Prosecutors as punishers: A case study of Trump-era practices

Abstract

Recent punishment and society scholarship has addressed the limits of policy reforms aimed at reducing mass incarceration in the U.S. This work has focused in particular on the political dimensions of penal legal reform and policy-making, and the compromises and shortcomings in those processes. Nearly absent in this scholarship, however, has been empirical and theoretical engagement with the role of front-line prosecutors as facilitators and/or resistors to downsizing efforts. Using the case of the U.S. federal criminal legal system's modest efforts to decrease the system's racially disparate and punitive outcomes, this paper elucidates the fragile nature of such reforms by delineating the critical role that front-line prosecutors play in maintaining punitive approaches. Focusing specifically on federal prosecutorial policy and practices in the Trump era, I draw on a subset of data from an interdisciplinary, multi-methodological project set in distinct federal court jurisdictions in the U.S. to examine how front-line prosecutors were able to quickly reverse course on reform through the use of their uniquely powerful charging and plea-bargaining tools. My findings illustrate how federal prosecutors pursued more low-level defendants, and utilized statutory “hammers,” including mandatory minimums and mandatory enhancements to ensure harsh punishments in a swift return to a war-on-crime.

Introduction

Recent punishment and society scholarship has addressed the limits of policy reforms aimed at reducing mass incarceration in the U.S. This work has focused in particular on the political dimensions of penal legal reform and policy-making, and the compromises and shortcomings in those processes (e.g., Aviram, 2015; Beckett et al., 2016, 2018; Campbell et al., 2020; Cate, 2016; Gottschalk, 2016; Seeds, 2017; Whitlock and Heitzeg, 2021). A nascent line of research has also explored how reforms play out on the ground (Lerman and Mooney, 2022; Martin, 2016; Verma, 2015), including some of the downsides of reform for those subject to punishment (Cate, 2022) and the illusory impact of reform on the use of imprisonment (Beckett et al., 2018; Beckett and Beach, 2021). Prosecutors, as political actors, have figured into this narrative, as they have largely lobbied to retain discretionary power (Degenshein, 2023), or otherwise thwart reform efforts (Hopwood, 2020; Lynch, 2016). Nearly absent in punishment scholarship, however, has been empirical and theoretical engagement with the role of front-line prosecutors as facilitators and/or resistors to penal downsizing efforts.

Using the case of the U.S. federal criminal legal system's modest efforts to decrease the system's racially disparate and punitive outcomes, this paper elucidates the fragile nature of such reforms by illustrating the critical role that front-line prosecutors play in maintaining punitive approaches. Focusing specifically on federal prosecutorial policy and practices in the Trump era, I use a subset of data collected from a multi-methodological project set in four distinct U.S. federal court jurisdictions to examine how front-line prosecutors were able to quickly reverse course on reform by wielding their uniquely powerful charging and plea-bargaining tools.

Prosecutors, punishment, and politics

The traditional conceptualization of the criminal prosecutor, at least in adversarial systems, is as a law enforcer, advocating on behalf of the state for conviction. While prosecutors also advocate at sentencing, the ultimate determination of penal sanction has long been understood to be the judge's purview, even if influenced by the courtroom workgroup (Schulhofer, 1980). However, since the late twentieth century, prosecutors have increasingly encroached on the judicial role, wielding significant power via charging and plea-bargaining tactics to lock in sanction outcomes prior to formal sentencing. Key to their power was the transformation of criminal statutes and sentencing structures in the 1970s through the 1990s, which increasingly baked in presumptive or mandatory sentence outcomes upon conviction (Alschuler, 1978; Davis, 2007; Lynch, 2016; Seeds, 2017). By virtue of charges filed, and charges agreed to in plea bargains, prosecutors working in such systems usurp much of the traditional role of judges to determine appropriate sanction, thereby becoming “first-look sentencers” (Berman, 2010: 430). This is especially the case with mandatory minimum laws, which are potent “statutory hammers” that ensure a specified incarceration sanction below which a judge may not sentence (Lynch, 2016).

The expansion of the prosecutor's role was partly due to the concurrent rise in political power of prosecutors, both as lobbyists (Campbell, 2012) and as elected officials, such as attorneys general who increasingly shaped state and federal criminal justice policy (Simon, 2007). Prosecutor groups lobbied for and facilitated legislative efforts in jurisdictions across the U.S. that increased punitiveness of criminal statutes; increased prosecutorial discretionary power; and decreased sentencing discretion of judges (Barkow, 2019; Campbell, 2012; 2014; Simon, 2007). In a symbiotic process, prosecutors wielded their growing political power in ways that ensured they amassed more punitive power at the front lines of the criminal legal system, which was then put to use as intended. Consequently, prosecutors were a key driver of mass incarceration in the U.S. (Barkow, 2019; Lynch, 2016; Pfaff, 2017).

A confluence of political, legal, and economic forces in the early 2000s gave rise to at least a mild reckoning with mass incarceration in the U.S. (Aviram, 2015; Beckett, 2018; Dagan and Teles, 2016; Simon, 2014). For the most part, this was realized in statutory changes whose effects on incarceration ranged from measurable (Campbell et al., 2020), as in California where the U.S. Supreme Court had mandated the state to reduce its prison population (Simon, 2014), to negligible (Beckett and Beach, 2021; Beckett et al., 2018). Many of the reforms only addressed sanctions for nonviolent crimes, such as drug offenses (Beckett et al., 2016; Gottschalk, 2016), and sometimes lawmakers even traded passing those reforms with increasing punitiveness for offenses deemed serious and violent (Seeds, 2017), thus limiting their overall impact on the use of incarceration as punishment.

As part of the reform movement, the public also began to elect some “progressive prosecutors” into office, although primarily in urban, Democratic jurisdictions (Romero, 2020). Jurisdictions that include cities such as Philadelphia, Los Angeles, San Francisco, Chicago, Houston, Baltimore, Boston, and Brooklyn have elected reform-minded prosecutors who campaigned on promises to dramatically change how their offices use their discretionary powers, especially regarding charging decisions. While progressive prosecutors have faced significant internal and external backlash (Butler, 2022; Davis, 2019), they have had some direct impact in mitigating punishment in their jurisdictions (Mitchell et al., 2022), and they have had a modest impact in lobbying for state legislation aimed at reducing punitiveness (Hessick et al., forthcoming). Nonetheless, as a political force, prosecutors have largely fought any reform efforts that would reduce prosecutorial discretion (Degenshein, 2023; Hopwood, 2020), and those that would lessen maximum punishments or eliminate mandatory minimum punishments (Hopwood, 2020; Lynch, 2016), while continuing to lobby for more funding, fewer restraints on their power, and more severe punishments in the current reform era (Hessick et al., forthcoming; Hopwood, 2020).

Frontline prosecutors as punishers

Even though most formally adopted statutory reforms in the U.S. promised only modest impact on punishment, they have nonetheless made even less of dent in mass incarceration than expected. Indeed, as Katherine Beckett and her colleagues have pointedly demonstrated, in the current era of reform, the use of incarceration has intensified when measured against actual crime incidence (Beckett and Beach, 2021; Beckett et al., 2018). A dramatic drop in U.S. crime rates has significantly outpaced the recent decline in incarceration rates, resulting in an annual net gain in the use of incarceration/crime incidence since 2000 (Beckett and Beach, 2021; Beckett et al., 2018). Punishment intensity increased most significantly in rural and suburban counties, and in jurisdictions within traditionally punitive states (Beckett and Beach, 2021).

To that end, Beckett and Beach's (2021) findings indicate that the interplay of specific elements of criminal statutes and local legal dynamics, including prosecutorial norms, function as a limit on the efficacy of formal policy reforms. In locales where the formal law has not substantially reined in penal capacity, statutory “hammers” can be used by zealous prosecutors to continue to fill prisons, whereas reforms that meaningfully limit punishment maximums can constrain even the most zealous prosecutors, suggesting to Beckett and Beach (2021: 24) that “formal sentencing policy and informal discretion are best understood as being intertwined than separate and distinct.” Given that in many jurisdictions, legislative reforms have left intact punitive elements of the criminal law, regional disparities have increased in the era of reform, and with a net result of increasing punishment intensity (Beckett and Beach, 2021).

Beckett and Beach's (2021) observation about the interplay of formal law and local context comports with other scholarship showing that the specific elements of criminal statutes function as the structured context in which frontline legal actors, such as prosecutors, construct the universe of possibility for enacting their roles, and serve as the tools they can wield in their day-to-day practices (Lynch, 2018; Ulmer, 2005). Those elements are also interpreted in accordance with the norms governing their local contexts—including local cultural “tool kits” (Swidler, 1986), organizational rules and constraints, and intergroup relations—which help define both social problems to be addressed and the organizational solutions to those problems. Consequently, legal policy is rarely translated into practice in a uniform and orthodox manner across diverse settings, as a line of now-classic scholarship on criminal courts has documented. For instance, local criminal court communities develop norms as to “going rates” for prototypical cases, what cases are triable, and the terms of negotiations in plea deals that both persist over time, and differ across place (e.g., Church, 1985; Dixon, 1995; Eisenstein et al., 1988). The specifics of these norms are conditioned and constrained by the structure and mandates of the larger jurisdiction's formal law (i.e., the penal code), but practices can and do vary considerably by intra-jurisdictional locale (Johnson et al., 2008).

Pat Carlen's (2002) concept of “carceral clawback,” is also instructive in understanding why reforms aimed at reducing punitive power struggle to make an impact. Applied to the case of reform in women's prisons, Carlen (2002: 116) defines carceral clawback as “the power of the prison constantly to deconstruct and successfully reconstruct the ideological conditions for its own existence.” In the context of prosecutors, we should not be surprised by the active resistance to, and reconstruction of, retrenchment efforts. Local criminal legal system actors, including prosecutors, and the organizational units in which they are embedded, are invested in their own legitimacy and power, so can be expected to both ideologically and operationally adapt to changing policies in order to maintain their size and stature (Lynch & Omori, 2014). They do so, though, in ways that respond to local norms, logics, and historical practices, so such efforts will not be uniform across locales (Lynch & Omori, 2014; Verma, 2015). Nor will the reforms be as substantively impactful as they might look on-the-books, given the strong organizational pressure to survive and thrive (Carlen, 2002).

Due to the nature of their role, frontline prosecutors are especially able to blunt the impact of punishment-reducing reforms. Their decision-making is highly insular and in the U.S. context, nearly unreviewable (Davis, 2007). They decide who, among the pool of suspected and arrested potential defendants, to charge and who not to charge; the content of the charges; and the terms of potential plea agreements (Barkow, 2019; Pfaff, 2017). Consequently, a significant body of research has pinpointed early-stage prosecutorial decision-making as an important driver of punitive and racially-unequal outcomes in the mass incarceration era at both the state and federal levels (Kutateladze, 2018; Metcalfe and Chiricos, 2018; Pfaff, 2017; Rehavi and Starr, 2014; Shermer and Johnson, 2010; Stemen and Escobar, 2018).

Federal prosecutors, who share concurrent jurisdiction with state prosecutors in many areas of criminal law, have added discretion in who to prosecute, rendering their case selection and charging power even more potent (Heller, 1997). Even in the area of immigration, where federal prosecutors have primary jurisdiction, they have wide discretion as to whether to criminally charge, whether to refer potential defendants for civil deportation proceedings, or whether do nothing (Chacón, 2012). They also have formidable statutory “hammers” at their disposal, in the form of mandatory minimums and mandatory sentence enhancements that must be imposed by judges upon conviction. These tools are also used as threats to obtain guilty pleas, and by virtue of their potency, push negotiated sentence outcomes higher than they would be otherwise (Lynch, 2016). It is the use of these two discretionary powers—case selection and the content of charges—that I examine in the present study.

The federal system as a case study: context, data & methods

Context & background

This study builds directly upon my previous research that examined jurisdictional differences in criminal case adjudication strategies in four distinct federal courts within the federal legal system, using both qualitative field methods and quantitative methods (Lynch, 2016). The original project was conducted during a period when several reforms in the federal system had recently been implemented or were ongoing. Regarding legislative reforms, in 2010, the U.S. Congress partially addressed the infamous crack-powder cocaine disparity in mandatory minimums by increasing the crack weight thresholds that triggered mandatory minimum penalties. Following that, beginning in 2013, a bipartisan effort to more broadly address harsh federal drug penalties ensued, eventually resulting in the passage of the First Step Act of 2018. Among other things, this law shortened drug recidivist and some gun mandatory enhancements, and expanded eligibility for exemption from drug mandatory minimums (U.S. Sentencing Commission, 2020). In the same time period, with the assent of Congress, the U.S. Sentencing Commission reduced the recommended guideline sentences for drug trafficking offenses in 2014 and for certain immigration defendants in 2016.

President Obama's first Attorney General, Eric Holder, also introduced several internal Department of Justice (DOJ) policy reforms that aimed to restrain frontline prosecutors in the 94 U.S. Attorneys’ offices from maximizing their penal power. First, in 2010, Holder issued a memorandum that reversed existing Bush-era policy requiring prosecutors to pursue the most serious provable charges (Holder, 2010). Three years later, Holder issued a memorandum specific to drug cases, directing prosecutors not to charge drug weight that would trigger a mandatory minimum prison sentence in cases involving low-level drug defendants who met qualifying criteria (Holder, 2013). The next year, Holder issued a memorandum specific to a particularly punitive drug mandatory enhancement applicable to drug defendants with prior drug convictions, prohibiting prosecutors from using the enhancement as a threat to induce guilty pleas, and limiting when it could be imposed (Holder, 2014). Among other things, up until the passage of the 2018 First Step Act, this enhancement, colloquially referred to as an “851,” doubled the mandatory minimum for people with one prior drug sales conviction, and mandated a life-without-parole sentence for those with two or more priors and who otherwise would face a 10-year mandatory minimum. In this same period, President Obama initiated a clemency program primarily targeting drug defendants that ultimately commuted the sentences of over 1700 prisoners. ¹ Consequently, the federal prison population began to decline, from a high of 217,815 prisoners at year-end 2012 to 189,192 by year-end 2016 (Carson, 2020).

While the Trump administration could not directly undo the policy reforms enacted by Congress or put in place by the U.S. Sentencing Commission (and it even supported the First Step Act of 2018), under Attorney General Jeff Sessions, the Obama-era internal DOJ policies were swiftly rescinded and frontline prosecutors were encouraged to reclaim their role as punishers. Thus, in a May 2017 memorandum, Sessions directed federal prosecutors to charge and seek convictions on “the most serious, readily provable offense” (Sessions, 2017a), maximizing the potential punishment exposure in every case that federal prosecutors bring. The memo also rescinded the 2013 Holder policy that directed prosecutors not to seek mandatory minimums against specified low-level drug defendants, requiring instead that mandatory minimums be sought in all eligible cases, without regard for individual defendants’ relative culpability or mitigating circumstances. And the memo rescinded the 2014 Holder policy that barred prosecutors from using threats of 851 mandatory sentencing enhancements to compel guilty pleas from defendants (Sessions, 2017a).

Sessions also directed prosecutors to reprioritize “illegal immigration and violent crime, such as drug trafficking, gang violence and gun crimes” (Sessions, 2017b), resulting in a deemphasis on white collar crimes, a traditional priority in the federal system (Hurtado et al., 2020). His rhetoric in doing so was racialized, explicitly linking unauthorized immigration to drug and gang crime and implicitly racializing drug traffickers as cartel members and/or inner-city gangsters and violent thugs (Sessions, 2017b). Quantitative analysis of national sentencing data that examined how these DOJ policy changes impacted federal drug case outcomes indicates that the use of drug mandatory minimums significantly declined under the Holder policy, then shot back up after the Sessions policy mandates (Lynch et al., 2021). Both sentence lengths and percentage of drug defendants sent to prison also significantly increased following the policy changes, wiping out the reform benefits of the Holder policies (Lynch et al., 2021). A key driver in the increase in punitiveness was Trump's installation of new U.S. Attorneys at the district level, suggesting that a localized, on-the-ground implementation process was key to the success of the new punitive policies (Lynch et al., 2021).

Despite the new punitiveness of the Trump-era DOJ, the federal prison population continued to decline, dropping to 175,116 by year-end 2019 (Carson, 2020). This was at least partly due to several factors, particularly related to those convicted of drug offenses, a primary driver of federal mass incarceration including the racial disproportionality that especially disadvantaged Black defendants (Lynch, 2016). First, the earlier drug policy changes such as the 2010 Fair Sentencing Act and the 2014 drug guideline reductions, ensured earlier release dates for drug defendants sentenced in the intervening years, thus lowering annual year-end counts in future years (Gotsch, 2019). Second, many of the Obama commutations of drug defendants were staggered so those granted were not released until various times during the Trump administration. Third, the 2018 First Step Act enabled a number of early releases from prison for those previously convicted on crack charges, among other early releases (Gotsch, 2019).

The shrinking year-end count also masked a countervailing trend. After steadily declining over Obama's second term after peaking in 2011, annual federal prison admissions began to creep up during the Trump administration, growing from 44,682 in 2016 (Carson, 2018) to 46,051 in 2019 (Carson, 2020). Given that the number of people sentenced for non-petty immigration convictions increased by nearly 50% during the Trump administration (Lynch, 2020), the year-end counts would not be expected to explode given the relatively shorter prison sentences imposed in immigration cases.

Study design & data

The present study is a follow-up to my original study of the federal system, which was conducted between 2012 and 2014 (Lynch, 2016). In the current study, I used both quantitative and qualitative data to assess how criminal legal practices had changed under the Trump administration. I focused on the same four research sites as in the original study to see how local legal actors responded to the new law-and-order regime that reversed the Obama-era reforms, and that reprioritized aggressive federal law enforcement in the areas of immigration, drugs, and street crime. The four districts differ from each other on several key dimensions including size, geographic location, population demographics, caseload characteristics, and sentence outcomes.

I collected and analyzed multiple data sources. First, I made field observations in the four districts’ courts during periodic visits over a period from late 2017 to late 2019. In 2018 and 2019, I also interviewed 35 legal actors across the four districts, after the Trump-appointed U.S. Attorney had been appointed in each district. My primary interviewees were federal defenders, supplemented by interviews with federal prosecutors and judges. ² The interviews aimed to assess current practices, including how things had changed from 2014 (when I had last conducted observations and interviews in each district) to the time of the current interview. I had previously interviewed 27 of these legal actors in the original study, so I was able to get their perspectives about what specifically had changed in their experience. In addition, I analyzed US Sentencing Commission data from FY2007 through FY 2019, both specific to the four districts and nationally, to assess longer-term trends in criminal case features and outcomes. Finally, I sampled actual case files from the four districts to assess whether and how the DOJ policy reforms impacted caseload characteristics and adjudication practices. The selected cases were initiated in the last three years of the Obama administration (2014–2016) and the first three years of the Trump administration (2017–2019). Twenty cases per year in eight primary offense categories (targeting drug, gun, immigration, and child pornography crimes) were randomly drawn from each district and downloaded from PACER, the federal courts database. In some categories and in some years, a given district had fewer than 20 cases.

In this paper, I report on findings from two of the districts, and limit my analysis to immigration and drug trafficking cases, ³ both prioritized by the Trump DOJ, and where countervailing reforms threatened the punitive power held by federal prosecutors. The first district, Northeastern, ⁴ is in a politically liberal state in the northeastern part of the United States. It is composed of both urban and rural communities, and has had among the highest rates of judge-initiated, below-guideline sentences since the federal sentencing guidelines became advisory in 2005. Federal prosecutors in this district work in a local court environment that more fully embraced reform efforts aimed at reducing the reliance on incarceration (Lynch, 2016). The second district, Southeast, is in the U.S. South and includes urban, suburban, and rural communities. Its sentencing patterns have been consistently punitive over time, regardless of policy mandates. Unlike in Northeastern, judges here largely embrace the sentencing guidelines, and prosecutors have fully exercised their punitive power with little pushback, including using threats and actual imposition of mandatory statutory “hammers” (Lynch, 2016). Northeastern has always had a significantly smaller annual caseload size than Southeast, even though the two jurisdictions have a similar general population size. The two selected districts offer an excellent contrast in legal cultures and historical practices, while not so different in structural conditions to render them incomparable (Lynch, 2016). ⁵

In this analysis, I draw on my field notes from observations in the respective district courts, the 20 legal-actor interviews conducted in the two districts, court file data from the sampled criminal case files in four of the eight sampled offense categories (felony immigration, aggravated identity theft, drug trafficking, and the “851” drug enhancement), as well as the U.S. Sentencing Commission data specific to these districts. I specifically assess how charging changed under the Trump administration, including the priorities in case type and the characteristics of defendants, and examine how statutory “hammers,” including mandatory minimums and mandatory enhancements were being used at charging and in plea negotiations.

Findings

Immigration prosecutions

Neither Northeastern nor Southeast has traditionally had a large immigration caseload, as is typical along the Southwest border and in districts with major international cities. Yet both districts experienced an uptick in immigration cases in the Trump era. In Southeast, the share of the total criminal caseload devoted to non-petty immigration prosecutions stood at about 13% in 2014 and grew to 19% in 2019. In Northeastern, the growth in share of caseload over the same period was more modest, from about 11% in 2014 to 14% in 2019. But more critically, and not reflected in these statistics, was the nature of the defendants being charged with felony immigration violations.

In Northeastern, the immigration caseload was where the impact of the Sessions policies was perceived as most dramatic. It was the first change mentioned by most of the defense attorneys I interviewed in the district. Several related that for the first time in their federal careers, they were getting cases of people with no criminal record, but just prior deportations, who were being criminally charged with felony illegal re-entry. The overwhelming majority of these defendants were from Central America, reflecting the broader trend of targeting non-White immigrants from the Global South. As one federal defender shared, “we’re seeing a lot more illegal re-entry cases, a lot more immigration prosecutions. We’re seeing them charge multiple crimes. We’re seeing people who don't have any record get picked up on the way to school to drop their kids off. Or, on their way to work in the morning… I know other districts, they use the misdemeanor alternative a lot. We never get that, that's never on the table here. It's always a felony.”

Several interviewees indicated that previously, prosecutors generally declined to prosecute where the detained person did not have a notable criminal record; the detainee would instead proceed through the civil deportation process or simply be released. The U.S. Sentencing Commission data confirm there was an uptick in prosecuting defendants with minimal or no prior criminal records in Northeastern. The share of sentenced immigration defendants who were in the lowest criminal history category ⁶ declined to 13% in 2015, then rose to 33% in 2018. Three interviewees referred to these cases as the new “low-hanging fruit” in the district: cases that prosecutors could easily charge and obtain convictions to get their case numbers up (i.e., “statistics-driven policy,” in one defender's words). Another defender characterized the current situation as “just open season on all undocumented people.”

Prosecutors in this district also began utilizing a particularly punitive strategy against undocumented immigrants of filing an aggravated identity theft charge (18 U.S.C. § 1028A), in addition to predicate charges such as passport fraud or social security fraud, in cases where the defendant had used another person's name and social security number to get employment, obtain a passport or driver's license, or receive health benefits or related government services. The aggravated identity theft charge operates as a statutory enhancement, requiring that the judge impose a sentence of at least one day in prison on the predicate charge, and then impose a mandatory minimum sentence of 2 years, consecutive to the underlying sentence, for the aggravated identity theft.

The aggravated identity theft statute was enacted in 2004, in response to both rising digital financial crimes through the use of identity theft, and in the context of the war on terror. It specified a series of terror-related and nonterror-related underlying felonies that can serve as predicate, or underlying offenses, and was not conceived of as an immigration enforcement tool (Wixted, 2009). In 2017, however, reviving a practice used during the Bush administration, Attorney General Sessions (2017c) directed prosecutors to use this specific statutory hammer against undocumented immigrants who would be eligible to be charged with a predicate document fraud.

Prosecutors in Northeastern took this directive to heart, and used the 1028A charge both as a plea-bargaining tool to lock in sentences significantly higher than the guidelines for the document fraud, and in many cases, to ensure a sentence of at least 2 years and a day. U.S. Sentencing Commission data indicate that Northeastern had the second-highest share of 1028A sentences/overall caseload of all 94 federal districts in 2019, growing from 1% of the total sentenced caseload in 2016 to 7% in 2019. Multiple federal defenders raised concerns about this new practice in interviews. As one defender shared:

They’re charging minimum mandatories on cases they never used to. So, in a lot of illegal re-entry cases or misuse of social security cases, they had evidence to potentially charge aggravated identity theft, which carries a two-year mandatory minimum. And, they never did that really. Now, they’re doing it fairly routinely…[The US Attorney's Office] recently did a big publicity press release when they arrested 20 people connected with registry and motor vehicle documents. They charged all those people with aggravated identity theft. And these are people who basically have no record, who really weren’t committing any other crime outside of getting a form of identification. And, they’ve charged all of those with the minimum mandatory.

The US Sentencing Commission data and the sampled case files also reflect this change. While prosecutors in Northeastern were still pursuing more traditional aggravated identity theft defendants in the Trump period—those who steal identities to commit major financial fraud or other serious crimes—the quantity and quality of 1028A cases shifted. From 2014 until the end of 2016, a total of only 18 1028A cases were sentenced in the district, of whom only a minority (39%) were noncitizens, and whose pre dicate charges were typically major financial fraud. In the subsequent 3-year period, there were 58 cases sentenced that received the 1028A mandatory penalty, of whom 72% were noncitizens. The sampled cases from 2017 to 2019 indicated a majority of these defendants were convicted of simple document fraud (i.e., using another person's identity to get legal identification) as the predicate offense. These cases do not appear in the illegal reentry prosecution statistics, thus masking the degree to which prosecutors in Northeastern waged a war on undocumented immigrants in the Trump era.

The sampled cases also reveal how the 1028A charge was used to induce guilty pleas. Because the guideline-recommended sentences on the predicate charges were often in the range of 0–7 months, plea agreements would exchange dropping the 1028A charge for an agreed-upon sentencing recommendation of 12–15 months. In the Trump era, the plea agreements also began to include a requirement that undocumented defendants in such cases (and even in the regular illegal reentry cases) waive their right to a hearing on the civil side of immigration enforcement.

To illustrate how these cases proceeded in Northeastern, 58-year-old Umberto Hernandez-Padilla, an undocumented Central American, used a social security number of a Puerto Rican citizen to get work as a laborer when he had entered the district a decade prior to his arrest. Mr. Hernandez-Padilla had no criminal history or criminal involvement, and paid taxes and social security on his income in the intervening period. When he sustained a back injury at work that disabled him, he applied for social security benefits, which prompted his arrest. He was charged with passport fraud, social security fraud, aggravated identity theft (1028A), and theft of government funds related to healthcare benefits he had received via Medicaid. In order to get out from under the 2-year mandatory enhancement, the plea agreement required him to agree to request a sentence of 12 months from the judge, as well as enter into a “stipulated judicial order of removal” from the U.S. that additionally required the defendant to “assist ICE [Immigration and Customs Enforcement] in the execution of his removal.” Finally, he had to agree to forfeit and repay approximately $45,000 for the health benefits he received.

In Southeast, the majority of interviewees also noted that immigration had become a larger share of the caseload and described the defendants as more likely to be lower level. As the head federal defender in the district reported when asked about how things had changed:

The less serious [illegal reentry] cases, the person who has been here for 20 years, no criminal history but for some reason was deported and then came back, that person… their guidelines are going to be zero to six months or one to seven months… They’d stopped prosecuting basically towards the end of the Obama Administration. During probably the second year of the Trump Administration, they started picking those cases back up again. So, we went from 59 [illegal reentry] cases in fiscal year 2017 to 190 cases in fiscal year 2018.

He shared that the cases typically initiated in local jails, where an authorized officer identified undocumented arrestees (often picked up for DUI or driving without a license), called the U.S. Attorney's office, and asked if they would like the referral for prosecution. In his view, the previous practice would be for prosecutors to ask whether the arrestee had a serious criminal history. If the answer was no, they would typically pass; whereas in the new regime, “they’re like, ‘All right. We’ll take all of them.’” The U.S. Sentencing Commission data also indicate that more defendants in the lowest criminal history category were prosecuted in the Trump era relative to Obama's second term in Southeast district. The number of sentenced illegal reentry defendants in the lowest criminal history category dropped as low as 11% (in 2014) in the district, then climbed to 24% in 2019. On the other hand, a prosecutor I interviewed in the district insisted that their office had maintained a kind of “zero-tolerance” steady state in immigration prosecutions: “You’re here illegally, we prosecute them. Period.”

The other way noncitizens were targeted in this district was if they were in the country legally but without citizen status, and they picked up a criminal offense that could rescind their legal status. A defender shared a story about her client, a DACA recipient (so holding legal status) who was a military spouse with two children. She was arrested for a theft of some items from the military commissary, and prosecuted federally. The defender felt that under the previous administration, this case would likely have been handled in a way to ensure a plea agreement that would not jeopardize her client's ability to stay in the country. Under the new policies, the line prosecutor would not commit to doing so, even knowing the defendant will be deported if she is convicted of the charge he filed.

The new, hardline policies also changed the way immigration cases were resolved in this district. Unlike the previous norm, where almost all cases across the spectrum were resolved with plea agreements, defendants had started to plead guilty in illegal re-entry cases straight up without an agreement. In Southeast, some judges were declining to expedite immigration cases, as had become a standard practice in many jurisdictions, so defendants often spent many months in custody before sentencing. Consequently, the cases typically resolved with sentences of credit for time served, given the low guidelines for such cases. Moreover, prosecutors in Southeast also started to impose waivers of civil immigration rights in their plea offers. Without the plea agreement, those defendants could still contest their deportation. Ultimately, there was little to gain from a plea agreement and not much to lose. Unlike Northeastern, however, there was no evidence that prosecutors had adopted the practice of charging undocumented defendants with the 1028A aggravated identity theft mandatory minimum in Southeast.

Drug prosecutions

Both Northeastern and Southeast districts had a longstanding practice of proactively prosecuting drug trafficking cases, using sting operations and confidential informants to target parties of interest, and adopting cases that were being prosecuted in state court to ensure a more punitive sentence. These efforts disproportionately targeted Black and Latinx potential defendants. Drug cases have consistently composed about twice the share of the total caseload in Northeastern compared to Southeast; drug trafficking is generally the single largest category of cases in Northeastern. Notwithstanding Northeastern's aggressive pursuit of drug cases, prosecutors in the Holder era were more tempered in seeking maximum punishments, and were not as aggressive in how they used mandatory minimums and mandatory enhancements compared to prosecutors in Southeast district. In response to Holder's (2013) directive regarding the use of mandatory minimums, the number of Northeastern drug cases sentenced to a drug mandatory minimum more than halved by 2016, compared to the year prior to the policy change: Only 29% of all drug defendants faced a mandatory minimum in 2016, down from 59% facing a mandatory minimum in 2012. By 2019, the percentage of sentenced drug defendants in Northeastern facing a mandatory minimum sentence had climbed to 51%.

In Northeastern, the drug case types were bifurcated, a pattern that persisted in the Trump era. Prosecutors here pursued very small, retail dealers, who in most jurisdictions would have been left to the state system; and they prosecuted multi-defendant drug conspiracy cases. There was one prosecutor in the district who specialized in prosecuting the small cases involving hand-to-hand sales of tiny amounts of drugs that were often set up by law enforcement near schools, parks, or public housing complexes, to be able to obtain a short mandatory minimum due to that proximity. The defendants in these cases were nearly exclusively persons of color, and the predominant drug was crack cocaine. This prosecutor continued this practice, unabated from the Holder period into the Sessions period, and crack was still the predominant drug. The conspiracy cases, however, were increasingly dominated by heroin and fentanyl trafficking.

When asked about the current state of drug prosecutions, most interviewees mentioned increasing aggressiveness in drug cases, especially cases involving heroin and fentanyl. This was not surprising, as Sessions himself had publicly asserted that “gangsters” from a predominantly Black and Latinx region in this district were responsible for supplying the drugs that were killing people in a neighboring, overwhelmingly White district. One defender shared that, “we’re seeing more and more fentanyl cases, in terms of drug cases. But, more importantly, they’re charging every mandatory minimum that they can. And they’re filing 851s [the mandatory drug enhancement].” This same interviewee indicated that at the same time, prosecutors were less willing to go easy on sympathetic or lower-level drug defendants than they had been under the previous administration. The new expectation was that to get any break from the prosecutor, defendants had to “cooperate” with the government by informing others.

The sampled cases confirmed a resurgence in the filing of the “851” mandatory drug enhancements. In 2014 and 2015, a total of 6 defendants faced mandatory enhancements (4 in 2014 and 2 in 2015). A total of 14 defendants who were originally charged in 2016 also had 851s filed against them, but most of those enhancements (10 of the 14) were filed in the Trump era, after Sessions rescinded Holder's policy. In cases initiated from 2017 through 2019, a total of 58 defendants had mandatory drug enhancements filed against them; 30 of those were in 2018 alone. The majority of the defendants with “851s” filed in the Trump era (45/68; 66%) faced charges of distributing heroin, fentanyl, or both substances. Not all of these cases ended with mandatory enhancements imposed, however. In some cases, the mandatory enhancement was rescinded as part of a plea agreement. In others, the filing occurred in cases where the drug weight was insufficient to trigger the mandatory minimum, however the filing ensured the defendant would face much higher recommended guideline sentences under the “career offender” provision of the U.S. Sentencing Guidelines (see Lynch, 2016). And in some cases, where the defendant actually had two or more prior convictions so could potentially face a life without parole sentence, prosecutors only used one prior conviction against them, mitigating the punitive outcome.

The Sessions drug prosecution policies did not sit well with all of the line prosecutors in Northeastern. A federal defender related a story of a case where a sympathetic prosecutor helped the defender avoid both the mandatory minimum and the 851 enhancement, against the wishes of the prosecutor's new supervisors:

I had a guy who - in a seven defendant conspiracy case - was charged in 2016. So, during Obama. The indictment didn’t allege weight. It dragged out a long time because it was a wiretap case… And my guy was a career offender [facing very high sentencing guidelines]. She [the prosecutor] kept trying to get a plea agreement approved. And then, we didn’t get into a plea agreement until after the administration changed, and they [her supervisors] rejected everything she did. Finally, they said to her, ‘he has to plead to an information, where the drug weight's alleged, which would have been a five-year mandatory minimum, and you have to file an 851,’ which would have meant a minimum mandatory of ten years. I said to her, ‘that's ridiculous,’ she's like, ‘I know.’ So I’m like, ‘I’m going to call the judge's clerk and see if we can get a plea hearing for tomorrow.’ I said, ‘if I do that, are you going to tell anybody?’ She's like, ‘nope.’ So, we went in the next day, and he pled straight up without a plea agreement to the indictment that didn’t allege weight; she hadn’t filed an 851, and he just got sentenced the other day… But, if they had superseded and filed the 851, [the judge] would have had to give him ten. Instead, she gave him five and ten months. And, then [the prosecutor] got screamed at, and she transferred back to San Francisco.

The imposition of the aggressive policies in both drugs and immigration cases triggered a fair amount of turnover in Northeastern's U.S. Attorney's office, allowing the office to hire line prosecutors who were more ideologically-aligned with the Sessions regime. The office created designated hiring lines for immigration and “violent crime,” including drug cases, to ensure that the new administration's priorities were carried out.

Changes to drug prosecutions in Southeast were also evident, but this district had neither reformed as much as Northeastern under Holder's policies, nor was it a jurisdiction that had shied away from seeking and imposing harsh sentences in drug cases (Lynch, 2016). In Southeast, the relative decrease in the use of mandatory minimums under the Holder policy was smaller than it was in Northeastern: 74% of drug defendants were sentenced to a mandatory minimum in the year prior to the policy change; by 2016, that share was 57%. By 2019, the share of drug defendants sentenced to a mandatory minimum had grown to 84%, suggesting that Sessions’ directive to seek mandatory minimums in all eligible drug cases was taken seriously in Southeast. The high level of drug mandatory minimums in Southeast was notable given the changing nature of the drug caseload in the district. It had been one of the most aggressive districts in the nation in imposing the draconian crack cocaine mandatory minimums, primarily on Black defendants, up until those laws were changed in 2010 making it harder to cross the drug weight thresholds. Heroin and fentanyl had now replaced crack as the targeted drug, both of which require more weight to trigger mandatory minimums than even the revised crack thresholds.

My interviews, observations, and the case file data revealed the new contours of drug prosecutions in Southeast. There was general consensus among the Southeast interviewees that drug mandatory minimums and 851 mandatory enhancements were more frequently being applied to eligible defendants in the Trump era. In the two southern divisions within the district, where the drug war had been most virulent (Lynch, 2016), prosecutors had started to charge very small cases involving heroin, even single hand-to-hand sales, if any fentanyl was detected in the product. As one defender told me, “heroin is the flavor of the month now.” Even though opioid use is much more prevalent among Whites, drug defendants continued to be disproportionately Black in this district. Indeed, as one defender cynically suggested, these cases were being prosecuted federally “because white people are dying” in the opioid epidemic. She went on to say, “You have suburbanites, who are generally not people of color to be perfectly candid, getting addicted to pain meds. And yet, where they’re focusing their enforcement attention seems to be the consistent supply. Which is, you know, the same ‘neighborhoods,’ the impoverished communities.”

A sentencing I observed in Southeast district well-illustrated the racial divide in who was prosecuted federally and who was spared. The defendant, Leticia Franks, was a 26-year-old Black woman, and mother of two children, whose criminal history included only a couple of misdemeanors. She was a low-level dealer who got arrested with three men, including her live-in boyfriend for whom she dealt relatively small amounts of heroin and cocaine. Because she was charged in a conspiracy, she could be held responsible for the full weight of the drugs being dealt by her boyfriend, his partner, and their supplier. Therefore, her original charges, if sustained, would have triggered a 10-year mandatory minimum sentence. She ultimately agreed to plead guilty to charges that came with a 5-year mandatory minimum. She faced sentencing from one of toughest judges in the district, who had earned the informal nickname of “the Hammer,” so the prosecutor was already poised to get a sentence imposed above that minimum. After the plea, the prosecutor argued in his sentencing memorandum that the sentencing guideline range should be 235–293 months, alleging that two people experienced nonfatal overdoses after buying drugs from Ms. Franks. The federal defender contended that the applicable guideline range was 63–78 months, and contested the legal and factual basis for the allegation about the overdoses.

At the sentencing hearing, the prosecutor called three witnesses. The first was a White man who testified he bought .2 grams of heroin from Ms. Franks for himself and his girlfriend. He stated that his girlfriend overdosed after shooting up, but was taken to the hospital and survived. On cross-examination, he admitted that he had been picked up by the police while in possession of pain pills that he was planning to sell, but was neither arrested nor charged in exchange for providing information about Ms. Franks and the other defendants in the case. The next witness was the girlfriend who had allegedly overdosed, a 23-year-old White woman. The defender brought out her extensive prior criminal record, including that she had just been released from jail on a probation violation, and presented her medical records from the alleged overdose that indicated she had no opiates in her system. The final witness was an older White man who testified he was a long-time retail heroin dealer, and had purchased from Ms. Franks on multiple occasions. He claimed that on one occasion, he gave a “taste” of the heroin to a customer who immediately passed out, but he was able to revive the customer. There was no independent evidence of this alleged incident. He, too, admitted he was cooperating with federal prosecutors by providing information about Ms. Franks and her co-defendants to avoid federal charges and to minimize his own punishment in state court, where he was charged.

In light of the defense's evidence and argument, the prosecutor withdrew his motion pertaining to the alleged overdoses, but nonetheless argued for an “upward variance” from the guidelines. The judge complied, declaring that a sentence in the range of 97–121 months was appropriate, ultimately settling on 108 months—9 years—for Ms. Franks’ sentence.

Prosecutors in this district also stepped up their use of a 20-year mandatory minimum against defendants who supplied drugs resulting in a fatal overdose. A prosecutor I interviewed pointed to rise of fentanyl and fentanyl analogs as the reason for the increasing number of these prosecutions, citing numerous overdose deaths in rural parts of the district. She shared that during the crack epidemic, “we didn’t do a lot of overdose death cases arising out of the crack cases… this sounds terrible, but usually, the ones who died of a cocaine overdose were the ones who were like the street level dealers. And then the police would pull up and they’d, like, swallow their drugs and run, and they’d end up dying from it because, you know, there was too much and it would get absorbed.” She also shared that they were considering expanding the use of the 20-year mandatory charge in cases where a person overdosed but was revived by Narcan: “That doesn’t mean it's not a serious bodily injury [which is also subject to the 20-year mandatory]. I mean, a person died and now they’re back… We don’t charge them, [but] we’ve been looking at charging those as well.” Multiple defenders mentioned that prosecutors were generally not willing to deal on the heroin/fentanyl overdose cases, and line prosecutors often blamed either their office leadership or the Trump administration for their inability to negotiate. As one defender put it, “They’re not being very reasonable or flexible with those [overdose cases] because of the politics involved.”

Both prosecutors and defense attorneys shared that the 851 enhancement was also back to being used as a threat to obtain increased punishment in a plea agreement and/or to gain cooperation in drug trafficking cases. Several defense attorneys mentioned that prosecutors were also insisting that the 851 be applied and imposed in more cases than had been the case after Holder's policy change. The case data confirm that Southeast experienced an uptick in the use of the 851 drug-mandatory enhancement in the Trump period, although fewer overall filings relative to Northeastern. There were five drug mandatory enhancements imposed on defendants in 2014, before the Holder policy discouraging their use was implemented, then only four total in the 2015–2016 period. In the Trump period, 13 defendants received the enhancement, 12 of whom were convicted of distributing heroin or fentanyl. Since all but one of these were imposed before the law changed in December 2018, the defendants were subject to a doubled mandatory minimum with one prior drug conviction, and to life-without-parole with 2 or more prior convictions. Two of the 13 defendants subject to the enhancement in the Trump era received the mandatory life-without-parole sentence, one who was convicted at trial and the second one who had planned to go to trial, but entered into a plea agreement just days before the trial was to commence. The majority of the other cases were subject to a 20-year mandatory minimum. In that sense, unlike in Northeastern, the 851s actually filed in Southeast were almost always used to obtain very long mandatory minimums in the Trump era (as had been the case here during the height of the crack war).

A prosecutor disputed that there had been any change, “it's the same policy that we’ve always had, you know, that you file an 851 unless it becomes a part of a plea negotiation. And sometimes it does, sometimes it doesn’t.” This prosecutor did concede that the Holder policy, which was not well-received by prosecutors in this district, had required the office not to file: “under Holder, we were basically told, ‘Stop enforcing the law as written.’ [The law] was written so that, look, if you’ve got somebody who this is not their first time around and they’ve been dealing drugs before and they’re dealing significant amounts, then their sentence should be enhanced… And we were told, ‘No, you can’t do that.’” Under the Sessions’ regime, though, “we can go back to doing our jobs.” And that they did.

Discussion

The changing practices in the two districts profiled here demonstrate how the interplay of local legal culture and formal law and policy produces distinct outcomes, but that nonetheless move together to thwart the impact of reform. As such, they confirm both Beckett and Beach's (2021) and Carlen's (2002) observations about the challenges to achieve carceral retreat when powerful organizational actors are invested in maintaining their size and strength. Even where the DOJ could not directly rescind reformist policies such as the First Step Act or the sentencing guideline reductions, the constituent prosecutors’ offices could mobilize other legal tools to achieve their new punitive goals. In both Northeastern and Southeast, frontline prosecutors used the statutory tools at their disposal to implement the regressive war-on-crime that Sessions pledged to wage as Attorney General, and that William Barr then maintained during his tenure as Attorney General (Lynch, 2020). They also wielded their case-selection power in immigration cases to expand the pool of defendants, and keep their case numbers up with easy convictions. People who previously would either have been left in the community, or at most sent to immigration court for potential deportation, were now facing felony criminal charges, which were nearly always easy convictions for prosecutors.

Yet in both substantive areas addressed here—immigration and drug trafficking, two key targets in Sessions’ war—the two districts looked slightly different from each other. Northeastern and Southeast have consistently diverged, both in who they prosecute and in how they wield the punitive tools available to them in the federal system, and that divergence was growing under the Holder policies (Lynch, 2016). The change in administration seemed to halt that divergence, moving Northeastern in particular to take a much more punitive approach than had been the norm during the Obama years. In that regard, the changes in Northeastern appeared much more radical and aggressive than those in Southeast, even though Southeast continued to have more punitive outcomes, at least in drug cases.

In Northeastern, especially notable was the use of the 1028A aggravated identity theft charge, requiring a 2-year mandatory minimum, against undocumented immigrants. Their federal conviction and incarceration ensured these defendants would move directly to deportation proceedings upon completing their sentence (and indeed some had to agree to “assist” in that process to avoid the full 2-year sentence), so the custodial term in a U.S. prison was largely untethered from any traditional penal justification other than to make the defendant suffer a bit more pain. This practice was flourishing in Northeastern, but had yet to take hold at all in Southeast, suggesting that prosecutors in Northeast were using as many tools as possible to reassert their discretionary power. In the immigration cases, judges in Northeastern had become disillusioned with the pattern of low-level defendants being prosecuted and some had begun to routinely release these defendants in the pretrial stage, something that previously would rarely happen with any undocumented defendant. The 1028A conviction reasserted some prosecutorial sentencing power over judges in this ideological power struggle, forcing judges to impose 2-year-and-a-day sentences even if they did not want to.

This study's implications are sobering in regard to the potential for reforming the criminal legal system through piecemeal statutory changes, and the findings highlight the key role frontline prosecutors play in maintaining mass incarceration. As Beckett and Beach (2021) show, a system's penal code needs to be substantially restrained across the board, where statutory “hammers” are tailored to only a small subset of very serious cases, and where the ceiling on punishment for most convictions is not astronomical. In the federal system, even with recent reforms like the First Step Act, this is not the case. Statutory maximums remain very high, especially for drug cases, and the many built-in hammers, like the 851s and the 1028As, are there to be exploited by zealous prosecutors. So while the overall federal prison population did decline during the Trump administration, the effects of the Trump-era punitiveness in drug cases are cumulative and will increasingly be felt in the years to come, as those who received mandatory and enhanced sentences have no chance of early release absent executive clemency action. Moreover, while those convicted of immigration offenses will have less of an impact on year-end counts due to the generally short prison terms, Trump's DOJ ensured that many thousands more people churned through the federal prison system than would have under the previous administration.

Because none of the legislative reforms that have been implemented in the federal system rein in frontline prosecutorial discretion in case selection and charging, they will have modest benefits, at best, in reducing the reliance on incarceration. As Barkow (2019) has argued, comprehensive reform must also address the nearly-unchecked power of frontline prosecutors. She recommends a slate of changes that would make prosecutors more fiscally responsible for their contributions to mass incarceration, provide administrative oversight of their policies, and disaggregate case-level decision-making to mitigate bias. These kinds of reforms have yet to take hold, but will undoubtedly face pushback by prosecutors and other law enforcement groups as they directly target discretionary power.

The lesson from the federal system, then, is not good news for the rollback of mass incarceration. While both districts were more restrained in how they prosecuted drug cases under the Holder policies, that restraint quickly disappeared with the change of administration. Southeast reconstructed a renewed war on drugs that was eerily similar to its prior racialized war on crack. The demon substances were new—heroin and fentanyl—but the people subjected to the harsh mandatory minimums and 851 enhancements were mainly the same, predominantly Black men and women from impoverished neighborhoods who got involved in drug sales to survive. In the new version, however, the “victims” of the drug trade are predominantly White addicts whose death and suffering emboldened prosecutors to ensure even more punishment for those federally charged. In Northeastern, too, the new regime had brought the district back to the “bad old days,” in the words of one defender, with undocumented immigrants of color as the new targets for enhanced punishment.