Despite deportation’s devastating effects, the Immigration and Nationality Act (INA) specifies deportation as the penalty for nearly every immigration law violation. Critics have regularly decried the INA’s lack of proportionality, contending that the penalty often does not fit the of­fense. The immigration bureaucracy’s implementation of the INA, how­ever, involves a spectrum of penalties short of deportation. Using tools such as administrative closure, orders of supervision, and deferred ac­tion, agency bureaucrats decide who is deported and who stays, and on what terms, on a purely ad hoc basis. In this “shadow system,” immi­grants, their advocates, and the broader public lack basic information about what penalties are being imposed and why.

This Article argues for reframing the problem of immigration law’s disproportionality as a problem of insufficient justification—one remedi­able only by building the infrastructure for reason giving in the immigra­tion bureaucracy. Deportation strikes many as disproportionate because the government often lacks satisfactory reasons for imposing such a dras­tic penalty. But in the system of shadow sanctions today, the government not only fails to offer good reasons: It fails to offer any at all. As a result, the system of shadow sanctions represents a classic case of an arbitrary exercise of government power. Looking to examples of procedural innova­tion across the administrative state, this Article backs prudential reforms to create immigration law’s missing reason-giving infrastructure. With it in place, the public can demand better reasons or proportionality. But the first step is addressing immigration law’s arbitrariness problem.

The full text of this Article can be found by clicking the PDF link to the left.


Since the nineteenth century, judges have recognized deportation as a penalty “beyond all reason in its severity.” 1 See Fong Yue Ting v. United States, 149 U.S. 698, 759 (1893) (Field, J., dissenting). One of the plaintiffs failed to obtain the certificate of residence because he could not pre­sent the requisite “white witness” to vouch for his lawful presence. Id. at 729–31. In Fong Yue Ting v. United States, the Chinese immigrant plaintiffs allegedly failed to obtain certifi­cates of residence required for lawful presence under section 6 of the Chinese Exclusion Act of 1892, rendering them deportable. 2 Id. at 725–26. Justice Stephen Johnson Field argued in dissent that “neglect to obtain a certifi­cate of residence” hardly warranted forced expulsion without indictment and trial. 3 Id. at 758–59. Such a punishment involves removal from one’s place of resi­dence and the “breaking  up  of  all  the  relations  of  friendship,  family,  and  business  there contracted.” 4 Id. at 759.

Like Justice Field, immigration law scholars have long bemoaned the cruelty of “the deportation state,” 5 See, e.g., Adam B. Cox & Cristina M. Rodríguez, The President and Immigration Law 79 (2020) [hereinafter Cox & Rodríguez, The President and Immigration Law]. focusing specifically on immigration law’s overreliance on deportation and the resulting disproportionality in many cases. 6 See, e.g., Angela M. Banks, The Normative and Historical Cases for Proportional Deportation, 62 Emory L.J. 1243, 1298 (2013) [hereinafter Banks, Cases for Proportional Deportation] (arguing that immigration law disproportionately sanctions longer-term per­manent residents with minor convictions); Jason A. Cade, Enforcing Immigration Equity, 84 Fordham L. Rev. 661, 665 (2015) [hereinafter Cade, Enforcing Immigration Equity] (argu­ing that “[t]he removal system . . . imposes dire penalties on the basis of a broad range of civil infractions”); Jill E. Family, The Future Relief of Immigration Law, 9 Drexel L. Rev. 393, 406 (2017) (arguing for adopting the European Convention of Human Rights’ proportion­ality analysis in immigration); Laila Hlass, The Adultification of Immigrant Children, 34 Geo. Immigr. L.J. 199, 256 (2020) (noting the need for a system of graduated sanctions in immigration law, especially to distinguish between child migrants and adults); Daniel Kanstroom, Smart(er) Enforcement: Rethinking Removal, Structuring Proportionality, and Imagining Graduated Sanctions, 30 J.L. & Pol. 465, 466 (2015) [hereinafter Kanstroom, Smart(er) Enforcement] (arguing for “tak[ing] the notions of proportionality and gradu­ated sanctions seriously in structural—rather than in discretionary—ways”); Juliet Stumpf, Fitting Punishment, 66 Wash. & Lee L. Rev. 1683, 1684 (2009) (noting that, in immigration law, “[n]either the gravity of the violation nor the harm that results governs whether depor­tation is the consequence for an immigration violation”); Michael J. Wishnie, Immigration Law and the Proportionality Requirement, 2 U.C. Irvine L. Rev. 415, 416 (2012) (noting widespread public perception of immigration law as “too harsh”). The black letter law of the Immigration and Nationality Act (INA) specifies deportation as the sanction for nearly all transgressions of immigration law, no matter how minor, and regardless of the personal cir­cumstances of the immigrant. 7 See, e.g., 8 U.S.C. § 1227(a)(1)(C) (2018) (listing a violation of nonimmigrant sta­tus or a condition of admission, such as overstaying a visa, as a deportable offense). The INA contains waivers from inadmissibility and removability that can soften the application of deportation, but they are not systematic, and little data is available on how frequently they are granted. See Shoba Sivaprasad Wadhia, Darkside Discretion in Immigration Cases, 72 Admin. L. Rev. 367, 402 n.164 (2020); see also U.S. Citizenship & Immigr. Servs., Number of Service-Wide Forms by Fiscal Year To-Date, Quarter, and Form Status 2019, https:// [https:
//] (last visited July 26, 2021).
In failing to recognize that deportation lacks adequate justification in many cases, the criticism goes, immigration law contemplates life sentences for the proverbial parking violation. 8 See Rummel v. Estelle, 445 U.S. 263, 274 n.11 (1980) (“This is not to say that a proportionality principle would not come into play in the extreme example . . . if a legisla­ture made overtime parking a felony punishable by life imprisonment.”). Crit­ics have often pressed analogies to criminal law, arguing that some deport­able immigrants warrant penalties short of deportation: 9 See Banks, Cases for Proportional Deportation, supra note 6, at 1266–78 (compar­ing criminal law to immigration law); Stumpf, supra note 6, at 1720–40 (using criminal law as a model for proportionality and discussing alternative sanctions such as fines and delayed access to immigration benefits); see also Wishnie, supra note 6, at 418–24 (describing juris­prudence of proportionality in criminal law). Just as those convicted of crimes must receive proportionate sentences in some cases under the Eighth Amendment, deportation is unwarranted in at least some immigration cases. 10 See Solem v. Helm, 463 U.S. 277, 299–300 (1983) (reviewing life sentence without parole for proportionality). But see Rummel, 445 U.S. at 274 (recognizing that there is legis­lative prerogative when imposing felony sentences).

Although criminal procedure offers a model for protecting targets of enforcement from civil liberties violations, 11 Michael Kagan, Immigration Law Is Torn Between Administrative Law and Criminal Law, Yale J. on Regul.: Notice & Comment (Feb. 12, 2016), https://www.yalejreg.
com/nc/immigration-law-is-torn-between-administrative-law-and-criminal-law-by-michael-kagan/ [] (“Criminal procedure offers tools that are espe­cially well suited to address the urgent civil liberties concerns in immigration law, especially the manner in which [DHS] arrests and detains people on immigration allegations.”).
analogies to criminal law have foundered conceptually and in the courts. Conceptually, criminal law does not offer a compelling example of proportionality. 12 See Stephen F. Smith, Proportionality and Federalization, 91 Va. L. Rev. 879, 883 (2005) (arguing that federal criminal law lacks fidelity to the principle of proportionality, largely because federal judges have construed federal crimes expansively, increasing the number of potential defendants). Legislatures retain vast discretion to punish felonies with lengthy terms of imprisonment, and proportionality in criminal law has limited traction outside of capital pun­ishment and life sentences without parole. 13 Harmelin v. Michigan, 501 U.S. 957, 965 (1991) (rejecting the notion that the Eighth Amendment contains a “proportionality guarantee”); see also Youngjae Lee, The Constitutional Right Against Excessive Punishment, 91 Va. L. Rev. 677, 681 (2005) (charac­terizing the Supreme Court’s proportionality jurisprudence in criminal law as “highly un­satisfactory and disappointing; the body of law is messy and complex, yet largely meaningless as a constraint, except perhaps in a few instances in the capital context”). As the jurisprudence stands today, the Eighth Amendment offers a weak, inadequate check on dispro­portionate punishment, underscored by rampant overpunishment and overcriminalization. 14 See Excessive Punishment, Equal Just. Initiative, [] (last visited July 26, 2021) (describing “dec­ades of harsh and extreme sentencing”); see also Jones v. Mississippi, 141 S. Ct. 1307, 1317–19 (2021) (holding that, for juveniles who commit homicide, the Eighth Amendment does not require a separate finding of permanent incorrigibility to sentence someone to life in prison without parole).

Federal courts have also roundly rejected efforts to apply constitu­tional proportionality doctrines to removal orders. Courts have specifically rejected application of the Eighth Amendment to removal orders, instead affirming immigration law’s “civil” rather than “criminal” character, 15 See, e.g., Hinds v. Lynch, 790 F.3d 259, 266 (1st Cir. 2015) (noting “removal’s civil character”). not­withstanding recognition of the overlap. 16 See Padilla v. Kentucky, 559 U.S. 356, 365–66 (2010) (noting that “deportation is nevertheless inherently related to the criminal process”); see also Daniel Kanstroom, Deportation Nation: Outsiders in American History 122 (2007) [hereinafter Kanstroom, Deportation Nation] (discussing the Supreme Court’s “bright-line” between detention in­cident to removal and “infamous punishment” in Wong Wing v. United States, 163 U.S. 228 (1896)). Courts have similarly rejected applying the Fifth Amendment Due Process Clause’s proportionality rule for punitive damages to removal orders. 17 See BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 562 (1996) (noting that “[t]he Due Process Clause of the Fourteenth Amendment prohibits a State from imposing a ‘grossly excessive’ punishment on a tortfeasor” (quoting TXO Prod. Corp. v. All. Res. Corp., 509 U.S. 443, 454 (1993))); see also A. Benjamin Spencer, Due Process and Punitive Damages: The Error of Federal Excessiveness Jurisprudence, 79 S. Cal. L. Rev. 1085, 1123 (2006) (cri­tiquing BMW v. Gore’s “guideposts” for evaluating punitive damages awards). Because immi­gration law is federal law, the Fifth Amendment’s Due Process Clause applies, but courts and scholars have presumed that the Fifth and Fourteenth Amendment Due Process Clause standards are equivalent. See, e.g., Hinds, 790 F.3d at 268 n.8 (treating claims under the Due Process Clauses of the Fifth and Fourteenth Amendments equivalently); Angela M. Banks, Proportional Deportation, 55 Wayne L. Rev. 1651, 1655 (2009) (noting that the Fifth Amendment ensures proportionality “in the civil context” and citing Fourteenth Amendment Due Process cases). At the end of the day, efforts to apply constitutional proportionality doctrines in immigration law have reached a dead end. 18 See, e.g., Hinds, 790 F.3d at 263 (explaining that “federal courts have long de­scribed removal orders as non-punitive and, therefore, not punishment”).

This Article takes a different approach to the INA’s failure to mean­ingfully distinguish among deportable immigrants. Rather than pressing constitutional proportionality doctrines or analogies to criminal law, it looks inward to the apparatus of the immigration bureaucracy by examin­ing the many informal, discretionary tools that bureaucrats use to show lenience. These tools include deferred action, administrative closure, and orders of supervision, all of which defer the issuance or execution of a removal order, sometimes indefinitely. 19 See infra section II.B. These tools comprise a system of shadow sanctions, 20 One might view forms of lenience as “benefits” rather than “sanctions,” but they have elements of both. Although some shadow sanctions come with work authorization and other protections, immigrants pay a price for coming out of the shadows and onto the gov­ernment’s radar. See Asad L. Asad, On the Radar: System Embeddedness and Latin American Immigrants’ Perceived Risk of Deportation, 54 Law & Soc’y Rev. 133, 135 (2020) (arguing that immigrants view “documents” as enhancing their visibility to immigration en­forcement, fueling the government’s power to “surveil and expel” them). Accordingly, this Article characterizes these tools of lenience as “shadow sanctions.” one that remains largely unregulated and hidden from public view. 21 See, e.g., Jennifer Lee Koh, Removal in the Shadows of Immigration Court, 90 S. Cal. L. Rev. 181, 184 (2017) (discussing how, in fiscal year 2013, “approximately 83% of all formal removal orders took place through either reinstatement of prior removal orders or expedited removal of individuals seeking admission at the border”—summary proceedings outside of immigration court). Koh further argues that longstanding critiques applicable to standard removal proceedings in immigration courts “are far more pronounced, and far more common, in immigration court’s shadows.” Id. at 186. Most noncitizens likely lack any knowledge of what these shadow sanctions are, how to apply for them, or the standards by which the government imposes them. 22 See Shoba Sivaprasad Wadhia, Demystifying Employment Authorization and Prosecutorial Discretion in Immigration Cases, 6 Colum. J. Race & L. 1, 5 (2016) [hereinaf­ter Wadhia, Demystifying Employment Authorization] (noting that there is no form or other information available about deferred action, other than for the Deferred Action for Childhood Arrivals (DACA) program). For discussion of DACA, see infra section II.B.1.) Public data about the use of these tools are also limited, save for occasional Freedom of Information Act (FOIA) requests. 23 See, e.g., Shoba Sivaprasad Wadhia, The Aftermath of United States v. Texas: Rediscovering Deferred Action, Yale J. on Regul.: Notice & Comment (Aug. 10, 2016), https:
// [] [hereinafter Wadhia, The Aftermath of United States v. Texas] (discussing the results of a FOIA request on USCIS grants of deferred action by a field office).
As evidenced by this lack of openness and consistency—key val­ues traditionally associated with the law—the system of shadow sanctions suffers from serious rule of law deficits. 24 See Gillian E. Metzger & Kevin M. Stack, Internal Administrative Law, 115 Mich. L. Rev. 1239, 1258 (2017) (describing “values traditionally associated with the rule of law—specifically the values of authorization, notice, justification, coherence, and procedural fairness”). Rather than requiring bureau­crats to make reasoned, consistent, and nonarbitrary judgments when dis­tinguishing among deportable immigrants, 25 Cf. Judulang v. Holder, 565 U.S. 42, 55 (2011) (applying administrative law’s requirement that agency action be based on nonarbitrary factors relevant to immigration policy). the system of shadow sanctions creates a zone of unregulated agency discretion. 26 See infra section II.C.

Structuring, constraining, and checking agency discretion is a core project of administrative law. 27 See Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 25–26 (1971). This Article argues for putting administra­tive law’s doctrinal and conceptual resources to use in mitigating the flaws of the shadow system. Looking to agency practices in other areas of regu­lation as well as the example of the European Union, this Article argues for specific prudential reforms to promote reasoned immigration administration.

Reasoned administration serves as a necessary foundation for, but does not equal, proportionality. 28 See infra section I.B. As discussed below, there are many con­ceptions of proportionality. The immigration law literature has gener­ally taken proportionality to mean that the severity of a penalty should track the severity of the offense and should consider noncitizens’ partic­ular cir­cumstances, such as their length of residence in the United States. This Article argues for sanctions better tailored to these considerations and highlights the role of reason giving in attaining this objective.