States multiple times, but his recommended sentence was dramatically increased because his first deportation happened to occur between two felony convictions. Had his first deportation happened before 1995 or after 1998, his Guidelines range would have been at least five years lower.³⁶⁴ Second is the importance of supervision violations. Because they extend the clock on very old convictions, large swings in sentence length can turn on precisely when a probation or parole violation occurred. And because the Guidelines count violations as part of the original conviction, two completely concurrent sentences (one for a new crime and one for a supervision violation) are treated as separate sentences of the same length. Third is how crucial the Guidelines make difficult-to-find state records. Mr. R’s defense lawyer did not know about the parole violations that, according to the probation officer, made the 1995 convictions count against Mr. R in 2018. And neither the lawyer nor the probation officer found the document in Mr. R’s prison file that proved the convictions were, in fact, too old. When large sentence swings turn on the precise details of state cases, defendants are at the mercy of state agencies’ variable recordkeeping practices.

Mr. R’s case is exceptional in some respects, including that his sentence was unusually long for a reentry case.³⁶⁵ But the problems it illustrates are endemic to the reentry guideline. They emerge in one form or another in thousands of cases every year. And his case also follows a familiar pattern for reentry prosecutions. Mr. R spent nearly his entire life in the United States. It is where his family lives and it is the only country he considers home. But he also accumulated a serious criminal record in his late teens and early twenties. Now, as a man in his forties whose entire life is in the United States, he keeps trying to return.³⁶⁶ And, because of the reentry guidelines’ double enhancements, those convictions from his youth keep triggering lengthy sentences every time he comes back.

IV. Sentencing Constitutionalism

This Article has shown that the federal sentencing system discriminates against undocumented immigrants. So what can be done about it? Naturally, Congress could amend the First Step Act so that immigrants receive earned time credit. The Sentencing Commission could also rewrite the reentry guideline to remove its extra criminal history enhancements. Such legislative fixes would be straightforward. This Part will thus focus instead on what can be done by federal judges and the lawyers who appear before them.³⁶⁷ A constitutional Equal Protection challenge seems like an intuitive strategy.³⁶⁸ But that approach faces significant obstacles under current case law. The Supreme Court has disallowed racial disparate impact claims, and it grants the federal government substantial leeway to discriminate by immigration status.³⁶⁹ Such doctrinal hurdles may be surmountable, but advocates challenging these sentencing policies have yet to overcome them.

This Part proposes another approach, herein labeled “sentencing constitutionalism.” The basic idea is that sentencing decisions are less constrained than other judicial decisions. In the sentencing context, judges exercise normative discretion over the size of the punishment and the reasons that justify it. And judges can include amongst those reasons a desire to advance underenforced constitutional norms.³⁷⁰ Because so much deference is built into modern Equal Protection doctrine, many instances of clear discrimination go unremedied in formal constitutional review. Recognizing this, several scholars have developed a theory of “administrative constitutionalism.”³⁷¹ These scholars have observed that executive branch agencies use their policymaking discretion to combat forms of discrimination, such as racial disparate impact, that cannot be remedied in court. Judges should do the same in the immigrant sentencing context. Specifically, judges should counteract the racial and immigration-status-based discrimination built into federal sentencing law by (1) refusing to apply the reentry guideline’s recidivist enhancements, and (2) discounting immigrants’ sentences to account for the fact that they cannot earn First Step Act credits.³⁷² Many federal judges have already embraced this kind of approach in the crack cocaine sentencing context. The crack/powder disparity in federal sentencing law is racially discriminatory, and Equal Protection challenges have entirely failed to remedy it. However, federal judges have somewhat counteracted this discrimination in their sentencing decisions. Numerous judges have explicitly rejected the higher crack guideline as a matter of policy. And overall, federal judges vary downward from the Guidelines range in most crack cocaine cases. This systematic judicial rejection of the crack guideline is motivated by constitutional equality concerns. Judges should do the same with reentry sentences.

A. A Two-Track Model of Equal Protection at Sentencing

Immigrant defendants can challenge the sentencing discrimination described above as violating Equal Protection. This argument can be made as a constitutional challenge to the validity of the laws themselves. It can also be made as a quasi-constitutional appeal to judges’ discretionary sentencing power. The former strategy runs into major obstacles in the form of deference principles established by the Supreme Court. The latter strategy lacks the compulsory force of law, but avoids the limitations of contemporary Equal Protection doctrine. Here both strategies will be explored in turn.

Both the First Step Act and the reentry guideline explicitly discriminate against defendants based on their immigration status.³⁷³ Numerous defendants have argued that these laws are therefore unconstitutional.³⁷⁴ To date, such arguments have not found success. Defendants have been unable to overcome the Supreme Court’s deference to the federal government when it comes to immigration policy. Under the plenary power doctrine, the Court gives Congress and the President broad leeway to enact policies that discriminate by immigration status.³⁷⁵ The Court has directed greater scrutiny at state laws that harm immigrants, but that is not helpful in the federal sentencing context.³⁷⁶ Perhaps more usefully, the Court in Hampton v. Wong did apply heightened scrutiny to federal agency regulations that discriminate against immigrants.³⁷⁷ Wong involved a Civil Service Commission regulation excluding all noncitizens from federal employment.³⁷⁸ The Court applied heightened scrutiny to this regulation, reasoning that the plenary power doctrine does not protect agency rules unless they are mandated by Congress or the President.³⁷⁹ The Court held that when an agency discriminates by immigration status, it must demonstrate both an “overriding national interest” and evidence that the policy was “actually intended to serve” that interest.³⁸⁰ Because the Guidelines are written by an independent agency (the Sentencing Commission), Wong may support a constitutional challenge to the reentry guideline.³⁸¹ The post-deportation enhancement seems especially vulnerable, since it (1) is not connected to Congress’s statutory maximum increase, which applies only to pre-deportation felonies, and (2) does not meaningfully deter reentry because it punishes only post-reentry conduct. Nonetheless, Wong-based challenges to the reentry guideline have yet to find success in court.³⁸²

Defendants could also argue that the reentry guideline and the First Step Act discriminate by race and ethnicity. Nearly every person harmed by these rules is of Latin American ancestry.³⁸³ And policies that discriminate by race are subjected to strict scrutiny.³⁸⁴ The problem, however, is that these rules are facially neutral as to race, and the Supreme Court has foreclosed Equal Protection challenges based on disparate impact claims.³⁸⁵ A law could thus apply against only one race ninety-nine percent of the time and still be upheld in court because it is facially neutral. The Court has carved out one important exception to this rule: under the Arlington Heights test, a formally race-neutral law violates Equal Protection if it was enacted with racist purposes and has a racially disparate impact.³⁸⁶ A defendant could try to bring such a challenge to the First Step Act or the reentry guideline. In the guideline context, this would likely mean emphasizing the racist panic in Florida that gave rise to the double criminal history enhancements. In the First Step Act context, it would mean focusing on the Trump Administration’s 2018 nativism. However, Arlington Heights challenges face two significant hurdles. First is the problem of proving intent—federal courts adopt a strong presumption that government action is not racially motivated.³⁸⁷ Second is the problem of reenactment—if a rule is enacted for discriminatory reasons at Time One, and then reenacted for different reasons at Time Two, courts often treat Time Two as decisive. For example, many defendants have argued that the unlawful reentry statute is itself unconstitutional due to its racist history.³⁸⁸ While federal courts have largely conceded that the law’s original enactment was racist, they have unanimously upheld the law because it was later reenacted.³⁸⁹

Modern Equal Protection doctrine thus erects significant barriers to immigrants’ antidiscrimination claims. While these barriers may be surmountable in some cases, they significantly limit the potential of constitutional challenges against the First Step Act and the reentry guideline. Another option would be to appeal to district judges’ sentencing discretion. After Booker, sentencing judges do not have to follow the Guidelines.³⁹⁰ They can sentence a defendant anywhere within the statutory range, subject only to deferential appellate review for reasonableness.³⁹¹ Because the judge is the key decisionmaker at sentencing, rather than Congress or the executive, the deference doctrines discussed above lack purchase. A judge can use their discretion to counteract race-based and immigration-status-based discrimination built into the federal sentencing rules. And judges should do so to vindicate a constitutional equality norm that, because of those deference doctrines, goes widely underenforced.³⁹² Constitutional norms can thereby have wider play in the discretionary sentencing context than they do in the judicial review context. This idea will here be called “sentencing constitutionalism.”

Administrative constitutionalism is a close analogue to sentencing constitutionalism. In numerous contexts, scholars have observed administrative agencies using their policymaking discretion to advance constitutional norms beyond the level required by courts. For example, in the 1960s both the Federal Communications Commission and the Federal Power Commission interpreted the Equal Protection Clause as requiring regulated companies to achieve racial equality in employment.³⁹³ In the 1940s and 1950s, the Federal Social Security Board enforced an Equal Protection norm safeguarding the welfare rights of the poor.³⁹⁴ During the Obama Administration, the Department of Housing and Urban Development adopted rules requiring recipients of federal funding to affirmatively reduce racial housing disparities.³⁹⁵ And, in the criminal justice context, contemporary federal (and some state) prosecutors have adopted a wide array of self-limiting rules that protect defendants’ constitutional rights beyond what courts require.³⁹⁶ In these examples and others, agencies go above the floor set by judicial doctrine and implement a more robust vision of constitutional rights. They do so through their own discretionary rulemaking authority. Judges, when they sentence defendants, operate in a similar context. Deference doctrines like plenary power over immigration, the prohibition on disparate impact claims, and others lack purchase because judges are themselves determining sentences. So long as they do not violate the law, judges can go above the constitutional floor just as agencies have.³⁹⁷ Policymaking discretion thus lets them implement constitutional antidiscrimination norms to a greater degree than formal doctrine requires.

Sentencing constitutionalism is also similar to the concept of an imperfect defense.³⁹⁸ In criminal trials, defenses are binary. The judge either will or will not let the defendant argue a defense, and the jury either will or will not acquit based on a defense. Because a successful defense exonerates completely, the formal elements of defenses are often written quite narrowly. The federal duress defense, for example, only applies where the defendant faces an “immediate threat of death or serious bodily injury.”³⁹⁹ The most common version of the insanity defense is limited to situations where the defendant either (1) did not know what they were doing or (2) was incapable of understanding that their action was wrongful.⁴⁰⁰ And California’s entrapment defense only applies where the police conduct would cause a “normally law-abiding person” to commit the crime.⁴⁰¹ Because these and other defenses are defined narrowly, they fail to exonerate the defendant in many cases where it seems like they should apply. Take, for example, someone who committed a crime on the orders of their abusive partner, but was not in immediate peril at the time of the crime. That person will not satisfy the elements of a duress defense, even though the moral purpose underlying the defense clearly applies.⁴⁰² In cases like these, many U.S. jurisdictions lower the defendant’s sentence to account for the “imperfect” defense.⁴⁰³ The Federal Sentencing Guidelines, for example, enumerate imperfect duress as a reason for downward departure.⁴⁰⁴ Several federal circuits have also recognized imperfect entrapment as a legitimate basis for downward departure.⁴⁰⁵ And numerous states’ sentencing statutes specifically list imperfect defenses as mitigating factors.⁴⁰⁶ Such imperfect defenses are structurally similar to sentencing constitutionalism. Both involve situations where a defendant’s legal argument has strong moral force (e.g., a duress defense or discrimination claim). However, the argument fails in court for technical reasons (e.g. the defense is too narrowly defined, or the discrimination claim is denied out of deference and administrability concerns). In such circumstances, judges can use their sentencing discretion

to compensate for the narrowness of formal law. They can lower a defendant’s sentence to account for the injustice that otherwise goes unremedied.⁴⁰⁷

Sentencing constitutionalism also finds conceptual support in Justice Anthony Kennedy’s concurrence in Beckles v. United States.⁴⁰⁸ In Beckles, the Supreme Court held that defendants cannot bring constitutional vagueness challenges to the Sentencing Guidelines.⁴⁰⁹ The majority reasoned that because the Guidelines are purely advisory post-Booker, they do not raise the same due process concerns as a vague criminal statute.⁴¹⁰ Justice Kennedy wrote separately to note that constitutional vagueness challenges should still have a place in discretionary sentencing decisions. However, he posited that “the realm of judicial discretion in sentencing” required “some other explication of the constitutional limitations.”⁴¹¹ He thus suggested that the Constitution should be operationalized differently at sentencing. That is also the intuition behind sentencing constitutionalism. Indeed, Kennedy’s reasoning could cover a variety of criminal procedure rights that do not normally apply at sentencing.⁴¹² The right to confront witnesses, the prohibition on double jeopardy, the right not to be punished for acquitted conduct, the privilege against self-incrimination, and other constitutional principles can be incorporated into the discretionary sentencing context (perhaps in a modified form with more judicial flexibility).⁴¹³ This would particularly make sense in the federal system, where trials have all but disappeared and there is far more litigation at sentencing than at the guilt-or-innocence phase.⁴¹⁴ Thus sentencing constitutionalism, in addition to providing an alternative remedy for Equal Protection claims, might also bring some of the Constitution’s criminal procedure architecture into the sentencing process.

B. Sentencing Constitutionalism and The Crack/Powder Disparity

The history of federal crack cocaine sentencing provides a real-world case study of sentencing constitutionalism. The story begins in 1986, with the tragic death of a basketball player named Len Bias. Bias died after using cocaine on the night he was drafted into the NBA.⁴¹⁵ The public reaction to his death launched a nationwide media-driven moral panic over crack cocaine.⁴¹⁶ This panic had a clear racial dimension, as crack cocaine abuse was widely perceived to be an epidemic amongst African-American individuals.⁴¹⁷ Politicians and the wider public feared that crack use was spreading from the inner cities to suburbia.⁴¹⁸ For example, Senator Lawton Chiles inserted a Florida newspaper article into the Congressional Record that reported: “Less than a block from where unsuspecting White retirees play tennis, bands of young Black men push their rocks on passing motorists, interested or not.”⁴¹⁹

Responding to this panic, Congress enacted the Anti-Drug Abuse Act of 1986.⁴²⁰ Senator Chiles was one of the “prime movers” behind the law.⁴²¹ It created new mandatory minimum penalties for federal drug trafficking crimes, which were triggered by a much lower quantity of crack than of powder cocaine. The law provided a ten-year mandatory minimum sentence for crimes involving five kilograms of powder cocaine, but just fifty grams of crack.⁴²² Similarly, it provided a five-year mandatory minimum for 500 grams of powder but just five grams of crack.⁴²³ This 100-to-1 ratio was not the product of any systematic study of the drugs’ relative effects—Congress’s factfinding process was perfunctory.⁴²⁴ Rather, legislators engaged in a one-way auction, ratcheting up the crack cocaine penalties in a competition to show which party could look tougher.⁴²⁵

The higher penalties for crack had a clear disparate impact on African American defendants. In 1993, 88.3% of federal crack cocaine defendants were Black and only 4.1% were White.⁴²⁶ By comparison, powder cocaine defendants were 27.4% Black and 32% White.⁴²⁷ The average federal drug sentence for a Black defendant went from 11% higher than that of a White defendant in 1986 to 49% higher four years later.⁴²⁸ Black defendants sentenced under the new law brought Equal Protection challenges, arguing that this crack/powder disparity was racially discriminatory. Such Equal Protection claims succeeded, albeit temporarily, in only two cases. Judge Clyde Cahill of the Eastern District of Missouri struck down the crack/powder disparity on Equal Protection grounds, finding that the 1986 Congress was motivated by unconscious anti-Black racism.⁴²⁹ Judge Cahill’s ruling was then reversed by the Eighth Circuit.⁴³⁰ The Minnesota Supreme Court also struck down a state statute creating higher penalties for crack than powder cocaine, concluding that it violated the state constitution.⁴³¹ The Minnesota legislature responded by increasing the penalties for powder cocaine to match those of crack.⁴³² Beyond those two outliers, such Equal Protection challenges failed. The federal courts of appeal unanimously upheld the disproportionate punishment of crack defendants.⁴³³ As the Fourth Circuit explained: “[T]here is evidence that the line Congress and the Sentencing Commission have drawn has a disproportionate impact upon blacks. But this is not sufficient to make out an Equal Protection violation.”⁴³⁴

When the Sentencing Commission published the first Guidelines Manual in 1988, it based drug trafficking sentences on the mandatory minimums set by Congress.⁴³⁵ The centerpiece of the drug trafficking guideline is a “Drug Quantity Table” that specifies the defendant’s base offense level.⁴³⁶ This Table gives the defendant an offense level between six and thirty-six depending on the type and amount of drug involved in the crime. Following Congress, the Commission set the amount of crack cocaine corresponding to each level on the Table at 1/100th the amount of powder cocaine at that same level. So, for example, a defendant would have a base offense level of 28 for 2.0–2.9 grams of crack cocaine or 200–299 grams of powder cocaine.⁴³⁷ This gave defendants with crack much higher sentences than defendants with an equivalent amount of powder. Appeals courts rejected defendants’ Equal Protection challenges to the Guidelines, using the same reasoning that they applied to the underlying statute.⁴³⁸ In 1995, the Sentencing Commission attempted to change the drug guideline in order to reduce the gap between crack and powder cocaine sentences.⁴³⁹ Congress, however, exercised its veto power over Guidelines amendments to keep the disparity in place.⁴⁴⁰

Then, in the 2005 case United States v. Booker, the Supreme Court made the Guidelines advisory.⁴⁴¹ This meant district judges could now choose to disregard the Guidelines and sentence defendants anywhere within the relevant statutory range. Federal judges soon took advantage of this new discretion to mitigate the crack/powder disparity. Citing racial discrimination concerns, numerous judges began to systematically sentence defendants below the Guidelines range in crack cocaine cases.⁴⁴² One such judge noted that “[t]his Court’s conclusion that a non-Guideline sentence is called for is . . . supported by the vast majority of district courts that have evaluated the crack/powder cocaine sentencing disparity in the wake of Booker.”⁴⁴³ The Supreme Court then confirmed judges’ power to categorically reject the crack cocaine guidelines in two decisions: Kimbrough v. United States and Spears v. United States.⁴⁴⁴ The Court also singled out the crack guideline for criticism in Kimbrough, noting that it was based solely on the mandatory minimums in the 1986 law, not on any empirical study of drug crimes or drug sentencing.⁴⁴⁵ Since then, judges have increasingly used their discretion to lower crack cocaine sentences. The Sentencing Commission’s statistics show that the rate of below-Guideline sentences in crack cases has consistently been the highest of all major drug types.⁴⁴⁶ The rate of downward variances has also steadily risen since Booker, and today more than half of all federal crack cocaine sentences are below the relevant Guidelines range.⁴⁴⁷

This history is a real-world example of sentencing constitutionalism. The crack/powder sentencing disparity is clearly racially discriminatory. It was produced during a racialized moral panic over Black drug use, and it disproportionately burdens Black defendants. Due to the hollowed-out nature of Equal Protection doctrine, defendants were unable to convince judges to declare the crack/powder disparity unconstitutional. However, some judges have exercised their sentencing discretion to lessen the constitutional harm.

Judges should do the same thing in unlawful reentry cases. Indeed, there are deep parallels between crack cocaine sentencing and unlawful reentry sentencing. Both involve formally neutral guideline provisions that produce overwhelming disparate impacts according to race and ethnicity. Neither guideline provision was the product of any systematic study by the Commission. Both were essentially copied from recently enacted statutes that increased the punishment ranges. In both cases, the underlying statute was enacted during a racialized moral panic in the 1980s. Both statutes were championed by the same Senator (Lawton Chiles), indeed both were even enacted through bills with the title “Anti-Drug Abuse Act.” And constitutional Equal Protection challenges to both statutes have failed in the federal courts. The major difference is that judges have systematically used their discretion to mitigate sentencing discrimination in crack cocaine cases. They have not yet done so in unlawful reentry cases.

Conclusion

This Article has shown that federal sentencing laws explicitly discriminate against undocumented immigrants. The Sentencing Guidelines give them duplicative recidivism enhancements that do not apply to U.S. citizens. The First Step Act denies them sentence reductions that U.S. citizens receive. This discrimination contributes to a racial and nationality-based hierarchy in the federal criminal justice system. Undocumented immigrants, nearly all of them Latin American, prosecuted for entering the United States, are treated as a legal underclass. They are punished more severely for past crimes and granted less mercy for current rehabilitation. Fortunately, federal district judges can help counteract this discrimination. Sentencing judges have the power to disregard the Guidelines’ duplicative enhancements. They also have the power to correct for immigrants’ exclusion from in-custody sentence reduction programs. For nearly twenty years, judges have mitigated a similar injustice in crack cocaine cases by systematically lowering sentences. They have done so out of recognition that the sentences for crack cocaine are racially discriminatory, and that this discrimination has gone unremedied in ordinary litigation. Judges should do the same for undocumented immigrants.

Appendix

99 S. Cal. L. Rev. 291

Download

Eric S. Fish*

*Professor of Law, University of California at Davis. For their invaluable help the author would like to thank Jessie Agatstein, Elizabeth Arford, Amber Baylor, Marcus Bourassa, Jack Chin, Haiyun Damon-Feng, Ingrid Eagly, Sheldon Evans, Jailene Gutierrez, Deborah Kang, Doug Keller, Litzy Martinez Rodriguez, Emmanuel Mauleon, Sean McGuire, Jacob Schuman, Rory Van Loo, Charles Weisselberg, Rebecca Wexler, and the participants in Crimfest, Southwest Crim, the Yale Law PhD program colloquium, and faculty workshops at the University of California Irvine and the University of California Berkeley. Thank you also to the editors of the Southern California Law Review for their excellent editorial work.