Abstract
This article analyzes the traffic stop–to-deportation pipeline in New York State, how it harms children of immigrants, and how New York’s Greenlight Law seeks to disrupt it but has been hobbled by an implementation gap. It first establishes the phenomenon of the traffic stop–to-deportation pipeline by documenting how traffic stops are a key cause of deportations in New York State. Second, it analyzes how the pipeline harms (mostly US citizen) children of undocumented immigrants in New York State, who are more than 7 percent (more than 300,000) of New York State’s children. The pipeline makes these children fear and mistrust the police; harms their educational, social, and brain development; and consumes family income with the Mexican driver tax (costs incurred because parents could not get a driver’s license). Third, the article analyzes how the Greenlight Law should help remedy these harms, and how an implementation gap leaves many parents and children vulnerable to the pipeline. The implementation gap is partly due to the pandemic, but also driven by political and other factors that could be addressed by policy. Finally, the article analyzes how variation in implementing the Greenlight Law could leave the pipeline undisrupted and lead to unequal protection of the law by place in New York State. The article makes policy recommendations for stronger enactment to reduce the pipeline’s harms.
Keywords
Introduction
The last time I got stopped [while driving], like they [her children] were crying really bad. [Now] they see the cops, and they are like ‘Mommy, you got to hide, you got to hide!’…My [ten-year-old] son still remembers, and that was five or six years ago.…My five-year-old wakes me up at night, saying, ‘I had a bad dream they had deported you.’ And my ten-year-old told me he can’t concentrate at school, because he’s thinking about it if I will come home from work [or get deported].…He was failing school. He was supposed to repeat [fifth grade]. [Getting] a license…will be a lot better and secure because then I know for sure I’m not getting deported any minute and leaving my kids behind.
“Juana” here describes preventable harms being done to her US citizen children by her inability to get a driver’s license because she is undocumented. Such harms to children were part of the rationale for New York State (NYS) to pass the Driver’s License Access and Privacy Act — the Greenlight Law — in June 2019 (Smith 2019; Smith and Yrizar 2019). The law enables established undocumented immigrants to get a standard driver’s license, valid only for driving (not the enhanced driver’s license permitting entry to federal buildings or boarding planes; DMV 2020). Preventable harms to these mostly US citizen children produced by undocumented parents’ inability to get a driver’s license include deep fear and mistrust of the police; toxic stress driven by fear their parents will be deported, impeding social and brain development; and family impoverishment via the “Mexican driver tax.” These children’s fears are well founded. Traffic stops have been a key cause of deportations in NYS, creating the “traffic stop–to-deportation pipeline,” which the Greenlight Law could help disrupt. The potential harm from the pipeline is immense. Seven percent of NYS children (more than 300,000) and more than 5 million nationally live with at least one undocumented parent (Capps, Fix, and Zong 2016). Moreover, most are US citizens — 89 percent of oldest children of undocumented Mexican parents in our NYS-wide research project (N = 1,707) were US citizens, and 91 percent of all US citizen children in these families were age 11 or younger. The average child in these families harmed by the pipeline is a 4–5-year-old US citizen little boy or little girl.
Despite the Green Light Law’s passage, most undocumented immigrants in NYS still lack driver’s licenses, because the Covid-19 pandemic closed Department of Motor Vehicles (DMV) offices across NYS for months and because an implementation gap (Pressman and Wildavsky 1984; Pülzl and Treib 2007) stops the law from preventing these harms and disabling the pipeline. This implementation gap is driven by political resistance and bureaucratic hindrances, and it exacerbates unequal treatment of undocumented immigrants.
The Greenlight Law was supported by police and elected officials to whom the principal investigator (PI) spoke, although others reject the law and its restrictions on data sharing with federal authorities (Gavin 2020). These police and elected officials believe that ticketing undocumented parents driving children home from the doctor — as Juana was doing — or to work can permanently separate families through deportation. One told me that no one becomes a police officer to separate children from their parents on the way to work. These police officers understand the key problem with local police enforcing immigration laws (Taylor et al. 2014) — the police cannot protect the community without trust, which cooperation with Immigration and Customs Enforcement (ICE) undermines. If community members — including US citizens in immigrant families — fear any police contact may lead family members to be deported, they hesitate to call police to report crimes against them (Menjívar and Bejarano 2004; Lysakowski, Pearsall, and Pope 2009; Vishnuvajjala 2012). Menjívar (2014) reports close police–ICE cooperation leads immigrants to call the police the poli-migra (policia and migra, Mexican slang for immigration authorities). This vision of policing as based in community trust runs counter to seeing the policed community as a threat or enemy to be controlled by force (Varsanyi et al. 2012; Provine et al. 2016; Wong et al. 2019; Pass et al. 2020).
This article analyzes why and how the NYS Greenlight Law was needed, and how it can protect children and disrupt NYS’s traffic stop–to-deportation pipeline. It analyzes how the law’s ability to protect these immigrants and their mostly US citizen children depends on implementation, which some evidence suggests has been and will be unequal across the state, and which has been complicated by the pandemic and by anti-immigrant, fear-mongering rhetoric.
The article first establishes that the pipeline has been a key driver of deportation in NYS, analyzes reasons why Mexican and Central American immigrants had a high risk of deportation due to traffic stops, and describes how traffic stops can turn into deportation. Second, it describes how the traffic stop–to-deportation pipeline caused preventable harms to immigrants and their mostly US citizen children. It argues that more racial profiling in some localities increases the chances one gets deported in a traffic stop. The article then talks about how the Greenlight Law should decrease preventable harms to children in NYS. Some evidence shows Greenlight Laws make driving slightly safer (Lueders, Hainmueller, and Lawrence 2017). Third, the article analyzes how variation in Greenlight Law implementation could leave the traffic stop–to-deportation pipeline undisrupted and lead to unequal protection of the law in different places in NYS. It makes policy recommendations for stronger enactment to reduce the pipeline’s harms.
The DACA Access Project (aka Mexican Initiative on Deferred Action [MIDA])
Since 2014, my researchers and I have studied how having, lacking, gaining, or losing legal status, especially the Deferred Action for Childhood Arrivals (DACA) program, affects immigrant families in NYS. The DACA Access Project was a combined service, capacity-building, and research project that screened more than 1,700 people (mostly Mexican) for DACA or other visas (e.g., U visas for victims of violence) in 33 of NYS’s 52 counties and more than 200 zip codes. The project helped several hundred youth apply for DACA or permanent visas, helped build organizational capacity, and established a longitudinal research into the effects of having, gaining, lacking, or losing legal status among immigrants and their families, when nearly everyone screened (over 95 percent) agreed to become study participants. We have done more than 240 interviews and intermittent fieldwork in sites in NYS since 2015. The project’s intake interviews gathered detailed data on participants’ and families’ legal status and police interactions. Repeat interviews and ethnography with key cases deepen insight on how the traffic stop–to-deportation pipeline works, including cases resulting in warnings, tickets, or deportations. We also interviewed local authorities, police, and community and immigrant leaders, and did site fieldwork (e.g., a local DMV). We use publicly available datasets and published research in analyzing the Greenlight Law’s rationale, its potential positive impact, and the implementation deficit it now faces.
The Traffic Stop–to-Deportation Pipeline in New York State
The most wrenching, oft-recurring story we hear from parents and children is how much the children fear their parents will be deported due to traffic stops by local police. Their fear is well based. US Customs and Immigration Service (USCIS) data available through Syracuse University’s Transactional Records Access Clearinghouse (TRAC) system report the most serious criminal conviction for each deported person. These data show that of 100,614 people deported from NYS from 2002 to 2017, the most serious conviction for 68,799 (69 percent) was “No Conviction.” Hence, 69 percent of those deported had not been convicted of any crime. This reality contradicts the common belief that most deported people have committed crimes, and its corollary that undocumented immigrants are more likely to commit crimes. 1 In fact, research shows that increases in the undocumented population correlate throughout time with lower levels of crime (Rumbaut 2008; Simes and Waters 2014; Light and Miller 2018; Ousey and Kubrin 2018; Marshall Project 2019).
The role of traffic stops by local police in causing deportations emerges from other TRAC data detailing specific types of crimes of those deported, and “Method of ICE Arrest,” which describes how the person came into ICE custody. Table 1 presents TRAC data for four counties we are studying closely — Orange, Westchester, Suffolk, and Nassau (Long Island) — showing that “Traffic Offense,” “No Conviction,” “Driving under the Influence of Liquor,” and “Disorderly Conduct” are four of the top six causes of deportation. While driving under the influence and disorderly conduct are serious violations, we include them here in the belief that most Americans do not believe parents should be permanently separated from their children for these offenses alone. Better policy would be to help such drivers or disorderly persons to stop drinking, thus helping their mainly US citizen children. 2
Most Serious Convictions for Four New York Counties — From October 2014 to May 2018 (Top 5–6, until Traffic Offense Is Reported).
Source: US Customs and Immigration Service (USCIS) data accessed via Syracuse University’s Transactional Records Access Clearinghouse (TRAC) system (TRAC 2020).
In Table 1, No Conviction is one of the top four most serious crimes of deported people in all four counties: it leads in Nassau County, is second in Orange and Suffolk Counties, and is fourth in Westchester County. Driving under the Influence is the first or second most frequent most serious crime in all four counties. Traffic Offense and Disorderly Conduct are among the top six most serious crimes for all four counties. Moreover, the percentage of total deportations due to these causes is large. In Suffolk County, Driving under the influence, No Conviction, Disorderly Conduct, and Traffic Offense, in that order, led to 49.5 percent (984 of 1,987) of all deportations. In Nassau County, 47 percent (429 of 911) of deportations fell within the No Conviction, Driving under the influence, Disorderly Conduct, and Traffic Offense categories, in that order. In Orange County, Driving under the Influence, No Conviction, and Traffic Offense accounted for 42 percent (63 of 150) of all deportations. Finally, in Westchester County, 26 percent (66 of 255) of deportations were due to Driving under the influence, No Conviction, or a Traffic Offense. These data show that traffic stops with or without any other conviction are a main cause of deportation in these counties.
TRAC also reports how the deportees came into ICE custody in its “Method of ICE Arrest” data (aka apprehension method and agency data) presented in Table 2. Table 2 shows that in three of these four counties, local police (“CAP [Criminal Alien Program] Local Incarceration”) deliver to ICE the overwhelming number of undocumented people who are deported; while in the fourth county, NYS law enforcement (“CAP State Incarceration”) delivers more than local law enforcement. The number of deportees located (“Located”) by ICE in three of the counties is very small: eight in Nassau County, two in Orange County, and three in Westchester County.
Method of ICE Arrest — From October 2014 to May 2018.
Source: US Customs and Immigration Service (USCIS) data accessed via Syracuse University’s Transactional Records Access Clearinghouse (TRAC) system (TRAC 2020).
Suffolk County differs from the other counties in two ways. Like other counties, its single biggest apprehension method is CAP Local Incarceration, but its third largest is Located, meaning that ICE alone or in cooperation with other law enforcement found and deported specific people. That Suffolk County had more located deportees is driven by targeted large-scale efforts to deport violent gang members (476 deported from Long Island, New York City, and the Hudson Valley in 2018; Katersky and Carrega 2019). Those deported for violent crimes are not part of the traffic stop–to-deportation pipeline.
The second largest category — Non-Custodial Arrest — is part of the traffic stop–to-deportation pipeline. Noncustodial arrests occur when local police detain and refer someone to ICE without placing them into custody. Such arrests led to about a third of all Suffolk County deportations. Noncustodial arrests include cases like those our study participants described in which police stopped a car with a “Mexican-looking” family; discovered the driver, the mother, was undocumented and lacked a driver’s license; and delivered her to ICE on the side of the road while her young children sat in the car, terrified. In another case, a mother told the PI that local authorities arrested her for driving without a license and sent her three children, all under age 10, to walk three miles home alone on a highway in the dark of night in winter.
The harm done by most deported immigrants being handed to ICE by local or state law enforcement is that so many NYS (and American) children rightly fear the police. Nearly every family we talked to experienced at least one traffic stop, and most know someone who was incarcerated or deported for driving without a license. This fear undermines the key relationship that police need to do their jobs: the trust of the people they are sworn to protect and serve. If young children learn to mistrust the police because they reasonably fear their mother could be deported just for driving them home from the doctor, as happened with Juana, it also teaches these children that the law is not meant to protect them and that they are worth less in the eyes of the police.
How Can a Traffic Stop Lead to Deportation? The Greenlight Law Should Help Immigrant Parents and Their US Citizen Children
But how can a traffic stop lead to deportation? Traffic tickets are minor offenses. Most Americans suffer no lasting harm from them. Traffic stops can become deportations due to two factors: antecedent contact with law enforcement creating a higher risk of deportation for parents; and close cooperation between local law enforcement and ICE, converting any encounter with police into potentially permanent family separation. This can be true even when one is a crime victim, as when a long-term immigrant was “hand-delivered” by local police to ICE after calling them to report a trespasser (Silva 2018). More egregiously, an immigrant mother of four was raped and then worked closely with police to convict the rapist, only to be deported, despite the existence of the Violence Against Women Act and U visas designed to aid such vulnerable women (Dias 2019).
Antecedent Vulnerability to Deportation and the High Impacts of Traffic Stops
Antecedent encounters with police in prior traffic stops, or with immigration agents, can raise risks of deportation in later traffic stops. DACA Access Project data show 28 percent of parents reported having been detained previously by immigration officials, 43 percent had previously been fingerprinted (mostly during border crossing), 40 percent had been given tickets, and 44 percent had prior interaction with police or immigration authorities. 3 Having any prior contact with immigration officials or police — especially being fingerprinted — makes deportation more likely if one is taken into custody in the local jail for driving without a license. In addition, having prior tickets for driving without a license makes one more likely to be taken into custody.
During the New York State Senate’s televised June 2019 Greenlight Law debate, some upstate Republican lawmakers rejected the bill by arguing that these new immigrants “did not want to assimilate” or obey American laws, saying they “refused” to secure legal status and then a driver’s license. Most undocumented Mexican immigrants cannot legalize their status, despite living in the United States on average for 15 years (Krogstad, Passell, and Cohn 2019; American Immigration Council 2019) — an average of ten years for those in our study. Even those with adult US citizen children to petition for them are not guaranteed legal status, and delays for visas can be more than two decades (Kerwin and Warren 2019) — a cruel wait to impose on any parent or child. Given the obstacles to legal status, getting a driver’s license was impossible for most undocumented people before the Greenlight Law.
Another question was why immigrants who lacked a driver’s license simply did not drive. Why did they not take public transportation or hitch a ride with others? Many have sought to avoid driving, both by asking their DACA recipient children (who can get driver’s licenses) to drive more, and by paying for rides (raites) from an informal industry of paid upstate ride-givers (raiteros). But asking why one needs a car to get to work in upstate New York, with its scant public transportation and long distances, seems disingenuous. Most New Yorkers outside New York City must drive to work or school or to pick up children. For nearly all Americans, cars are prerequisites to full adult competence (Boyce 2018).
Close Cooperation with ICE and Racially Profiled Traffic Stops, Whose Frequency Varies by Locality
The experience of having been pulled over by the police was so common in our upstate sites that when organization staffers — US-born citizen Latinxs — heard potential DACA applicants talking about it, they shared stories of police pulling them over for no reason and asking if they had “papers” (a visa) and a driver’s license.
Research shows how common racially profiled traffic stops are, how they are done, and how immigrants try to avoid them. Two books by political scientists help establish this larger phenomenon: Pulled Over: How Police Stops Define Race and Citizenship by Charles Epp, Steven Maynard-Moody, and Donald P. Haider-Markel (2014); and Suspect Citizens: What 20 Million Traffic Stops Tell Us about Policing and Race by Frank Baumgartner, Derek A. Epp, and Kelsey Shoub (2018). These books show Black and Latinx drivers are more likely than whites to be pulled over and are issued harsher sanctions for infractions. Epp, Maynard-Moody, and Haider-Markel (2014) show Black drivers are more likely than whites to experience “investigatory stops,” wherein police stop motorists to see what they can find, rather than due to a concrete infraction in a “traffic safety stop.”
The situation for undocumented Latinx, and especially Mexican, drivers is complicated by their lack of legal status, raising the stakes for any traffic stops and removing some legal protections. Amada Armenta’s (2017) book, To Protect, Serve, and Deport, analyzes how police practices in Nashville led to Latinx immigrants being deported for driving without a license. In her ride-alongs with police, Armenta observed how stops of Latinx drivers did not usually start with officers observing dangerous driving, but rather with them spotting Latinx drivers and seeking a pretext (e.g., a cracked taillight) to stop them. Once stopped, officers asked about legal status. Nashville participated in the 287(g) program, a federal program that deputized local law enforcement officials to function as ICE agents. Armenta’s research documents extensively how this pipeline works, from pretextual stops to routine referral to ICE from such stops. Moreover, the US Supreme Court has implicitly blessed such pretextual, profiling stops by ruling in Lopez v. Mendoza (1984) that unlawful traffic stops (e.g., stopping someone for “looking Mexican”) do not protect against deportation, because an undocumented person was deportable before the illegal stop. 4
In this way, the situation in upstate New York is much like Armenta’s description of Nashville, in that “Mexicans” are a fairly new population emerging in the prior 20–30 years. “Mexicans” here mean anyone who “looks Mexican” (Latinx populations upstate are mainly Mexican and Central American). Local authorities have come to know that many Latinx drivers lack legal status and hence (before Greenlight) lacked driver’s licenses. Our study participants believe police are stopping Mexican-looking drivers in fishing expeditions, then asking about a driver’s license or legal status. That police may have a higher chance of finding someone lacking a driver’s license by racially profiling “Mexican-looking” drivers should not justify such profiling, because police are supposed to enforce traffic laws to protect public safety, not to write more tickets. One’s phenotype should not constitute probable cause for a traffic stop, 5 although Epp, Maynard-Moody, and Haider-Markel (2018) show that court rulings require only a thin veil to justify racially profiled stops. Such practices reinforce a dangerous association between looking Mexican and being “illegal,” formalizing a negative status via public institutions like the police that should treat all members of society equally, as required under the equal protection clause of the 14th Amendment (Smith 2017).
Our research presents a reality wherein the stance of local institutions, and especially police, can increase or decrease the importance of undocumented status for immigrant families. New York is a “blue” state overall, consistently voting Democratic in statewide and national elections. It participated only briefly in the 287(g) program, under Governor David Paterson’s short, accidental tenure (2008–2010). 6 So how did a traffic stop–to-deportation pipeline develop in a pro-immigrant state?
NYS’s overall blue/Democratic stance on politics hides deep divides between upstate/red–downstate/blue and rural-red versus urban-blue jurisdictions, and different institutional stances (police vs. schools) toward immigrants (Marrow 2009; Taylor et al. 2014). The traffic stop–to-deportation pipeline has functioned more strongly in New York in what we call “zones of scrutiny,” where study participants say the police regularly make exploratory stops of Latinx (especially Mexican) drivers without apparent reason (i.e., not a traffic violation) to ask about legal status and a driver’s license. If those stopped lacked a driver’s license, police often issued tickets for not having a driver’s license, as well as for infractions that study participants said they had not committed. Police issued Juana tickets for not wearing a seat belt and not having her toddler in a car seat, even though she wore one and he was in a car seat. Conversely, in places we call “havens,” police racially profile Latinx drivers less and cooperate much less with ICE, yielding fewer deportations. Study participants in havens reported fearing police less than those in zones of scrutiny did, and fewer feared traffic tickets could lead to deportation.
We can discern where on the zone of scrutiny–haven continuum a county falls using data from the DACA Access Project. We asked study participants if they had ever had any contact with the police or immigration authorities or had ever received tickets. The percentage of all encounters with authorities being due to traffic stops is higher in counties where immigrants reported being targeted more in racially profiled stops. The percentage of all encounters with any authorities being due to traffic stops is the highest in Orange County (60 percent), then Suffolk County (51 percent), Nassau (45 percent), and Westchester (35 percent).
The traffic stop–to-deportation pipeline is also driven by the stance of local sheriffs and their jail practices. Nationally, jails and sheriffs have been recognized as key deliverers of undocumented immigrants to ICE because they are countywide, and sheriffs’ electoral victories are often delivered by whiter, more rural or suburban voters who are less supportive of immigrants than more Democratic voters in a county’s cities (Wessler 2020). Moreover, since the Secure Communities program went nationwide in 2013, the fingerprints of everyone booked into a jail are sent to the Federal Bureau of Investigation, and then shared with ICE to check for immigration violations (as per ICE’s website under “Secure Communities,” https://www.ice.gov/secure-communities). Coleman and Kocher (2011) argue this represented a major shift because criminal investigations were firewalled off from civil immigration enforcement until the late 1990s. The sheriffs and jails in Orange and Suffolk Counties have been more vocal, enthusiastic collaborators with ICE than those in Nassau and Westchester. Collaboration with ICE is financially lucrative because it guarantees full jails funded with federal dollars. One study participant reported the Orange County sheriff had said in a meeting that getting an ICE detention center was the best economic thing ever for the county jails (Gilman and Romero 2018).
Three Preventable Harms to (Mostly US Citizen) Children from the Traffic Stop–to-Deportation Pipeline
Juana’s US citizen children were not the only ones in our study to display signs of anxiety and to cry whenever they see the police. Because the police arrested Juana while she was driving them home from the doctor, they saw and heard the interactions between their mother and the police officer. They heard the officer ask if Juana had a driver’s license and if she had legal status. Moreover, Juana’s children have been with her in more than one traffic stop. Finally, Juana’s children — all US citizens by birth — know others whose parents have been detained by the police for months or deported after traffic stops that they believe took place just to create the chance to ask their mother about her legal status and driver’s license. In our conversations with undocumented parents and their US citizen children, it was experience-based common sense to believe that traffic stops were investigatory and could lead to deportation.
These experiences lead these mostly US citizen children to grow up fearing and mistrusting police, and believing police specifically target Mexicans for traffic stops. Such fear undermines the trust necessary for the police to protect the community, because ethnic or racial groups that mistrust the police are less likely to assist the police as witnesses or report crimes as victims (Epp, Maynard-Moody, and Haider-Markel 2014; Taylor et al. 2014; Wong et al. 2019; Pass et al. 2020). These children’s fear and mistrust underline the irreconcilable conflict facing police who are asked to do immigration law enforcement. That so many children fear the police should alarm every police officer.
Indeed, some police officers and chiefs in havens understand this fear and seek to generate trust among Latinx communities. Some do not cooperate with ICE except for serious crimes, so community members know they will not get delivered to ICE for a traffic ticket. In one haven, the police chief came out against applying for an ICE grant for extra police officers, because it would have prohibited such non-information-sharing policies. The police chief told the PI they had spent several years generating trust in the Latinx community and did not want to undermine their ability to protect the community by taking ICE funding. Immigrant leaders and community members talked a lot about the police chief’s stand, which they felt protected them and treated them as valuable members of the community.
Lack of parental legal status can also lead to other harms to children, including the avoidance of routine healthcare. For example, undocumented parents are less likely to bring their children to the doctor or dentist (Lambert 2018), and are more likely to forgo nutritional supports for their US citizen children (e.g., the Supplemental Nutrition Assistance Program [SNAP]). This situation has worsened under the Trump administration, which increased the scope of disqualifying public benefits, endangering parents’ ability to legalize their status (Berk et al. 2000; Watson 2014; Castañeda et al. 2015; Kerani and Kwakwa 2018; Philbin et al. 2018).
Juana’s five year old’s nightmares and her 10 year old’s inability to focus on school, for fear of their mother being deported and being separated from them, illustrate how the pipeline can harm children’s physical, emotional, and psychological development. Concretely, these negative effects include posttraumatic stress syndrome, trauma, and difficulty forming strong emotional attachments and adapting to changes in family structure (Torrico 2010; Rojas-Flores et al. 2017; Arellano et al. 2018). Other research finds that constant fear of separation from parents — or actual separation after parental deportation — makes it harder for children to learn in school and harms brain development (Zayas and Heffron 2016). Constant fear of losing their parents prompts children to become hyper-alert to any threat, continually engaging their fight-or-flight response and taking their focus and their body chemistry away from normal childhood tasks in school or with friends. Such trauma changes children’s brain structure in ways that cause physical, mental, and emotional harm into adulthood, even increasing the risk of substance abuse (Dye 2018; Enoch 2011).
Evidence emerged after DACA’s passage that having undocumented parents increases rates of anxiety for the second generation and causes intergenerational transmission of disadvantage. In particular, referral rates for mental health treatment fell among children of undocumented mothers with DACA versus those lacking DACA (Hainmueller et al. 2017). They also fell for DACA recipients (Venkataramani et al. 2017; Patler and Pirtle 2018). Threats and the prior administration’s attempt to end DACA, however, increased anxiety among DACA recipients (Patler et al. 2019). We posit these actions will also raise anxiety rates among their second-generation children.
Juana reported that her children could not stop worrying about their mother’s possible deportation. One worried constantly she would not be home after school, because ICE would take her. At home, he worried that someone might come and take his mother while he slept, and so he could not sleep. When a child cannot feel secure that his mother will be there when he returns home from school or goes to sleep in his own bed, and when this fear is based on his own experience, it makes sense his schoolwork would suffer. It also indicts the policies and practices causing that fear and trauma.
Juana was stopped several times and issued multiple tickets in each stop, finally owing $2,000 in tickets. Fearing deportation from traffic court — which friends told her could happen if she did not pay tickets, fines, and court fees — she hired a lawyer for $1,000 to pay everything and avoid referral to ICE. Her strategy worked, but she had to borrow $3,000, repaid in $300 monthly payments, plus a $3,000 payoff payment — a usurious, informal loan she felt compelled to take to avoid losing her children. Parental deportation affects families catastrophically — one study shows a 70 percent fall in income, for families already struggling financially and whose children were almost all US citizens (Chaudry et al. 2010; Boyce and Launius 2020).
In fear, Juana’s family sold their car and began paying for rides to work, school, the supermarket, and elsewhere. Both parents worked 60 hours per week and then paid $910 of their joint $3,000/month income — 30 percent — for rides. Given that maintaining and insuring their car had cost $300/month, this represented a $600/month loss in income. Her $300 monthly debt service raised what we call her “Mexican driver tax” to $1,200 of $3,000, or 40 percent of total income. Juana’s family suffered financially for years afterward. They mostly ate rice, beans, milk, and cereal, and rarely meat. Juana would not go to the doctor unless very sick, to save money. Juana’s family’s savings were fully spent in resolving her tickets. Other study participants reported higher Mexican driver taxes — one woman paid 66 percent of her income for rides after a traffic stop made her fear losing her children by deportation.
We call such extra expenditures due to lacking a driver’s license the Mexican driver tax because study participants felt they were stopped, as one put it, because the police “saw my Mexican face” and pulled them over. The Mexican driver tax describes all extra costs a family must bear because family members cannot obtain driver’s licenses, including paying for rides, tickets for driving without a license, and the family’s fear of permanent separation in the traffic stop–to-deportation pipeline. Given that the median family in our study has two children and earns $24,800/year (about $2,067/month), a Mexican driver tax of several hundred dollars would consume much of the family’s disposable income.
How the Greenlight Law Should Help Immigrant Families, including Their US Citizen Children
New York’s Greenlight Law should help these children in several ways. First, having one’s undocumented parents get a driver’s license could lower children’s anxiety and trauma. The children could experience fewer traumatic encounters with the police as the number of pretextual or investigatory stops falls. Having fewer terrifying interactions with police — and perhaps some positive interactions — could reduce fear among these parents and their US citizen children (Soss and Weaver 2017). Enabling established undocumented parents to secure driver’s licenses makes them visible to local institutions. It also means that they no longer have to break the law by driving without a license to bring their children to the doctor or visit their children’s teacher in school.
Second, having a standard driver’s license could remove the pretext that study participants believed motivated their traffic stops (i.e., that police saw a “Mexican face” and stopped them to check for legal status or a license). Moreover, study participants believe their lacking a license makes police feel they can give “extra” unmerited tickets. Enabling these parents to get driver’s licenses should reduce the moral hazard created when local governments fund municipal services too much using fines like traffic violations. The status quo strongly incentivizes police officers to write more tickets, because their own paychecks depend on this revenue. Even if racially profiled stops persist, however, parents with driver’s licenses should get fewer tickets, which should decrease their children’s terror of the police.
Third, less anxious children will suffer less from trauma or toxic stress, which could yield better grades and better health in the short run for these 300,000-plus NYS children, putting them on stronger academic and health trajectories as they launch into adulthood. Healthier children who receive more education will also pay more in taxes throughout their adult lifetimes (Ryo and Peacock 2021).
Fourth, enabling long-term, established NYS immigrant parents to get driver’s licenses should decrease the Mexican driver tax significantly because driving one’s own car is cheaper than paying for rides, and having a license should decrease or prevent tickets, fines, and legal fees for driving without one. Driver’s licenses will prevent diversion of hard-earned family income to the Mexican driver tax, enabling families to buy better food, medicine, or books.
Fifth, enabling established undocumented parents to get driver’s licenses should lessen the increasingly dangerous association of looking “Mexican” with “illegal” in New York, where most newly arrived immigrants are Mexican or Latinx, and many are undocumented.
Sixth, driver’s licenses should improve immigrant families’ health across the state, as parents become less fearful of driving to doctor visits and more likely to drive their children to preventative care visits.
Finally, we have an interesting counterfactual finding in DACA recipients who have obtained licenses. They report still being pulled over, but they less frequently get tickets after showing their license. This pattern ironically confirms the racial profiling at work, because if they were getting stopped for traffic violations, they would still get tickets (Epp, Maynard-Moody, and Haider-Markel 2014). Yet they do not receive tickets because they are not, in fact, being stopped for traffic violations. It also highlights the “extra” tickets participants described to us (e.g., for not having a car seat, when the child was in one).
Implementation Deficits Threaten to Perpetuate Preventable Harms to New York State Children
Implementation gaps (Pressman and Wildavksy 1984; Hudson, Hunter, and Peckham 2019) could stymy the Greenlight Law’s positive impacts and perpetuate preventable harms to NYS children that the law sought to remedy. Passed in June 2019, the law went into effect in December 2019 but has been rolled out haltingly, creating an implementation gap.
Implementation theory argues that the distinction between politics and public administration breaks down under complex systems of dispersed government (e.g., federalism) in which some policy process actors oppose the policy (Hudson, Hunter, and Peckham 2019). Opposition to New York’s Greenlight Law exists at federal and state levels. Implementation theory predicts that any large-scale policy will have coordination issues and inefficiencies intrinsic to street-level bureaucracy (Lipsky 1980). Elster (1992) and Heinrich (2018) note that local institutions and agents can impose higher administrative burdens on users of state services (immigrants) who they want to discourage or exclude entirely from those services.
Political opposition to the law has emerged in some places upstate and on Long Island, and was promoted by ICE during the prior federal administration. Soon after New York’s law passed, several county clerks — who would implement the law — declared they would not do so, echoing county clerks’ refusals to issue marriage licenses to same-sex couples (Goldbaum 2019). Foreseeing the law would pass, upstate Republican State Senator Jim Tedisco proposed in May 2019 the unsuccessful County Clerks Protection Act, enabling clerks to refuse to implement the Greenlight Law and protecting them from the consequences for sharing information about Greenlight drivers with federal authorities (prohibited by Greenlight). Other NYS elected officials sought unsuccessfully to repeal the Greenlight Law (DeWitt 2020). Two county clerks sued to prevent the law’s implementation. Nine states’ attorneys general and the New York Civil Liberties Union (NYCLU) presented amicus briefs describing the public safety and human rights benefits of Greenlight Laws (in Kearns v. Cuomo, 2019). Evidence to date shows Greenlight Laws in other states make roads slightly safer and lower insurance premiums slightly because there are fewer uninsured drivers (Cáceres and Jameson 2015; Lueders, Hainmueller, and Lawrence 2017). The county clerks’ suit stalled during the pandemic.
These anti–Greenlight Law efforts were strongly supported by the prior federal administration. In a February 2020 visit to New York, ICE’s then–acting director declared, without evidence, that the law would make New York more dangerous (Campbell 2020; Fink 2020). New York Governor Andrew Cuomo rejected these assertions, arguing that ICE would otherwise use the DMV information to deport undocumented people “living peacefully, who have not committed a crime, who are not violent” (Campbell 2020). ICE’s acting director argued that ICE does targeted enforcement, a claim not backed up by USCIS’s own data via the TRAC system. 7 The law has so far withstood these challenges and remains in effect.
The Greenlight Law implementation gap worsened in the pandemic. The Greenlight Law was beginning to be used in early 2020, when Governor Cuomo closed all DMV offices in mid-March 2020 due to the pandemic (WYNT 2020). DMVs have reopened for online services. But because obtaining a standard driver’s license requires taking a written test and a road test, it is unclear if one can even apply for this license now or (if so) at how many DMVs. One of our county’s DMV websites reports that it now provides in-person service by appointment but does not mention Greenlight licenses.
The implementation gap is also driven by reluctant or incomplete enforcement of the law due to limited resources or purposeful limiting of Greenlight licenses via bureaucratic discretion (Lipsky 1980). In one of our counties, only one of five DMV sites offered Greenlight licenses. The local elected official setting this policy argued that NYS did not provide DMVs extra funding for the influx of applicants. Moreover, our fieldwork shows this one DMV reduced the number of Greenlight applicants allowed from only 100 per day, then 50, then only 35, while not limiting applications for other services. Reports confirm this happens in other counties as well (Smith 2020). Applicants told us they lined up the night before or at 5 a.m. in February in upstate New York to secure a place in line. Research on policy implementation that focuses on street-level service delivery (Lipsky 1980) shows that bureaucrats use discretion to manage case overloads. While we sympathize with bureaucrats tasked with offering new services without more funding, the “fix” of restricting only Greenlight Law applicants denies equal protection to the intended beneficiaries of this law, and signals they are less protected and valued than those seeking other services. It impermissibly imposes an extra, targeted, administrative burden on a specific class (Elster 1992; Heinrich 2018).
The DMV security guard’s line management (which authors Smith, Besserer Rayas, and Flores observed at the single site offering Greenlight licenses) reinforced the perception of unequal access: he gave the first applicant in line a legal pad to start a list of the first 35 arrivals, who would get help; others would be sent away. He spoke mostly in English (although in close proximity to us he spoke fluent Spanish) while knowing that nearly everyone in line spoke only Spanish. To the extent that those administering the law are engaged in unequal limitation of services offered, or slowdowns, or they place special conditions on access to that service (getting a standard driver’s license), they risk engaging in unconstitutional discrimination.
The implementation gap is also driven by insufficient capacity to deliver services in Spanish. NYS’s DMV states, “All DMV Offices will provide language access assistance to individuals with limited English proficiency.” Yet in this county, responsible officials report they can offer Greenlight licenses in Spanish at only one of five DMV offices.
Implementation gaps can be driven on the “demand” side by fears among potential applicants. As noted above, usage of public programs or services by immigrants decreases in response to anti-immigrant rhetoric or policies. Visits to doctors or dentists for US citizen children of immigrant parents, for example, decrease with anti-immigrant policy changes or rhetoric. Hence, expanding the list of public benefits that can be used to find immigrant parents a “public charge” led to less service and benefit usage for immigrants’ US citizen children (Bernstein et al. 2019; Zallman et al. 2019). ICE’s hostility to the Greenlight Law, and prior federal attempts to gain access to NYS’s DMV records, could scare many into not applying. News and social media reports of the Department of Homeland Security’s ban of NYS from Global Entry and the Trusted Traveler program likely led many to fear the state could not prevent their data from being accessed by ICE. Finally, there could be a lack of demand for Greenlight licenses due to fear of COVID-19, including fear of going to crowded places.
A final related fear is the inability of immigrants to know which state or federal government offices or officials to trust, and when, as immigrant leaders explained to us. For example, the DMV we visited in early February 2020 was in the same complex as a county sheriff’s office, which is the site from which ICE agents operate. Study participants told us they thought the county was purposefully offering Greenlight licenses at only this site to discourage applications. Moreover, study participants told us they feared using inclusive government services because ICE has targeted them. For example, one study site implemented a municipal identification (ID) law, which gives undocumented parents a government ID to enter their children’s public schools or visit them in the hospital. Yet in 2019, ICE agents — without informing local authorities — tried to apprehend a man who had come to pay a traffic ticket in the building where municipal IDs are issued. The encounter created a dangerous situation of visibly armed, unidentified men in plainclothes chasing a ticket-paying community member, with local police responding to reports of unknown armed men chasing a local resident through the building.
Policy Recommendations and Conclusion
In proposing policy or practice changes to address implementation gaps for the Greenlight Law, we differentiate between different types of gaps and their causes, and, when possible, we propose both full, more expensive solutions and practical, no-cost or low-cost solutions. For example, the official running the county DMV argued that the real problem was the state imposing an unfunded mandate. Thousands of new people now qualify for DMV services; but without additional funding, this official did not have enough bilingual staff at each DMV site. One fix would be to fund more bilingual DMV staff. This problem is widespread, however, and is increasing in NYS and nationwide, as Latinx immigrants, especially, settle in previously mostly white suburban or rural areas. Just as NYS’s Office of New Americans created an online English language program staffed by live teachers (Cell-Ed) and has conducted much business online in the pandemic, it could create a NYS Translator Corp to translate remotely for any NYS office. This service could be provided economically. Staff lines in a NYS Translator Corp could be created for common languages (Spanish, Chinese, and Haitian Creole), but translators for less common languages could be contracted for remote availability. This Translator Corp could be located within the state’s Office of New Americans, which already has a hotline for immigration questions and reporting fraud. We propose a state-level fix because most localities cannot resolve this problem themselves, and they often ask any bilingual staff to help or require clients to bring their own translators, often their children, which is a bad fix for many reasons.
Second, practices we observed seemed to violate at least the spirit of the 14th Amendment’s equal protection clause. These practices included the separate lines for people applying for all other services and those applying for the standard license; restrictions on how many could apply for standard licenses but not how many could access other services; and making those in the standard line (who were all Latinx) stand outside in February in upstate New York, while others were ushered inside. Such practices create de facto unequal access to government services and protection of law, which reinforces the perception of Latinxs that they get unequal protection. Indeed, in our upstate fieldwork, a Latinx US citizen who went to renew his license joined the Greenlight Law line, telling us he believed it was the line for Latinxs. Local Latinx leaders remarked that siting the DMV office with bilingual capacity so near the local ICE field office had a chilling effect on standard license applications.
One simple short-term fix to this problem would be to disseminate a “Dear Colleague” letter to DMV staff outlining how such policies or practices violate equal protection law and agency standards and discourage use of DMV services. NYS might also have DMV staff offer a brief training on the issue and require directors of DMV offices to certify that they do not permit such policies or practices. In an interview posted on Facebook, the county administrator’s answers seemed to suggest nothing was wrong with having separate lines; she was doing her best with a tight budget. A Dear Colleague letter and certification requirement could foster the bureaucratic imagination to resolve the problem. DMV could analyze data it already collects (if such data exist) on average wait time from entry to service for different kinds of clients and services, which might identify unequal access across service type or racial lines.
The above recommendations focused on the Greenlight Law itself, but we also suggest policy and practice changes at the state level to prevent or disrupt the traffic stop–to-deportation pipeline. NYS could issue statewide law enforcement guidance or regulations on how to deal with undocumented immigrants and traffic or other minor violations. The district attorneys in some field sites have created immigration units to advise immigrants with such minor violations of possible immigration impacts of their cases. New York City and State have funded programs offering legal representation to immigrants in deportation or in routine legal proceedings in which minor convictions can have severe consequences for their families. NYS could adopt a statewide policy that local law enforcement not book people into jail for routine traffic offenses, which would prevent the jail from fingerprinting them and triggering delivery to ICE, and explain that police cannot adequately protect a community that does not trust them and sees them as a threat. If NYS does not have the power to make such a statewide policy, a Dear Colleague letter outlining the rationale for practices that disrupt this pipeline could help change thinking, and would give local advocates leverage in demanding change in local or county policing.
Expanding school busing in communities with many immigrants or nondrivers would reduce time spent driving and chances of a police encounter, and decrease children of immigrants’ anxiety and fear of the police (though it would not fix underlying legal status issues).
Passing the Greenlight Law was an important way that New York’s governor and its elected leaders showed immigrants and their children that they are valued members of the NYS community and worthy of equal protection of the law (de Graauw and Vermeulen, 2021; Provine and Varsanyi 2020). 8 Currently, most undocumented parents in NYS have not been able to secure a Greenlight license. We hope fuller implementation of the law will reduce the preventable harms the traffic stop–to deportation pipeline has caused, and help these NYS children thrive and become productive adults. Finally, our research demonstrates the need to integrate fair policing of Latinxs and undocumented immigrants in current debates over police reform. Thoughtful police reform would reduce the harms caused to (mostly US citizen) children of undocumented immigrants by the traffic stop–to deportation pipeline and help these children to trust the police.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Acknowledgments and Funding
The PI for this project (Rob Smith) would like to acknowledge funding and inspiration that supported research for this article. The following sources helped fund this research: the National Science Foundation (SES-1026580 in 2012-2014, and a 1997-2000 grant “Gender, race and ethnicity and social organization: determinants of second generation Mexican American educational and work mobility in New York City”), the William T. Grant Foundation (Major Research Grant # 187567 and Major Research Grant-6662), the Russell Sage Foundation (Grant G-6601), and PSC-CUNY in funding most rounds 49-52. Manny Castro, then a Baruch MPA student, first pointed out to me how traffic stops are a key first step towards deportation, confirmed by how traffic stop stories abounded in our DACA project interviews.
