The Case For Chevron Deference To Immigration Adjudications

Volume 71 October 2021
THE CASE FOR CHEVRON DEFERENCE
TO IMMIGRATION ADJUDICATIONS

Patrick J. Glen Adjunct Professor of Law, Georgetown University Law Center; Senior Litigation Counsel, United States Department of Justice, Civil Division, Office of Immigration Litigation. The author would like to thank Chris Walker and Smita Ghosh for their helpful comments on an initial draft of this article. The views and opinions expressed herein are the author’s own, and do not represent those of the Department of Justice or any component thereof.

PDF
PDF

Abstract

Chevron skepticism is in vogue in legal academia, as Professors Shoba Wadhia and Christopher Walker’s recent entry in the genre demonstrates. They place their project within the broader academic trend of arguing for limitations on the application of deference to various administrative decisions, but their aim is ultimately narrower—to show that “this case against Chevron has . . . its greatest force when it comes to immigration.”

The Professors are incorrect. Immigration adjudication presents one of the strongest cases for deference to administrative adjudication. This case is founded in the text of the statute itself and its myriad general and specific delegations of authority to the Attorney General, the expertise of the agency which has honed its interpretive enterprise through adjudicating tens of thousands of cases annually, and the ultimate political accountability of the agency head in immigration adjudication. For these reasons, the Supreme Court has applied Chevron deference to immigration adjudications since the very foundation of that framework. Although they advance an interesting contrarian thesis, the Professors ultimately provide no sound basis for retreating from four decades of established jurisprudence.

Introduction

In a recent article, published as part of Duke Law Journal’s annual Administrative Law Symposium, Professors Shoba Sivaprasad Wadhia and Christopher Walker argue that Chevron deference should not be applied to immigration adjudications, specifically, the decisions of the Board of Immigration Appeals. [1][1]. Shoba Sivaprasad Wadhia & Christopher J. Walker, The Case Against Chevron Deference in Immigration Adjudication, 70 Duke L.J. 1197 (2021). The removal proceeding is the proceeding held before an immigration judge to determine a noncitizen’s removability from the United States, as well as any eligibility for relief or protection from removal. See 8 U.S.C. § 1229a (2018) (providing the statutory authority and framework for the conduct of removal proceedings); see also Patrick J. Glen, Judulang v. Holder and the Future of 212(c) Relief, 27 Geo. Immigr. L.J. 1, 4 (2012) (providing a historical overview of removal proceedings). The immigration judge conducts the removal proceeding in the first instance, see 8 U.S.C. § 1229a(a)(1), with the Board of Immigration Appeals exercising appellate jurisdiction over the decisions of immigration judges, see 8 C.F.R. § 1003.1(b)(3) (2021) (providing the bases for appellate jurisdiction). This is a provocative thesis, contrary as it is to nearly four decades of federal court decisions, including repeated affirmations of Chevron’s applicability to immigration adjudications by the Supreme Court. [2][2]. See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56 (2014) (applying Chevron deference to uphold a decision of the Board); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (same). It is sure to prompt further debate in academia, as well as the federal courts, and to the extent it forces all interested parties to reconsider the justifications for and application of deference principles, it will be an important article. [3][3]. See, e.g., Amaya v. Rosen, 986 F.3d 424, 438 n.1 (4th Cir. 2021) (Richardson, J., dissenting) (noting the article and its argument that Chevron deference should not apply to immigration adjudications, while opining that Chevron must still be applied as a matter of precedent).

If the article has a fundamental shortcoming, however, it is perspectival: the Professors write largely from outside the system they are critiquing. That does not undercut the strength that many of their points carry, but it does provide an opening for a riposte from within the system. Having litigated the Chevron issue for the government in the immigration context before both the federal courts of appeals and the Supreme Court, my views differ significantly both in how the justifications for deference are framed and with the bottom-line conclusion that adjudications should categorically be denied deference. On my account, I would place significant weight on Congress’s delegations to the Attorney General in the Immigration and Nationality Act and the background principle of plenary power, which may be weakened in its contemporary form but is far from dead. Even assuming deference must be justified with resort to principles beyond Congress’s explicit delegations, immigration adjudications pass that bar: such adjudications are the poster child for political accountability, as the Attorney General sits at the apex of the adjudicatory bureaucracy, and the agency possesses both the craft and legislative expertise that the Professors view as integral to the agency’s decision-making process. Finally, it is important to note that rulemaking—the Professors’ preferred avenue for immigration policy-making—and adjudication often serve different purposes, a point with special importance under the INA, a statute that makes clear those circumstances where Congress mandated rulemaking and those where the agency has a freer hand to choose the method for implementing policy. The agency should be entitled to deference regardless of the path it chooses, so long as its decision is otherwise reasonable.

Part I begins by placing the Chevron issue in context. Although the immigration bureaucracy is a sprawling system of adjudication encompassing hundreds of thousands of adjudications each year before both immigration judges and the Board of Immigration Appeals, Chevron deference is concerned only with the vanishingly small minority of cases in which the Board issues a precedential decision interpreting an ambiguous provision of the INA. And in that small minority of cases, the decision-making of the Board is robust and comprehensive. Part II then proceeds to the meat of the issue, arguing that Chevron deference to immigration adjudications is appropriate. This section addresses the main points already highlighted above: the explicit general and specific statutory delegations to the Attorney General and the background principle of the Executive Branch’s plenary power in immigration; the political accountability in immigration adjudication via the Attorney General’s referral authority; and the significant institutional expertise the Board brings to bear in resolving cases through precedent-setting adjudications. Part III argues that there is no substantial reason to prefer rulemaking to adjudication when it comes to affording deference to the agency’s determinations, while highlighting the complementary roles each often plays in the development and promulgation of policy. Finally, Part IV turns to the Professors’ recommendations for reform, while posing the question of what effect, if any, would Chevron’s recalibration have in immigration cases?

I. Placing Chevron in Context: Immigration Adjudications and the Board of Immigration Appeals’ Decisional Process

Professors Wadhia and Walker lay out at some length the structure of the immigration adjudicatory system and its operational realities. [4][4]. See Wadhia & Walker, supra note 1, at 1204–13. Rather than retread that ground, it is worthwhile to address two additional points that are especially relevant for addressing the Chevron question: the scope of Chevron’s potential applicability to immigration adjudications, and the decisional process the Board engages in when issuing decisions entitled to Chevron deference.

First, Chevron’s potential scope. The immigration adjudication system is massive, encompassing 65 immigration courts and 529 immigration judges, [5][5]. See Executive Office for Immigration Review (“EOIR”) Immigration Court Listing, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/immigration-court-administrative-control-list [https://perma.cc/3LD5-KVPW] (last updated Aug. 2, 2021); U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Immigration Judge (IJ) Hiring (2021), https://www.justice.gov/eoir/page/file/1242156/download [https://perma.cc/8HQB-B57N]. as well as the Board of Immigration Appeals, an appellate body that now has 23 permanent members. [6][6]. See Board of Immigration Appeals, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/board-of-immigration-appeals-bios [https://perma.cc/5HPB-FZX2] (last updated July 19, 2021). In 2020, immigration courts received 367,038 new cases, completed 231,435, and ended the year with a backlog of 1,256,954 cases. [7][7]. See U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Pending Cases, New Cases, and Total Completions (2021), https://www.justice.gov/eoir/page/file/1242166/download [https://perma.cc/9QMQ-4N3X]. In the same year, the Board received 51,300 new appeals, resolved 33,974 appeals, and ended the year with 84,769 appeals pending. [8][8]. EOIR, Adjudication Statistics, Case Appeals Filed, Completed, and Pending, U.S. Dep’t Just. Exec. Off. For Immigr. Rev., https://www.justice.gov/eoir/page/file/1248501/download [https://perma.cc/3QRT-7WYZ] (last updated July 8, 2021). Such a high workload may, from a 30,000-foot view, lend credence at the threshold to an argument that deference is not appropriate to immigration adjudications. Professor Wadhia and Walker hint at this argument throughout their article, noting not only high caseloads but also certain variances in case outcomes based on which immigration court is conducting the proceeding and whether the alien is represented. [9][9]. See Wadhia & Walker, supra note 1, at 1228–30. These factors may undercut the contention that deference is warranted because they point to a system that is incapable of engaging in the type of reasoned adjudication that should undergird a decision claiming entitlement to deference. [10][10]. See id.

For that reason, it is worthwhile drilling down through the system to focus on that narrow class of decisions—out of the hundreds of thousands of cases that will be resolved by immigration judges and the Board in any given year—for which the question of Chevron deference would be a live issue. Deference applies only when the adjudicator “acts in its lawmaking capacity and, in the case of the BIA’s adjudications, that means only when the BIA’s decision is binding precedent within the agency.” [11][11]. Flores-Molina v. Sessions, 850 F.3d 1150, 1157 (10th Cir. 2017) (quoting Rangel-Perez v. Lynch, 816 F.3d 591, 597 (10th Cir. 2017)). The Board’s decision is binding as precedent only in circumstances where the Board issues a published decision, [12][12]. See 8 C.F.R. § 1003.1(g) (2021). and that occurs only rarely in any given year. In 2020, the Board issued 25 precedential decisions, a number on par with its average yearly output over the last eight years: 16 (2019), 23 (2018), 27 (2017), 26 (2016), 28 (2015), 29 (2014), 19 (2013). [13][13]. See Agency Decisions, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/ag-bia-decisions [https://perma.cc/BJ8H-JXYQ] (listing the Board’s precedential decisions for each volume) (last updated Sept. 8, 2021). It is this class of cases—representing .07% of the total decisions issued in 2020—that is relevant for Chevron purposes. But even this calculation overstates the universe of cases to which the government may ultimately seek deference. Although the Board may issue a precedential decision on any issue, and although that decision will bind agency adjudicators in the discharge of their responsibilities, many of these cases will not be entitled to deference before the courts of appeals. For instance, in cases where the Board decides on the divisibility of state criminal statutes or whether such a statute is a categorical match to the generic federal offense, [14][14]. See, e.g., Matter of Nemis, 28 I. & N. Dec. 250 (BIA 2021) (adjudicating application of the categorical approach to the federal conspiracy statute, 18 U.S.C. § 371, and a visa-fraud provision, 18 U.S.C. § 1546(a)); Matter of Dikhtyar, 28 I. & N. Dec. 214, 214–16 (BIA 2021) (addressing the divisibility of the Utah state drug schedules by specific controlled substance); Matter of P-B-B-, 28 I. & N. Dec. 43 (BIA 2020) (addressing the divisibility with regard to Arizona’s criminalization of possession of a “dangerous drug”). the decision will not be entitled to deference before the courts of appeals even though the decision is a precedential decision that binds agency adjudicators. This is because the Board is held to have no particular expertise on state criminal laws, and because the categorical approach implicates a legal analysis to which deference is not relevant. [15][15]. See, e.g., Omargharib v. Holder, 775 F.3d 192, 196 (4th Cir. 2014) (“Although we generally defer to the BIA’s interpretations of the INA, where, as here, the BIA construes statutes [and state law] over which it has no particular expertise, its interpretations are not entitled to deference.” (alteration in original) (quoting Karimi v. Holder, 715 F.3d 561, 566 (4th Cir. 2013))); Ramos v. U.S. Att’y Gen., 709 F.3d 1066, 1069 n.2 (11th Cir. 2013) (“We owe no Chevron deference to the Board’s interpretation of the Georgia [criminal] statute, which the Board has no power to administer.”); Denis v. U.S. Att’y Gen., 633 F.3d 201, 208 (3d Cir. 2011) (“[I]f the issue turns on the meaning of a federal statute other than the INA, we possess the requisite expertise to interpret a federal criminal statute such that no deference is due.” (citation omitted)); see also Almeida v. Holder, 588 F.3d 778, 785 (2d Cir. 2009) (deferring to the BIA’s interpretation of “theft offense,” but reviewing questions of a particular crime defined by state law de novo). For present purposes, it is enough to note that the universe of Chevron-eligible decisions in the immigration context is exceedingly narrow, and implicates less than one-tenth of one percent of all Board decisions issued in any given year.

Second, a corollary of the first point, the decisional-process undergirding the issuance of a precedential decision is more intensive and deliberative than issuance of single-member decisions. Precedential decisions are only issued in cases where the Board has sat as a three-member panel or en banc, and the decision may only be issued as a precedent where the entire Board membership discusses the case and votes in the affirmative to issue it as such. [16][16]. See 8 C.F.R. § 1003.1(g) (2021). In this sense, every Board decision is essentially en banc—no precedent issues unless a majority of the Board members believes it should be so issued. In this way, the work of the Board parallels that of the federal courts of appeals. The Ninth Circuit may issue only a handful of precedential decisions in any given week, while issuing 100 unpublished memorandum dispositions. [17][17]. For March 15, 2021, for instance, the Ninth Circuit issued 21 unpublished decisions and 3 published decisions. See Log of Unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021); Opinions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/opinions [https://perma.cc/8ST8-G668] (last visited Oct. 7, 2021). On March 12, 2021, when it issued no published decisions, it issued 13 unpublished decisions. See Log of unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021). Nobody would allege that a published opinion issued by the Ninth Circuit is somehow lacking in relative deliberative processes just because the court also issued 20 or more unpublished decisions the same day. What matters for assessing the adequacy of deliberation is the case at issue, and for the Board and Chevron purposes, the vanishingly small number of precedential decisions it issues in a year, not the thousands of other orders and decisions it will reach.

II. The Case for Chevron Deference

When a court is confronted with an issue resolved by the Board in a precedential decision, it should afford deference to that resolution under familiar principles of Chevron deference. This deference stems from the statute’s own explicit delegations to the Attorney General, coupled with background principles of judicial deference to the Executive Branch in immigration matters, the political accountability the agency enjoys in the course of discharging its adjudicatory functions, and its comparative expertise in the immigration issues subject to adjudication. Although Professors Wadhia and Walker attempt to cast doubt on the legitimacy of these justifications for deference in the immigration context, their arguments are ultimately misguided.

A. Explicit Delegations and Structural Cues

The “statutory” case for Chevron deference in immigration adjudications is simple and straightforward. The INA broadly provides the Attorney General with authority to “establish . . . regulations, prescribe such forms of bond, reports, entries, and other papers, issue such instructions, review such administrative determinations in immigration proceedings, delegate such authority, and perform such other acts as [he] determines to be necessary for carrying out this section.” [18][18]. 8 U.S.C. § 1103(g)(2). The Act also provides that the Attorney General’s “determination and ruling . . . with respect to all questions of law shall be controlling.” [19][19]. 8 U.S.C. § 1103(a)(1). These grants of authority are important markers for Chevron purposes, as they represent “express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.” [20][20]. United States v. Mead Corp., 533 U.S. 218, 229 (2001). Beyond the broad delegations framing the Attorney General’s authority under the INA as a whole, the statute contains many more specific, express delegations of authority. For instance, the INA delegates to the Attorney General the authority for establishing procedures for consideration of asylum applications, provides authority to establish mandatory bars to eligibility and other conditions for consideration of such applications, and uses open-ended terminology that allows for interpretation by Executive Branch adjudicators and rule-makers. [21][21]. See 8 U.S.C. § 1158(d)(1) (“The Attorney General shall establish a procedure for the consideration of asylum applications filed under [this section].”); 8 U.S.C. § 1158(d)(5)(B) (“The Attorney General may provide by regulation for any other conditions or limitations on the consideration of an application for asylum not inconsistent with this chapter.”); 8 U.S.C. § 1158(b)(2)(B)(ii) (“The Attorney General may designate by regulation offenses that will be considered to be a [disqualifying crime for purposes of asylum eligibility].”); 8 U.S.C. § 1158(b)(2)(C) (“The Attorney General may by regulation establish additional limitations and conditions, consistent with this section, under which an alien shall be ineligible for asylum under paragraph (1).”). See generally Patrick J. Glen, In re L-A-C-: A Pragmatic Approach to the Burden of Proof and Corroborating Evidence in Asylum Proceedings, 35 Geo. Immigr. L.J. 1, 34 (2020) (“The asylum statute . . . does not generally mandate specific procedures for the consideration of asylum applications. Rather, that section is composed almost entirely of explicit and implicit delegations.”). And the INA otherwise tends towards broad and general language for key points that necessitates interpretation by the Attorney General, whether that pertains to what is encompassed by the term “public charge,” [22][22]. See 8 U.S.C. §§ 1182(a)(4), 1227(a)(5) (stating public charge ground of inadmissibility and deportability, respectively); see also Field Guidance on Deportability and Inadmissibility on Public Charge Grounds, 64 Fed. Reg. 28,689, 28,691 (Mar. 26, 1999) (providing public guidance as to how the Immigration and Naturalization Service would apply the public charge grounds of removability); Inadmissibility on Public Charge Grounds, 84 Fed. Reg. 41,292, 41,315 (Aug. 14, 2019) (revising the Executive Branch’s interpretation of the public charge inadmissibility ground). how “good moral character” should be assessed, [23][23]. See 8 U.S.C. § 1101(f) (listing per se categories that would involve a finding that the alien lacked good moral character, but allow the Attorney General discretion to conclude that aliens outside those categories also lack good moral character); see, e.g., 8 U.S.C. § 1229b(b)(1)(B) (requiring good moral character in order to establish eligibility for cancellation of removal). or what offenses fall within the parameters of crimes of “domestic violence, stalking, and child abuse,” [24][24]. See 8 U.S.C. § 1227(a)(2)(E)(i); see also Matter of Velazquez-Herrera, 24 I. & N. Dec. 503, 504 (BIA 2008) (interpreting this provision). to offer only a few examples. The point is that the statutory framework established by Congress not only explicitly delegates broad authority to the Attorney General to resolve both general and specific interpretive issues, but almost invariably uses language that invites resolution through interpretation.

The Supreme Court has itself based its decisions extending Chevron deference to immigration adjudications on this statutory framework. In Immigration and Naturalization Service v. Cardoza-Fonseca, [25][25]. Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421 (1987). the Court recognized that although there will still be strictly legal questions to which deference will not apply, there are also statutory ambiguities in the INA “which can only be given concrete meaning through a process of case-by-case adjudication,” and in the process of filling the gaps, the courts “must respect the interpretation of the agency to which Congress has delegated the responsibility for administering the statutory program.” [26][26]. Id. at 448 (internal quotation marks and citations omitted). The Court clarified its rationale for applying Chevron to immigration adjudications in INS v. Aguirre-Aguirre. [27][27]. Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 424–25 (1999). Given that the INA explicitly provided that the “determination and ruling by the Attorney General with respect to all questions of law shall be controlling,” [28][28]. 8 U.S.C. § 1103(a)(1). the Court deemed it “clear that principles of Chevron deference are applicable” to immigration adjudications. [29][29]. Aguirre-Aguirre, 526 U.S. at 424. In Aguirre-Aguirre, the Court found further support in the statute’s specific conferral of “decisionmaking authority on the Attorney General” in determining an alien’s eligibility for withholding of removal, [30][30]. Id. at 424–25. as well as its prior recognition “that judicial deference to the Executive Branch is especially appropriate in the immigration context where officials ‘exercise especially sensitive political functions that implicate questions of foreign relations.’” [31][31]. Id. at 425 (quoting Immigr. & Naturalization Serv. v. Abudu, 485 U.S. 94, 110 (1988)).

These latter two points should be emphasized. First, the statute broadly provides authority to the Attorney General to interpret the statute he is charged with administering, while containing numerous other delegations of decisional authority akin to what the Supreme Court found compelling in Aguirre-Aguirre, including provisions relating to asylum, cancellation of removal, adjustment of status, and numerous other waivers and additional forms of relief or protection from removal. [32][32]. See Patrick J. Glen, Interring the Immigration Rule of Lenity, 99 Neb. L. Rev. 533, 560–61 n.191 (2021) (noting that “[t]he statute is . . . replete with express delegations of decisional authority to the Attorney General,” and citing examples relating to asylum, cancellation of removal, and other forms of relief). In other words, the INA is not a statute where courts must discern whether there is an implied or implicit delegation of authority to resolve statutory ambiguity; Congress has provided its explicit instruction that the Attorney General’s determinations, made in the course of discharging his responsibilities and decision-making authority under the INA, should be given “controlling” weight. [33][33]. See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”).

Second, judicial deference in the immigration context must be placed within the history of Executive authority over immigration; the Executive’s authority stems not only from the statutory scheme, but constitutes a background principle inherent in the nature of immigration law. In 1950, the Supreme Court noted that “[w]hen Congress prescribes a procedure concerning the admissibility of aliens, it is not dealing alone with a legislative power. It is implementing an inherent executive power.” [34][34]. United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 542 (1950) (emphasis added). In Fiallo v. Bell, [35][35]. Fiallo v. Bell, 430 U.S. 787 (1977). the Supreme Court opined that its cases “have long recognized the power to expel or exclude aliens as a fundamental sovereign attribute exercised by the Government’s political department largely immune from judicial control.” [36][36]. Id. at 792 (citation omitted). This simply restated what it had said a year earlier in Hampton v. Mow Sun Wong [37][37]. Hampton v. Mow Sun Wong, 426 U.S. 88 (1976).: “the power over aliens is of a political character and therefore subject only to narrow judicial review.” [38][38]. Id. at 101 n.21 (citation omitted). For this reason, the Supreme Court has traditionally, and even still, extended broad deference to the Executive’s immigration-related decisions because of the specific context of immigration law.

Beyond these principles, subsequent decisions of the Supreme Court have focused on the specific expertise of the agency in resolving complicated issues of immigration law, as well as the policy-oriented disposition of certain issues that permissibly rests with the agency. For instance, Justice Kagan recently described an immigration case as

[T]he kind of case Chevron was built for. Whatever Congress might have meant . . . it failed to speak clearly. Confronted with a self-contradictory, ambiguous provision in a complex statutory scheme, the Board chose a textually reasonable construction consonant with its view of the purposes and policies underlying immigration law. [39][39]. Scialabba v. Cuellar de Osorio, 573 U.S. 41, 75 (2014) (plurality opinion).

And in the related context of the ordinary remand-rule, the Court used agency expertise as a rationale for remanding to the agency rather than a court of appeals to address legal or statutory interpretation questions in the first instance. [40][40]. See Negusie v. Holder, 555 U.S. 511, 523 (2009) (explaining that remand rules exist, in part, “because ‘ambiguities in statutes within an agency’s jurisdiction to administer are delegations of authority to the agency to fill the statutory gap in reasonable fashion. Filling these gaps . . . involves difficult policy choices that agencies are better equipped to make than courts.’” (quoting Nat’l Cable & Telecomm. Ass’n. v. Brand X Int. Serv., 545 U.S. 967, 980 (2005)). Justice Kennedy also noted that on remand,
[t]he agency’s interpretation of the statutory meaning of “persecution” may be explained by a more comprehensive definition, one designed to elaborate on the term in anticipation of a wide range of potential conduct; and that expanded definition in turn may be influenced by how practical, or impractical, the standard would be in terms of its application to specific cases. These matters may have relevance in determining whether its statutory interpretation is a permissible one.
Id. at 524; see also Gonzales v. Thomas, 547 U.S. 183, 186–87 (2006) (per curiam) (noting that on remand, “[t]he agency can bring its expertise to bear upon the matter; it can evaluate the evidence; it can make an initial determination . . . it can, through informed discussion and analysis, help a court later determine whether its decision exceeds the leeway that the law provides” (quoting Immigr. & Naturalization Serv. v. Ventura, 537 U.S. 12, 17 (2002) (per curiam))).

This history provides a firm foundation for Chevron’s application. Congress has explicitly delegated authority to the Attorney General, including to conduct adjudications under the INA, and this delegation bolsters an inherent Executive authority in the immigration context. Add to that the expertise the agency has in its sole subject matter focus, and it is little wonder that deference has been applied to immigration decisions continually since the advent of Chevron.

B. Political Accountability and Attorney General Referral

Immigration adjudication, perhaps more than any other form of contemporary agency adjudication, offers exactly the type of political accountability that should underpin Chevron deference. The Professors argue that two types of political accountability should be relevant for administrative purposes, an “elections matter” conception that advances the “electoral accountability in the administrative state,” and a “deliberative accountability” that is centered on participatory goals, including obtaining the views of affected or interested parties. [41][41]. See Wadhia & Walker, supra note 1, at 1231–32 (citations omitted). The Professors only make the conclusory assertion that “policymaking through adjudication may not be an adequate substitute for rulemaking under an ‘elections matter’ accountability theory,” seemingly basing this contention on a comparison of adjudication versus rulemaking rather than the inherent lack of political accountability in immigration adjudication. [42][42]. See id. Regarding deliberative shortcomings, they rely on purported shortcomings in the decisional process of the Attorney General on review. The Professors’ concerns are overblown, and that political accountability is certainly present in the immigration system.

The Attorney General is, under the text of the statute, the chief administrator of the INA. [43][43]. See 8 U.S.C. § 1103(a), (g). The Board acts only as his delegate, and only within the context in which he has delegated his authority to the Board to act. [44][44]. See 8 C.F.R. § 1003.1(d)(1) (1958) (“The Board shall function as an appellate body charged with the review of those administrative adjudications under the Act that the Attorney General may by regulation assign to it.”); see also Delegation of Powers and Definition of Duties, Immigration and Naturalization Service, 5 Fed. Reg. 2454, 2454 (July 1, 1940) (“[T]he Board of Review of the Immigration and Naturalization Service shall have authority to exercise the powers of the Attorney General [in certain delineated cases].”). As the Attorney General’s delegate, the Board must still exercise its own independent judgment, [45][45]. See 8 C.F.R. § 1003.1(d)(1)(ii) (“Board members shall exercise their independent judgment and discretion in considering and determining the cases coming before the Board.”). and its decisions are deemed its own and not attributable to the Attorney General. [46][46]. See Tefel v. Reno, 972 F. Supp. 608, 613 n.1 (S.D. Fl. 1997) (“[T]he decision of the BIA is not factually, nor legally, the decision of the Attorney General.”). Moreover, a corollary to the regulatory requirement that the Board’s decisions are a reflection of its own independent judgment, the Attorney General may not attempt to influence or dictate the decisions of the Board. [47][47]. See United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260, 266–67 (1954) (“[T]he word ‘discretion’ means that the recipient must exercise his authority according to his own understanding and conscience. In short . . . the Attorney General denies himself the right to sidestep the Board or dictate its decision in any manner.”). But the regulations do safeguard the Attorney General’s fundamental role in immigration adjudication, by permitting him to decide cases he opts to refer to himself, or in which the Board or the Department of Homeland Security requests his review. [48][48]. See 8 C.F.R. § 1003.1(h). This mechanism allows the Attorney General to decide both discrete issues and broader questions of law through adjudication, and means that “the Attorney General is the final arbiter of the immigration agency’s interpretation of a statute.” [49][49]. Sanchez-Penunuri v. Longshore, 7 F. Supp. 2d 1136, 1149 (D. Colo. 2013).

Exercise of this authority fits squarely with the “elections matter” conception of Executive Branch accountability. [50][50]. See Wadhia & Walker, supra note 1, at 1232. The Trump Administration Attorneys General utilized the authority to implement an immigration policy that tracked with what the campaign promised it would do, a policy implementation that followed the election returns. [51][51]. See Patrick J. Glen, Robert F. Kennedy and the Attorney General’s Referral Authority: A Blueprint for the Biden Administration, 49 Hofstra L. Rev. 719, 721 (2021) (noting how the Trump Administration utilized Attorney General referral to advance an immigration policy consonant with its campaign promises, including restricting eligibility for asylum, streamlining enforcement and adjudications to limit the possibilities of delay in removal proceedings, and focusing on detention of aliens both during and after the conclusion of removal proceedings). Likewise, I have recently argued that the Biden Administration should utilize the authority to implement its preferred immigration policy within the limits the authority permits. [52][52]. See generally id. (discussing Biden administration’s policy).

Other Attorneys General have also utilized the referral authority in a manner that squares with this aspect of the political accountability theory. A few examples: Attorneys General in the Reagan and Bush Administrations referred the deportation case of a United Kingdom citizen based on foreign policy concerns and the impact on the U.S.-U.K. relationship; [53][53]. See Deportation Proceedings for Joseph Patrick Thomas Doherty, 13 Op. O.L.C. 1, 8 (1989); Deportation Proceedings of Joseph Patrick Thomas Doherty, 12 Op. O.L.C. 1, 8 (1988); see also Alberto R. Gonzales & Patrick Glen, Advancing Executive Branch Immigration Policy Through the Attorney General’s Review Authority, 101 Iowa L. Rev. 841, 882–86 (2016) (providing background and context to the Attorney Generals’ decisions in Matter of Doherty). Attorney General Reno referred the Board’s decision in Matter of R-A-, rejecting the claim of an asylum applicant that victims of domestic violence constitute a “particular social group,” based on the concerns of advocates, who had also led the Clinton Administration to propose rulemaking on the same issue; [54][54]. See R-A-, 22 I. & N. Dec. 906, 906 (Att’y Gen. 2001); see also Gonzales & Glen, supra note 53, at 886–89 (providing background and context to the Attorney General’s decision in Matter of R-A-). Attorney General Mukasey referred a decision on female genital mutilation and reversed the Board’s restrictive opinion after an outcry from advocates and direct pleas by lawmakers and advocates to intervene; [55][55]. See A-T-, 24 I. & N. Dec. 617, 617 (Att’y Gen. 2008); see also Gonzales & Glen, supra note 53, at 861–63 (providing background and context to the Attorney General’s decision in Matter of A-T-). and Attorney General Holder remanded a same-sex civil union case for further proceedings amid the Obama Administration’s deliberations over whether to continue defending the Defense of Marriage Act. [56][56]. See Dorman, 25 I. & N. Dec. 485, 485 (Att’y Gen. 2011); see also Glen, supra note 51, at 760–71 (providing background and context to the Attorney General’s decision in Matter of Dorman). This list is not exhaustive, [57][57]. For more on the history of Attorney General’s decisions, see generally Gonzales & Glen, supra note 53 (focusing on decisions issued in the Bush II and Obama administrations, with reference to additional decisions in the Reagan, Bush I, and Clinton administrations); Glen, supra note 51, at 726–63 (addressing the decisions issued during the Kennedy, Clinton, and Obama administrations). but it does establish the Attorney General as the final arbiter in immigration proceedings when necessary to take into account potential political ramifications of the decision. That it does not happen more often does not point to a lack of political accountability in immigration adjudication, but to the fact that that adjudication—unlike other areas of administrative law—rarely implicates questions that would place the need for political accountability front and center.

In the course of conducting these adjudications, the Attorney General also far more often than not (almost invariably, in fact) meets the conditions under the “deliberative democracy” conception of political accountability. When the Attorney General accepts a case on referral, briefing from the parties is normally contemplated, and immigration advocates have usually been invited to participate as amici, as well. [58][58]. See In re Castro-Tum, 27 I. & N. Dec. 187, 187 (Att’y Gen. 2018) (referring the case, directing briefing on specific issues, and inviting “interested amici to submit briefs”); In re A-B-, 27 I. & N. Dec. 227, 227 (Att’y Gen. 2018) (same); In re L-A-B-R-, 27 I. & N. Dec. 245, 245 (Att’y Gen. 2018) (same); In re M-G-G-, 27 I. & N. Dec. 469, 469 (Att’y Gen. 2018) (same); M-S-, 27 I. & N. Dec. 476, 476 (Att’y Gen. 2018) (same); In re Negusie, 27 I. & N. Dec. 481, 481 (Att’y Gen. 2018) (same); In re L-E-A-, 27 I. & N. Dec. 494, 494 (Att’y Gen. 2018) (same); In re Castillo-Perez, 27 I. & N. Dec. 495, 495 (Att’y Gen. 2018) (same); In re Thomas & Thompson, 27 I. & N. Dec. 556, 556 (Att’y Gen. 2019) (same); In re Reyes, 27 I. & N. Dec. 708, 708 (Att’y Gen. 2019) (same); In re A-M-R-C-, 28 I. & N. Dec. 7, 7 (Att’y Gen. 2020) (same). This is not a regulatory requirement, nor does every case elicit a request for responses from the parties. But in those cases where such briefs are solicited (the overwhelming majority), it is difficult to see why that public invitation and the briefing it may prompt is substantially less important from a participatory point of view than the publication of a proposed rule with its own invitation for comments. This process also has sufficient transparency. The public is not privy to the decision-making process of the Attorney General in the course of rendering his decision on a referred case, but neither is it privy to the consideration of submitted comments and what responses they may trigger from the interested agencies engaged in a rulemaking. In both cases, the public does see the final result—the Attorney General’s decision with arguments accepted or rejected and the reasoning for the relevant determinations, and the Final Rule with comments noted and the agencies’ response memorialized—but in neither is there any significant transparency concerning how the agency arrived at that final determination.

C. The Expertise of the Board of Immigration Appeals

The Professors also fault the agency for a lack of relevant “expertise” for Chevron purposes. [59][59]. See Wadhia & Walker, supra note 1, at 1215–24. Although there is some force to aspects of their argument—scientific knowledge and expertise may be entirely lacking in the immigration context, and technical expertise, too, may be less important than in other areas, other aspects seem incomplete or wrong. Here, I deal with two issues: first, the concept of “legislative expertise,” the expertise the agency obtains through its interactions with the legislative branch in the course of revising the statute, and which, in my (perhaps idiosyncratic) conception, the Board does possess; and second, “craft expertise,” the expertise the agency obtains through its focus on resolving issues under a particular statute, and which, given the case flows through the agency, the Board could be expected to have in spades. [60][60]. Although not a central focus of their argument, the Professors also argue that no “legal or policy expertise in immigration or foreign relations” is required to become a member of the Board. See Wadhia & Walker, supra note 1, at 1217–19. That is true in one sense, as a lack of such knowledge is not disqualifying, but untrue in another, as subject-matter expertise is an important part of the application process—a fact borne out by the complete job announcement that the Professors refer to only in part. See Immigration Judge, USAJobs, https://www.usajobs.gov/Get/Job?viewDetails/570894500 [https://perma.cc/ZU5H-49D7] (noting the requirement that applicants respond to the Quality Rating Factors, including “knowledge of immigration laws and procedures”). It is also worth noting that new hires are tested prior to the commencement of adjudicatory activities, ensuring there is a baseline of subject-matter specific expertise. See 8 C.F.R. § 1003.0(b)(1)(vi) (2021). Both Board members and immigration judges are required to undertake ongoing training “to promote the quality and consistency of adjudications.” See 8 C.F.R. §§ 1003.0(b)(1)(vii), 1003.1(a)(2)(i)(B) (2021). And throughout their time as adjudicators, Board members are assessed and evaluated to ensure they are properly discharging the functions of the office. See 8 C.F.R. §§ 1003.0(b)(1)(vi), 1003.1(a)(2)(i)(D) (2021).

First, the Professors argue that the Board lacks “[l]egislative [e]xpertise,” defined as “the expertise derived from the principal-agent relationship between Congress and the agency.” [61][61]. See Wadhia & Walker, supra note 1, at 1221. The Professors contend that “[t]his specialized knowledge of legislative purpose and process should only matter, from a Chevron-expertise perspective, if the agency statutory interpreter possesses that expertise—either directly because the interpreter helped draft the statute or indirectly because the interpreter interacts with the agency personnel who possess that expertise.” [62][62]. Id. at 1222. There is no question that Board members do not themselves assist with the drafting or review of legislation, and no reason to believe they have any interactions with agency personnel that do, so in one sense there is force to the Professors’ argument.

But the concept of “legislative expertise” is given too-narrow a definition in the Professors’ argument. In a more fundamental sense, the Board does have important insights into relevant legislative policies and purposes, because immigration enactments have largely unfolded in the past three to four decades as a complicated case of action and reaction, with the Board and courts of appeals constituting the “action” and Congress providing the “reaction.” The major amendments to the asylum statute made by the REAL ID Act of 2005, for instance, including new provisions relating to the burden of proof, credibility assessments, and corroboration requirements, were meant to codify existing Board standards as against contrary precedent in the Ninth Circuit. [63][63]. See H.R. Rep. No. 109-72, at 161–69 (2005) (Conf. Rep.). Now that Congress has looked to the Board itself for the statement of the proper standards, who is better than the Board to interpret any gaps or ambiguities left in those provisions? And, in fact, knowledge of this legislative background—the rules Congress wanted to reject and the policies it was interested in advancing—has informed the Board’s interpretation of these provisions, and produced exactly the type of decision that should qualify for deference under the terms of Chevron. [64][64]. See, e.g., Matter of L-A-C-, 26 I. & N. Dec. 516, 518–24 (BIA 2015) (interpreting the REAL ID Act’s corroboration requirement in light of its professed purpose of rejecting contrary court of appeals’ precedent and adopting the Board’s rule); Matter of C-T-L-, 25 I. & N. Dec. 341, 343–48 (BIA 2010) (applying a similar reasoning in regard to the REAL ID Act’s amendments to the statutory burden-of-proof provisions for asylum and withholding of removal).

The REAL ID Act is not an isolated instance, either. In the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA), [65][65]. Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, Div. C, 110 Stat. 3009-546 (1996) (codified as amended at 8 U.S.C. ch. 12). Congress made numerous changes to the INA premised on issues that arose through administrative and federal litigation, including: amending the definition of conviction to adopt certain aspects of the Board’s prevailing definition, while rejecting others; [66][66]. See H.R. Rep. No. 104-828, at 223–24 (1996) (Conf. Rep.) (citing Matter of Esposito, 21 I. & N. Dec. 1 (BIA 1995); Matter of Castro, 19 I. & N. Dec. 692 (BIA 1988); Matter of Ozkok, 19 I. & N. Dec. 546 (BIA 1988)). eliminating suspension of deportation and replacing it with cancellation of removal, including a heightened hardship standard deemed necessary because of lax application of the prior “extreme hardship” standard; [67][67]. H.R. Rep. No. 104-828, at 213–14 (citing Matter of O-J-O-, 21 I. & N. Dec. 381 (BIA 1996)); see Patrick J. Glen & Alanna R. Kennedy, The Strange and Unexpected Afterlife of Pereira v. Sessions, 34 Geo. Immigr. L.J. 1, 10–12 (2019) (recounting changes to the discretionary-relief regime made by IIRIRA). and clarifying certain issues relating to when a qualifying relationship must exist for purposes of certain waivers of inadmissibility. [68][68]. H.R. Rep. No. 104-828, at 228 (citing Matter of Farias, 21 I. & N. Dec. 269 (BIA 1996)); see Glen, supra note 51, at 750–51 (noting IIRIRA’s reversal of Matter of Farias); Gonzales & Glen, supra note 53, at 890 (same). In other words, the base-line for many of the shifts in IIRIRA was Board precedent, and thus the Board is well-placed to understand: 1) what the prior rule was; 2) how Congress altered that rule; and 3) what that alteration means for purposes of a permissible interpretation going forward. [69][69]. See, e.g., Matter of Cardenas-Abreu, 24 I. & N. Dec. 795, 796–802 (BIA 2009) (en banc) (interpreting IIRIRA’s statutory enactment of a definition for the term “convicted” in light of pre-1996 Board and federal court precedent, as well as legislative intent); Matter of Monreal-Aguinaga, 23 I. & N. Dec. 56, 58–63 (BIA 2001) (en banc) (interpreting the new “exceptional and extremely unusual hardship” requirement for cancellation of removal in light of the prior suspension-of-deportation standard and evident congressional intent in heightening the various eligibility criteria for discretionary relief from removal in IIRIRA). This of course tracks quite nicely with the concept of “legislative experience” offered by the authors. Their error, it seems to me, was reading that concept too narrowly to exclude the institutional knowledge the agency has by virtue of its central interpretive role in the relevant statutory scheme.

Second, the Professors posit the possibility that the Board may possess “craft expertise,” relying on the work of Professor Sidney Shapiro. [70][70]. See Wadhia & Walker, supra note 1, at 1217, 1223–24. Professor Shapiro advances a conception of “institutional expertise,” tied to the “unique wisdom of [the] regulatory agency,” and ultimately premised on “experience” rather than “formal knowledge.” [71][71]. Sidney A. Shapiro, The Failure to Understand Expertise in Administrative Law: The Problem and the Consequences, 50 Wake Forest L. Rev. 1097, 1099 (2015). With tens of thousands of appeals resolved each year, the Board could be expected to possess “craft expertise” in spades, and in fact, Professor Wadhia and Walker feint in that direction initially. [72][72]. See Wadhia & Walker, supra note 1, at 1223 (noting that BIA members become specialists on immigration law by interpreting INA and immigration policies. Compared to federal judges, the agency developed “expertise in reconciling and accounting for conflicting evidence and arguments, disciplinary perspectives, political demands, and legal commands.” (citation omitted)). Ultimately, they conclude that the issue of craft expertise does not weigh in the Board’s favor for two reasons: 1) statutory ambiguities in the INA rarely implicate the foundational issues of expertise relevant to the concept; and 2) even if the Board does possess some craft expertise, it is a lesser form of expertise compared to that of other agency actors in the rulemaking process. [73][73]. Id. at 1223–24.

Fairly construed, the calculus seems to weigh in the other direction. By dint of resolving tens of thousands of cases annually, presenting variations on a relatively firm set of themes, the Board does augment its understanding of the immigration laws generally, as well as how they specifically apply to certain circumstances. Considering the concept of “particular social group” in the refugee definition, [74][74]. Convention Relating to the Status of Refugees, Art. 1(A)(2), July 28, 1951, 189 U.N.T.S. 137. for instance, the Board may hone its understanding of the general concept as the issue is raised across a variety of circumstances, while also sharpening its application to specifics that may be presented over and over again. By having to resolve the question repeatedly, the Board’s expertise as a general and specific matter begins to far outstrip other actors in the bureaucracy. It may be true that some questions do not implicate technical or scientific expertise, but many, if not most, that are resolved through precedential decision-making will involve matters peculiar to the agency’s institutional mission, like the “particular social group” interpretation noted above. In cases where the Board issues a precedential decision, that decision is likely the end product of having considered the issue in hundreds, thousands, or even tens of thousands of other cases, with the precedent-setting decision marking the culmination of this expansive decisional process. Not only does the Board possess the institutional expertise posited by Professor Shapiro, because of the sheer breadth of its workload, but it is also difficult to imagine an agency that could claim greater “craft expertise.”

The comparative expertise argument also seems weak. The Professors do note that the focus of the Board on immigration issues and the stream of cases means that the agency possesses greater comparative expertise than the federal courts. [75][75]. See Wadhia & Walker, supra note 1, at 1223. But there is little reason to believe that the Board does not also possess greater “craft expertise” than other agency experts. Again tying this to work-flow, the Board will have seen a particular issue raised in virtually every possible circumstance in which it can be raised, giving it a more expansive view of the legal playing field than other experts tied solely to the language of the statute or legal imagination. Given the number of cases decided by immigration judges, and the possibilities of the full panoply of issues under the INA being appealed to the Board, it is the Board that has the most comprehensive view of immigration law of virtually any actor in the field. The number of appeals considered and resolved each year, and the range of issues presented therein, dwarf the scope of rulemaking. In other words, if there is institutional expertise borne of practice, the Board can comfortably claim that mantle in the immigration context.

III. Rulemaking versus Adjudication

The Professors’ arguments against Chevron deference to immigration adjudications are not free-standing attacks against the concept, but rather the framework they utilize to advance their main normative argument: that at least so far as immigration policy-making is concerned, the agencies should categorically prefer rulemaking to adjudication. [76][76]. See id. at 1202–03. For me, because I do not believe in the strength of their arguments against the applicability of Chevron deference to immigration adjudications, I remain at best agnostic about the choice between adjudication and rulemaking. But I also believe that, even on its own terms, the Professors’ argument relies on an idealized conception of rulemaking when, in reality, the purported gulf between the desirability of these options is substantially narrower (or non-existent).

At the threshold, there is no question—and the Professors do not argue to the contrary—that when given the choice to proceed via rulemaking or adjudication, the choice is entirely within the discretion of the agency. [77][77]. See, e.g., SEC v. Chenery Corp., 332 U.S. 194, 203 (1947) (“[T]he choice made between proceeding by general rule or by individual, ad hoc litigation is one that lies primarily in the informed discretion of the administrative agency.”); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 808 n.29 (1978) (“The Commission has substantial discretion as to whether to proceed by rulemaking or adjudication.” (citation omitted)). Likewise, under controlling precedent, deference is warranted in either case, so long as the agency is fulfilling its statutory mandate and Congress has not unambiguously foreclosed the interpretation or decision made by the agency. [78][78]. See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”) (emphasis added). The provision of this choice makes sense, as some issues may be amenable to either rulemaking or adjudication, for others, there are considerations that may weigh more heavily in favor of the one rather than the other, and for still others, the choice may be directed by statute or the nature of the policy question being resolved.

Using just the asylum statute as an example, all these dynamics can be seen at work. The Attorney General may, “by regulation,” “designate . . . offenses that will be considered to be” disqualifying particularly serious crimes or serious nonpolitical crimes for purposes of asylum eligibility; [79][79]. 8 U.S.C. § 1158(b)(2)(B)(ii) (emphasis added). he “may . . . establish additional limitations and conditions . . . under which an alien shall be ineligible for asylum”; [80][80]. 8 U.S.C. § 1158(b)(2)(C). and he “may provide . . . other conditions or limitations on the consideration of an application for asylum.” [81][81]. 8 U.S.C. § 1158(d)(5)(B). He also “shall establish a procedure for the consideration of asylum applications filed under” the statute. [82][82]. 8 U.S.C. § 1158(d)(1). These are issues that by their nature or by specific direction of the statute must be resolved via rulemaking. In contrast, the definition of “particular social group,” one of the statutorily protected grounds on which basis an alien may assert eligibility for asylum, is an issue that would be amenable to either rulemaking or adjudication. The Board has established broad criteria to consider in assessing whether a particular social group qualifies as such, [83][83]. See M-E-V-G-, 26 I. & N. Dec. 227, 227 (BIA 2014); W-G-R-, 26 I. & N. Dec. 208, 208 (BIA 2014). but it is possible that such criteria could have been promulgated via regulation, as well. No final rule on “particular social groups” has ever been promulgated, but the Clinton Administration did issue a draft rule that would have addressed some of these questions. [84][84]. See generally Asylum and Withholding Definitions, 65 Fed. Reg. 76,588, 76,593–95 (Dec. 7, 2000) (proposing standards to govern when an alleged particular social group is cognoscible under the INA). In contrast with the initial examples in this paragraph, however, there is nothing about resolving the ambiguity inherent in the term “particular social group” that makes rulemaking or adjudication the preferred course of action. Over the better part of four decades, the Board has, through adjudication, provided an interpretation of that term to guide determinations of when a “particular social group” may qualify under the statute, and has applied that definition in precedential decisions to render broadly applicable holdings regarding frequently raised “particular social groups.” [85][85]. See cases cited in supra note 83 (listing the Board’s most recent decisions on how and when a “particular social group” may be established, including historical overviews of the Board’s development of the “particular social group” analysis). The same goals could have been achieved through the rulemaking process, but that process would not have provided any better avenue for issuing Executive Branch policy on the question. Either forum allows complete consideration of the important issues and final resolution of the questions so that the law may be applied to future cases. In short, policy could be established through either mechanism. But then consider application of that generally promulgated framework to specific proposed social groups, whether it is victims of domestic abuse, the nuclear family, former gang members, or wealthy returning deportees. Rulemaking cannot exhaustively address each and every conceivable proposed social group, even if it could establish the parameters for considering when such a group could be recognized. [86][86]. See SEC v. Chenery Corp., 332 U.S. 194, 202 (1947) (explaining that not every principle can be immediately formed as a general rule because “[s]ome principles must await their own development, while others must be adjusted to meet particular, unforeseeable situations[,]” and therefore, concluding that “an administrative agency must be equipped to act either by general rule or by individual order. To insist upon one form of action to the exclusion of the other is to exalt form over necessity”). Instead, adjudication, by considering each question on a case-by-case basis, can provide greater form and context for interpretation of the ambiguous term, and in so doing should be entitled to deference consistent with the Supreme Court’s direction in cases such as Cardoza-Fonseca and Aguirre-Aguirre. [87][87]. See Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421, 448 (1987) (“There is obviously some ambiguity in a term like ‘well-founded fear’ which can only be given concrete meaning through a process of case-by-case adjudication. In that process . . . the courts must respect the interpretation of the agency . . . .” (internal quotation marks and citations omitted)); Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (adopting a similar reasoning).

Presenting rulemaking as an all-or-nothing option if an agency wants to obtain deference on review fails to take into account these differences inherent in the multitude of issues raised before and considered by agencies. At the very least, failing to meaningfully grapple with these questions undercuts the force of the Professors adamant turn to rulemaking in the immigration context. To be sure, on one level the question is not the choice itself but the result of that choice, and even there the Professors’ argument resorts to the extension of some deference (just not Chevron). [88][88]. See, e.g., Wadhia & Walker, supra note 1, at 1241. Regardless of these points, however, the end result is the same: for the agencies’ policy choices to receive the deference due to them under Chevron, they must proceed through rulemaking to the exclusion of adjudication.

Despite the foregoing, the Professors’ Manichean approach might make sense if rulemaking were some sort of panacea for all the ills they raise regarding immigration adjudication. On a clear-eyed view, though, it is not. The shortcomings of rulemaking may be different in kind or scope than those of adjudication, but they are shortcomings nonetheless. The concerns noted by the Professors ultimately provide no support for their aggressive argument in favor of rulemaking.

Returning, first, to the question of expertise; in arguing against relevant expertise by agency adjudicators, the Professors framed that perceived lack against the greater benefits of expertise offered by other individuals within the subject agencies. [89][89]. See id. at 1221–24 (focusing on the expertise of agency personnel directly involved with the drafting and consideration of legislation on behalf of the Executive Branch). The question of comparative expertise seems closer to a wash, however. There are undoubtedly other experts at EOIR, DHS, and DOJ, who have meaningful contributions to make in the course of rulemaking (as well as the antecedent step of statutory revision and drafting), but there is little reason to view any of these discrete experts as in possession of substantially more knowledge than the adjudicators and their adjutants. Regarding the aggregation of expertise, in the course of rulemaking the agencies may benefit from multiple different viewpoints being brought to bear on a single problem—EOIR may be able to explain adjudicative issues posed by the rule, DHS operational impact, and DOJ potential litigation fall-out. [90][90]. See id. at 1222. Here, the benefits of rulemaking in the abstract may seem formidable, but with any precedential decision the Board will have the views of its own expert members, the staff of EOIR who work for the Board, and the views of DHS via briefing and possibly argument. The slate of experts available to the Board is certainly less than what would be available to the administration via rulemaking, but this should also be considered within the scope of the decisions being rendered. As explained above, issues that are peculiarly subject to adjudication are also substantially less likely to benefit from the kind of aggregation of expertise encouraged by the rulemaking process. Interpreting the term “particular social group” and applying that definition in specific cases as aliens proffer a range of purported social groups is of a different kind of exercise than promulgating procedures to guide the filing and consideration of asylum applications. The former issues present questions of adjudicatory expertise that are dependent on the Board’s consistent engagement with both the general statute and specific questions that arise under the statute, whereas the latter issue presents more purely institutional questions where the aggregation of agency expertise may likely contribute to the development of an optimal regulation. The aggregation issue arises only through posing a false equivalency between issues considered in rulemaking and those considered in adjudication.

The nature of expertise is also a tricky concept in this area. As I argue above, I believe that the Board has substantial institutional expertise that arises because of the impetus behind much of Congress’s reforms over the past decades. [91][91]. See supra Part II.C. Those reforms have occurred with Board precedent in mind, and thus the Board is peculiarly well-placed to understand what Congress was doing and why. I also agree with the Professors, however, that others within the immigration bureaucracy may have important contributions to make regarding the same question. [92][92]. Wadhia & Walker, supra note 1, at 1222. The Professors note in this regard the Board’s inability to access this expertise because of the ethical and practical barriers that often exist between the agency adjudicator and other government policy-makers, [93][93]. Id. and use that fact as an argument against deference to the Board itself. [94][94]. Id. at 1222–23. But if that wall is blocking a flow of information, it is blocking that flow from both directions. The Board may not have access to other experts’ views, but then those same experts do not have access to the Board’s own expert views and what could be its singular insights into the motivating factors of statutory amendments. Where the Professors see only a shortcoming insofar as the Board itself is concerned, I see a two-way street whereby experts engaged in rulemaking have only a limited universe of expertise to consider, while the Board itself has only a slightly differently constituted limited universe of expertise to consider. Both procedures lack something, and given that—along with my more generally supportive views of Board expertise in the course of adjudication—I can find no expertise-based reason to categorically prefer rulemaking while withdrawing adjudication from the scope of Chevron deference.

Political accountability also seems more or less a wash. Given the possibilities of Attorney General referral, adjudication via that avenue provides as much political accountability as does Executive agency rulemaking. [95][95]. See supra Part II.B. In both cases, voters know where the buck has stopped. I also disagree that rulemaking is inherently preferable from the perspective of public participation. Here, the Professors are quite bullish on the benefits of rulemaking, which seem superficially clear: a rule is proposed, the public is invited to review and comment, and the administration then considers the comments (sometimes in the hundreds of thousands) individually (!) and drafts a final rule taking into consideration all these concerns and comments. The notion that this process is actually engaged in soliciting and considering public views to the ends of altering the agencies’ preferred rules is at best in tension with reality. Writing in an earlier version of the Duke Administrative Law Symposium, Professor E. Donald Elliott, formerly a Public Member of the Administrative Conference of the United States and Assistant Administrator and General Counsel of the Environmental Protection Agency, remarked that “[n]o administrator in Washington turns to full-scale notice-and-comment rulemaking when she is genuinely interested in obtaining input from interested parties.” [96][96]. E. Donald Elliot, Re-inventing Rulemaking, 41 Duke L.J. 1490, 1492 (1992); see David J. Barron & Elena Kagan, Chevron’s Nondelegation Doctrine, 2001 Sup. Ct. Rev. 201, 231–32 (2001) (“The more courts have required agencies to give detailed notice of proposed regulatory action to interest groups, the more pressure agencies have felt to complete the bulk of their work prior to the onset of the rulemaking process . . . the less flexibility they show during rulemaking.”); see also Wendy Wagner, Katherine Barnes & Lisa Peters, Rulemaking in the Shade: An Empirical Study of EPA’s Air Toxic Emission Standards, 63 Admin. L. Rev. 99, 123 (2011) (“The findings of this study [on the EPA’s adoption of the air toxic emissions standards] underscore both the accuracy and the importance of Professor Elliott’s remarks . . . [it] reinforce[s] the possibility that a great deal of interest group influence occurs outside of the glass box of notice and comment.”). No doubt informed by his own personal experience, Professor Elliott offered a telling simile: “Notice-and-comment rulemaking is to public participation as Japanese Kabuki theater is to human passions—a highly stylized process for displaying in a formal way the essence of something which in real life takes place in other venues.” [97][97]. Elliot, supra note 96, at 1492; see Stephanie Stern, Cognitive Consistency: Theory Maintenance and Administrative Rulemaking, 63 Pitt. L. Rev. 589, 597 (2002) (“The timing of rulemaking encourages agency lock-in by concentrating the bulk of decisionmaking in the pre-notice period. Notice occurs after the agency has completed substantial amounts of development, analysis, and review.”); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1669, 1775 (1975) (“Indeed, the content of rulemaking decisions is often largely determined in advance through a process of informal consultation in which organized interests may enjoy a preponderant influence.”). This is not to say that rulemaking is not informed by public concerns as well as interested constituents, but only that that process is not likely to occur within the confines of rulemaking itself. [98][98]. Elliot, supra note 96, at 1492–93 (listing several methods to secure public participation: “[F]rom informal meetings with trade associations and other constituency groups, to roundtables, to floating ‘trial balloons’ in speeches or leaks to the trade press, to the more formal techniques of advisory committees and negotiated rulemaking”). The notice-and-comment process is now quite formal and driven by compliance with statutory standards, and the bare-fact of meeting those standards (while also compiling the record on which the rule will be defended in the federal courts)—not “provid[ing] . . . public input into government thinking—is the “primary[]” function of the process. [99][99]. Id. at 1493; see Jud Mathews, Minimally Democratic Administrative Law, 68 Admin. L. Rev. 605, 636 (2016) (noting divergence between “the justifications that they will believe satisfy courts” and “the agency’s real reasons for acting” and concluding that “demanding judicial review standards contribute to the ossification of the rulemaking process, generating delays and reducing agencies’ functional rulemaking capacity”).

A comparison of those parties submitting comments to the prior administration in the immigration rulemaking context would, I would wager, align highly with the interested amici submitting briefs before the Attorney General in referred cases. Professor Wadhia specifically has been an active amicus participant in immigration cases before the Supreme Court, federal courts of appeals, and district courts. [100][100]. See, e.g., Shoba Sivaprasad Wadhia, Symposium: Reflections on the Travel Ban Decision, SCOTUSblog (June 26, 2018, 5:02 PM), https://www.scotusblog.com/2018/06/symposium-reflections-on-the-travel-ban-decision [https://perma.cc/WL5S-VS9W] (noting that Professor Wadhia co-authored an amicus brief for scholars of immigration law on the history of the Immigration and Nationality Act in support of Hawaii in Trump v. Hawaii, 138 S.Ct. 2392 (2018)). Why is the latter form of participation dramatically different from the former? The Professors do not pose this question and thus do not answer it, but it is worth considering. If the essence of the public-participation argument is the ability to be heard, the opportunities are similar as between rulemaking and adjudication in that narrow class of case that will prompt a decision entitled to deference on review. The opportunities are not identical, and I do not mean to argue to the contrary—the Board and Attorney General may decide cases without amicus participation or additional briefing from the parties. But fairly considered, the opportunities to participate in those adjudications that will lead to a precedential decision are more robust and important than the Professors acknowledge, while their own preferred course of rulemaking contains only a formalized mechanism of public participation whose actual substantive importance is open to debate.

All this is to say only that rulemaking has its own warts. It is a necessary mechanism for adopting certain policies, and the route that should be preferred in other classes of cases, as well. But it is not without its own shortcomings, including as to expertise, accountability, and participation, and the Professors fail to advance any compelling reason for categorically preferring rulemaking to adjudication in the advancement of administration policy, much less an argument for why we should continue to confer deference on rulemakings while withdrawing it from adjudications.

IV. Responsibility for and the Likely Outcome of Chevron’s Recalibration

This leaves possibilities for reform—assuming Chevron should be recalibrated for purposes of immigration adjudication, how should that recalibration be accomplished? First, if Congress wants to eliminate deference for all administrative cases or any subset thereof, there is no obvious impediment to that action, although I share the Professors’ skepticism that any substantial reform to deference principles will come through legislation. [101][101]. See Wadhia & Walker, supra note 1, at 1236. The Professors do note the possibility of comprehensive immigration reform in the Biden Administration, and this well may provide an avenue to action on the specific question of deference. But it additionally opens the door to other ways to cut back on the circumstances where such deference is relevant—revising the INA or enacting new provisions that more particularly and explicitly address the relevant questions rather than more open-ended provisions leaving interpretive discretion with the agency. Eliminating or cutting back on the chances an agency would have to render an interpretation ultimately entitled to deference before the federal courts is as wise a move towards “mitigating” deference’s reign as would be a wholesale revocation of deference.

Second, perhaps most provocatively, the Professors argue for the Executive Branch to take the lead by waiving deference in immigration adjudications and “shifting major immigration policymaking away from adjudication and into the realm of notice-and-comment rulemaking.” [102][102]. Id. at 1241. For support, they note a recent regulation limiting the circumstances where the Department of Justice will seek deference in the wake of Kisor v. Wilkie’s [103][103]. Kisor v. Wilkie, 139 S. Ct. 2400 (2019). reconceptualization of Auer deference. [104][104]. Wadhia & Walker, supra note 1, at 1241 (citing Processes and Procedures for Issuance and Use of Guidance Documents, 85 Fed. Reg. 63,200, 63200–04 (Oct. 7, 2020) (to be codified at 28 C.F.R. pt. 50)). I would assume there is no barrier to the Department also doing so in the context of Chevron deference and defense of immigration adjudications. But the Professors miss an important point in citing this regulation—the Department sought to ensure that its litigation unfolded consistent with governing Supreme Court law, and did not push the boundaries by asking for deference to guidance and internal documents that otherwise would not warrant deference. That argument is entirely lacking here—the Supreme Court has consistently and emphatically described immigration adjudication as entitled to deference on judicial review. In promulgating any contrary Departmental policy or rule, the Department would not be acting to ensure compliance with Supreme Court law, but would be itself catalyzing a shift away from that law.

This also implicates attorney-client obligations. The Department’s lawyers have higher obligations and duties regarding the rule of law than private practitioners. [105][105]. See, e.g., Berger v. United States, 295 U.S. 78, 88 (1935) (“The United States Attorney is the representative not of an ordinary party to a controversy, but of a sovereignty . . . and whose interest, therefore, in a criminal prosecution is not that it shall win a case, but that justice shall be done.”); United States v. Young, 470 U.S. 1, 25 (1985) (Brennan, J., concurring in part and dissenting in part) (noting that the Supreme Court has “long emphasized that a representative of the United States Government is held to a higher standard of behavior” than private counsel) (alteration in original). This means that in certain circumstances, the Department will decline to defend otherwise favorable precedent or may confess error. [106][106]. This obviously happens infrequently, and when it does happen the issue is usually not a substantive interpretation of the agency but a jurisdictional determination by the court of appeals. See Mata v. Lynch, 576 U.S. 143, 147 (2015) (noting government’s agreement with the petitioner contra the holding of the court of appeals); Kucana v. Holder, 558 U.S. 233, 241–42 (2010) (same); see also Abdisalan v. Holder, 774 F.3d 517, 523 (9th Cir. 2014) (en banc) (noting the confluence of the petitioner’s and government’s arguments on rehearing). Along with this obligation, however, is the Department’s obligation to defend its client-agencies within the limits permitted by the law. [107][107]. See generally Mark B. Stern & Alisa B. Klein, The Government’s Litigator: Taking Clients Seriously, 52 Admin. L. Rev. 1409 (2000) (noting that in representing agencies in the federal courts, Justice Department litigators are effectively representing “clients”). The limits of the law obviously entail not stretching the text of statutes, regulations, or judicial precedent beyond their reasonable import, but there can be no colorable argument that those limits also entail voluntarily declining to seek deference in a class of adjudications that the Supreme Court has repeatedly told us warrant deference. If Chevron continues to exist, and its application to immigration adjudications continues to be sanctioned, there is no legal rationale to support the Department’s waiver of deference, and any institutional interests are purely hypothetical. [108][108]. See, e.g., Wadhia & Walker, supra note 1, at 1241–42.

Most importantly, however, voluntary abdication of deference takes the burden off the entity that should be charged with rethinking the scope of Chevron: the Supreme Court itself. Deference, as currently conceptualized, stems from Supreme Court precedent and is a canon applicable to judicial review of agency action. It is thus with the judiciary that the mantle of reform must ultimately rest. Unless or until the Court seeks to alter the deference calculus, the Executive is entitled to continue to argue for deference consistent with that precedent. And, of course, this may already be happening in immigration cases. Former Justice Kennedy criticized the courts of appeals reflexive recourse to deference in a recent immigration case, where Justice Alito, in his solo dissent, argued that the majority’s decision was an implicit repudiation of Chevron deference. [109][109]. See Pereira v. Sessions, 138 S. Ct. 2105, 2120–21 (2018) (Kennedy, J., concurring). Justice Alito stated in his dissent:
Here, a straightforward application of Chevron requires us to accept the Government’s construction of the provision at issue. But the Court rejects the Government’s interpretation in favor of one that it regards as the best reading of the statute. I can only conclude that the Court, for whatever reason, is simply ignoring Chevron.
Id. at 2121 (Alito, J., dissenting).
The Court has also declined to take a position on whether deference is warranted to immigration adjudications that touch on so-called dual-use statutes, statutes that may entail both civil and criminal consequences, instead of conducting de novo review of the question presented. [110][110]. See Patrick J. Glen & Kate E. Stillman, Chevron Deference or the Rule of Lenity? Dual-Use Statutes and Judge Sutton’s Lonely Lament, 77 Ohio St. L.J. Furthermore 129, 133–40 (2016). Of course, in the midst of these cases, the Court has continued to apply Chevron deference. [111][111]. See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56–64 (2014) (applying deference to uphold the Board’s restrictive interpretation of the Child Status Protection Act); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (applying deference to uphold the Board’s interpretation of the continuous residence provision of the cancellation-of-removal statute). But to the extent there is an increasing number of cases where the Court balks at that application, any necessary recalibration is likely already underway.

It is also important to address a question the Professors do not raise—what is the practical effect if immigration adjudications no longer qualify for Chevron deference? Presumably, the Professors would argue that the elimination of deference would be of significant importance, otherwise there would be little reason to spend 47 pages arguing for that result in a prestigious law journal. I am skeptical. In the Supreme Court itself, I see little reason to believe that a lack of deference will negatively affect the rate at which the government prevails. First, again, many cases are already resolved without recourse to Chevron. Second, the cases where Chevron was applied were not likely to have come out the other way absent a robust conception of deference. The recently-concluded Supreme Court Term offers some support for these points. The Court decided five civil immigration cases, three of which presented some claim to Chevron deference. [112][112]. See Johnson v. Guzman Chavez, 141 S. Ct. 2271, 2291 (2021); Sanchez v. Mayorkas, 141 S. Ct. 1809, 1812 (2021); Niz-Chavez v. Garland, 141 S. Ct. 1474, 1480 (2021). The Court resolved all five on plain-text grounds, and the government prevailed in four of the five cases, including two of the three that presented the deference question. [113][113]. Of the cases presenting some form of deference, the government prevailed in Guzman Chavez, 141 S. Ct. 2271 and Sanchez, 141 S. Ct. 1809, and lost in Niz-Chavez, 141 S. Ct. 1474. The government additionally prevailed in Garland v. Dai, 141 S. Ct. 1669 (2021), and Pereida v. Wilkinson, 141 S. Ct. 754 (2021). Folding in an additional criminal-immigration case decided by the Court, and the government had a 5-1 record in immigration cases in the October 2020 Term. [114][114]. See United States v. Palomar-Santiago, 141 S. Ct. 1615, 1618 (2021). A more robust form of deference may well have led to a win for the government in the sole case it lost this Term. [115][115]. See Niz-Chavez, 141 S. Ct. at 1485. But that fact does not undermine the point that plain-text resolutions of immigration issues are often, even if not invariably, going to coincide with the disposition of the agency.

The issue may be more complicated in the courts of appeals, but the Professors certainly do not make any substantial case for a sea-change in judicial review of immigration adjudications in a hypothetical post-Chevron world. The points regarding the Supreme Court are likely to be just as important in the courts of appeals, i.e., many cases are already disposed of on non-deference-related grounds, while a “better-reading” argument will often favor the position advanced by the government even in the absence of Chevron deference (whether on de novo review or on application of Skidmore “deference”). Of course, this may just provide fuel for their argument—if the outcome will not have catastrophic consequences on the adjudicatory system, why not jettison deference and give a freer hand to the courts of appeals? That would be a rational counterpunch. For me, however, it argues for maintenance of the status quo, especially in these circumstances where I think most of the criticisms of deference lack strength and the proposed alternative of rulemaking offers no net benefits.

Conclusion

Professors Wadhia and Walker have written a thoughtful critique of the application of Chevron deference to immigration adjudications, but for me it ultimately misses the mark. In some sense, I have a dog in this fight, although my interests are more institutional than substantive. The actual practice of deference in immigration litigation has not engendered any significant problems in the four decades since Chevron, and unreasonable decisions or those prohibited by the plain language of the statute will be vacated on review. But where there is ambiguity or room for the agency to make a policy choice, it is entitled to pursue that decision through adjudication. And having done so, the courts should defer under the long-accepted principles of Chevron.


Copyright © 2021 Patrick J. Glen.

Adjunct Professor of Law, Georgetown University Law Center; Senior Litigation Counsel, United States Department of Justice, Civil Division, Office of Immigration Litigation. The author would like to thank Chris Walker and Smita Ghosh for their helpful comments on an initial draft of this Article. The views and opinions expressed herein are the author’s own, and do not represent those of the Department of Justice or any component thereof.

[1] Shoba Sivaprasad Wadhia & Christopher J. Walker, The Case Against Chevron Deference in Immigration Adjudication, 70 Duke L.J. 1197 (2021). The removal proceeding is the proceeding held before an immigration judge to determine a noncitizen’s removability from the United States, as well as any eligibility for relief or protection from removal. See 8 U.S.C. § 1229a (2018) (providing the statutory authority and framework for the conduct of removal proceedings); see also Patrick J. Glen, Judulang v. Holder and the Future of 212(c) Relief, 27 Geo. Immigr. L.J. 1, 4 (2012) (providing a historical overview of removal proceedings). The immigration judge conducts the removal proceeding in the first instance, see 8 U.S.C. § 1229a(a)(1), with the Board of Immigration Appeals exercising appellate jurisdiction over the decisions of immigration judges, see 8 C.F.R. § 1003.1(b)(3) (2021) (providing the bases for appellate jurisdiction).

[2] See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56 (2014) (applying Chevron deference to uphold a decision of the Board); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (same).

[3] See, e.g., Amaya v. Rosen, 986 F.3d 424, 438 n.1 (4th Cir. 2021) (Richardson, J., dissenting) (noting the article and its argument that Chevron deference should not apply to immigration adjudications, while opining that Chevron must still be applied as a matter of precedent).

[4] See Wadhia & Walker, supra note 1, at 1204–13.

[5] See Executive Office for Immigration Review (“EOIR”) Immigration Court Listing, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/immigration-court-administrative-control-list [https://perma.cc/3LD5-KVPW] (last updated Aug. 2, 2021); U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Immigration Judge (IJ) Hiring (2021), https://www.justice.gov/eoir/page/file/1242156/download [https://perma.cc/8HQB-B57N].

[6] See Board of Immigration Appeals, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/board-of-immigration-appeals-bios [https://perma.cc/5HPB-FZX2] (last updated July 19, 2021).

[7] See U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Pending Cases, New Cases, and Total Completions (2021), https://www.justice.gov/eoir/page/file/1242166/download [https://perma.cc/9QMQ-4N3X].

[8] EOIR, Adjudication Statistics, Case Appeals Filed, Completed, and Pending, U.S. Dep’t Just. Exec. Off. For Immigr. Rev., https://www.justice.gov/eoir/page/file/1248501/download [https://perma.cc/3QRT-7WYZ] (last updated July 8, 2021).

[9] See Wadhia & Walker, supra note 1, at 1228–30.

[10] See id.

[11] Flores-Molina v. Sessions, 850 F.3d 1150, 1157 (10th Cir. 2017) (quoting Rangel-Perez v. Lynch, 816 F.3d 591, 597 (10th Cir. 2017)).

[12] See 8 C.F.R. § 1003.1(g) (2021).

[13] See Agency Decisions, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/ag-bia-decisions [https://perma.cc/BJ8H-JXYQ] (listing the Board’s precedential decisions for each volume) (last updated Sept. 8, 2021).

[14] See, e.g., Matter of Nemis, 28 I. & N. Dec. 250 (BIA 2021) (adjudicating application of the categorical approach to the federal conspiracy statute, 18 U.S.C. § 371, and a visa-fraud provision, 18 U.S.C. § 1546(a)); Matter of Dikhtyar, 28 I. & N. Dec. 214, 214–16 (BIA 2021) (addressing the divisibility of the Utah state drug schedules by specific controlled substance); Matter of P-B-B-, 28 I. & N. Dec. 43 (BIA 2020) (addressing the divisibility with regard to Arizona’s criminalization of possession of a “dangerous drug”).

[15] See, e.g., Omargharib v. Holder, 775 F.3d 192, 196 (4th Cir. 2014) (“Although we generally defer to the BIA’s interpretations of the INA, where, as here, the BIA construes statutes [and state law] over which it has no particular expertise, its interpretations are not entitled to deference.” (alteration in original) (quoting Karimi v. Holder, 715 F.3d 561, 566 (4th Cir. 2013))); Ramos v. U.S. Att’y Gen., 709 F.3d 1066, 1069 n.2 (11th Cir. 2013) (“We owe no Chevron deference to the Board’s interpretation of the Georgia [criminal] statute, which the Board has no power to administer.”); Denis v. U.S. Att’y Gen., 633 F.3d 201, 208 (3d Cir. 2011) (“[I]f the issue turns on the meaning of a federal statute other than the INA, we possess the requisite expertise to interpret a federal criminal statute such that no deference is due.” (citation omitted)); see also Almeida v. Holder, 588 F.3d 778, 785 (2d Cir. 2009) (deferring to the BIA’s interpretation of “theft offense,” but reviewing questions of a particular crime defined by state law de novo).

[16] See 8 C.F.R. § 1003.1(g) (2021).

[17] For March 15, 2021, for instance, the Ninth Circuit issued 21 unpublished decisions and 3 published decisions. See Log of Unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021); Opinions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/opinions [https://perma.cc/8ST8-G668] (last visited Oct. 7, 2021). On March 12, 2021, when it issued no published decisions, it issued 13 unpublished decisions. See Log of unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021).

[18] 8 U.S.C. § 1103(g)(2).

[19] 8 U.S.C. § 1103(a)(1).

[20] United States v. Mead Corp., 533 U.S. 218, 229 (2001).

[21] See 8 U.S.C. § 1158(d)(1) (“The Attorney General shall establish a procedure for the consideration of asylum applications filed under [this section].”); 8 U.S.C. § 1158(d)(5)(B) (“The Attorney General may provide by regulation for any other conditions or limitations on the consideration of an application for asylum not inconsistent with this chapter.”); 8 U.S.C. § 1158(b)(2)(B)(ii) (“The Attorney General may designate by regulation offenses that will be considered to be a [disqualifying crime for purposes of asylum eligibility].”); 8 U.S.C. § 1158(b)(2)(C) (“The Attorney General may by regulation establish additional limitations and conditions, consistent with this section, under which an alien shall be ineligible for asylum under paragraph (1).”). See generally Patrick J. Glen, In re L-A-C-: A Pragmatic Approach to the Burden of Proof and Corroborating Evidence in Asylum Proceedings, 35 Geo. Immigr. L.J. 1, 34 (2020) (“The asylum statute . . . does not generally mandate specific procedures for the consideration of asylum applications. Rather, that section is composed almost entirely of explicit and implicit delegations.”).

[22] See 8 U.S.C. §§ 1182(a)(4), 1227(a)(5) (stating public charge ground of inadmissibility and deportability, respectively); see also Field Guidance on Deportability and Inadmissibility on Public Charge Grounds, 64 Fed. Reg. 28,689, 28,691 (Mar. 26, 1999) (providing public guidance as to how the Immigration and Naturalization Service would apply the public charge grounds of removability); Inadmissibility on Public Charge Grounds, 84 Fed. Reg. 41,292, 41,315 (Aug. 14, 2019) (revising the Executive Branch’s interpretation of the public charge inadmissibility ground).

[23] See 8 U.S.C. § 1101(f) (listing per se categories that would involve a finding that the alien lacked good moral character, but allow the Attorney General discretion to conclude that aliens outside those categories also lack good moral character); see, e.g., 8 U.S.C. § 1229b(b)(1)(B) (requiring good moral character in order to establish eligibility for cancellation of removal).

[24] See 8 U.S.C. § 1227(a)(2)(E)(i); see also Matter of Velazquez-Herrera, 24 I. & N. Dec. 503, 504 (BIA 2008) (interpreting this provision).

[25] Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421 (1987).

[26] Id. at 448 (internal quotation marks and citations omitted).

[27] Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 424–25 (1999).

[28] 8 U.S.C. § 1103(a)(1).

[29] Aguirre-Aguirre, 526 U.S. at 424.

[30] Id. at 424–25.

[31] Id. at 425 (quoting Immigr. & Naturalization Serv. v. Abudu, 485 U.S. 94, 110 (1988)).

[32] See Patrick J. Glen, Interring the Immigration Rule of Lenity, 99 Neb. L. Rev. 533, 560–61 n.191 (2021) (noting that “[t]he statute is . . . replete with express delegations of decisional authority to the Attorney General,” and citing examples relating to asylum, cancellation of removal, and other forms of relief).

[33] See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”).

[34] United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 542 (1950) (emphasis added).

[35] Fiallo v. Bell, 430 U.S. 787 (1977).

[36] Id. at 792 (citation omitted).

[37] Hampton v. Mow Sun Wong, 426 U.S. 88 (1976).

[38] Id. at 101 n.21 (citation omitted).

[39] Scialabba v. Cuellar de Osorio, 573 U.S. 41, 75 (2014) (plurality opinion).

[40] See Negusie v. Holder, 555 U.S. 511, 523 (2009) (explaining that remand rules exist, in part, “because ‘ambiguities in statutes within an agency’s jurisdiction to administer are delegations of authority to the agency to fill the statutory gap in reasonable fashion. Filling these gaps . . . involves difficult policy choices that agencies are better equipped to make than courts.’” (quoting Nat’l Cable & Telecomm. Ass’n. v. Brand X Int. Serv., 545 U.S. 967, 980 (2005)). Justice Kennedy also noted that on remand,

[t]he agency’s interpretation of the statutory meaning of “persecution” may be explained by a more comprehensive definition, one designed to elaborate on the term in anticipation of a wide range of potential conduct; and that expanded definition in turn may be influenced by how practical, or impractical, the standard would be in terms of its application to specific cases. These matters may have relevance in determining whether its statutory interpretation is a permissible one.

Id. at 524; see also Gonzales v. Thomas, 547 U.S. 183, 186–87 (2006) (per curiam) (noting that on remand, “[t]he agency can bring its expertise to bear upon the matter; it can evaluate the evidence; it can make an initial determination . . . it can, through informed discussion and analysis, help a court later determine whether its decision exceeds the leeway that the law provides” (quoting Immigr. & Naturalization Serv. v. Ventura, 537 U.S. 12, 17 (2002) (per curiam))).

[41] See Wadhia & Walker, supra note 1, at 1231–32 (citations omitted).

[42] See id.

[43] See 8 U.S.C. § 1103(a), (g).

[44] See 8 C.F.R. § 1003.1(d)(1) (1958) (“The Board shall function as an appellate body charged with the review of those administrative adjudications under the Act that the Attorney General may by regulation assign to it.”); see also Delegation of Powers and Definition of Duties, Immigration and Naturalization Service, 5 Fed. Reg. 2454, 2454 (July 1, 1940) (“[T]he Board of Review of the Immigration and Naturalization Service shall have authority to exercise the powers of the Attorney General [in certain delineated cases].”).

[45] See 8 C.F.R. § 1003.1(d)(1)(ii) (“Board members shall exercise their independent judgment and discretion in considering and determining the cases coming before the Board.”).

[46] See Tefel v. Reno, 972 F. Supp. 608, 613 n.1 (S.D. Fl. 1997) (“[T]he decision of the BIA is not factually, nor legally, the decision of the Attorney General.”).

[47] See United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260, 266–67 (1954) (“[T]he word ‘discretion’ means that the recipient must exercise his authority according to his own understanding and conscience. In short . . . the Attorney General denies himself the right to sidestep the Board or dictate its decision in any manner.”).

[48] See 8 C.F.R. § 1003.1(h).

[49] Sanchez-Penunuri v. Longshore, 7 F. Supp. 2d 1136, 1149 (D. Colo. 2013).

[50] See Wadhia & Walker, supra note 1, at 1232.

[51] See Patrick J. Glen, Robert F. Kennedy and the Attorney General’s Referral Authority: A Blueprint for the Biden Administration, 49 Hofstra L. Rev. 719, 721 (2021) (noting how the Trump Administration utilized Attorney General referral to advance an immigration policy consonant with its campaign promises, including restricting eligibility for asylum, streamlining enforcement and adjudications to limit the possibilities of delay in removal proceedings, and focusing on detention of aliens both during and after the conclusion of removal proceedings).

[52] See generally id. (discussing Biden administration’s policy).

[53] See Deportation Proceedings for Joseph Patrick Thomas Doherty, 13 Op. O.L.C. 1, 8 (1989); Deportation Proceedings of Joseph Patrick Thomas Doherty, 12 Op. O.L.C. 1, 8 (1988); see also Alberto R. Gonzales & Patrick Glen, Advancing Executive Branch Immigration Policy Through the Attorney General’s Review Authority, 101 Iowa L. Rev. 841, 882–86 (2016) (providing background and context to the Attorney Generals’ decisions in Matter of Doherty).

[54] See R-A-, 22 I. & N. Dec. 906, 906 (Att’y Gen. 2001); see also Gonzales & Glen, supra note 53, at 886–89 (providing background and context to the Attorney General’s decision in Matter of R-A-).

[55] See A-T-, 24 I. & N. Dec. 617, 617 (Att’y Gen. 2008); see also Gonzales & Glen, supra note 53, at 861–63 (providing background and context to the Attorney General’s decision in Matter of A-T-).

[56] See Dorman, 25 I. & N. Dec. 485, 485 (Att’y Gen. 2011); see also Glen, supra note 51, at 760–71 (providing background and context to the Attorney General’s decision in Matter of Dorman).

[57] For more on the history of Attorney General’s decisions, see generally Gonzales & Glen, supra note 53 (focusing on decisions issued in the Bush II and Obama administrations, with reference to additional decisions in the Reagan, Bush I, and Clinton administrations); Glen, supra note 51, at 726–63 (addressing the decisions issued during the Kennedy, Clinton, and Obama administrations).

[58] See In re Castro-Tum, 27 I. & N. Dec. 187, 187 (Att’y Gen. 2018) (referring the case, directing briefing on specific issues, and inviting “interested amici to submit briefs”); In re A-B-, 27 I. & N. Dec. 227, 227 (Att’y Gen. 2018) (same); In re L-A-B-R-, 27 I. & N. Dec. 245, 245 (Att’y Gen. 2018) (same); In re M-G-G-, 27 I. & N. Dec. 469, 469 (Att’y Gen. 2018) (same); M-S-, 27 I. & N. Dec. 476, 476 (Att’y Gen. 2018) (same); In re Negusie, 27 I. & N. Dec. 481, 481 (Att’y Gen. 2018) (same); In re L-E-A-, 27 I. & N. Dec. 494, 494 (Att’y Gen. 2018) (same); In re Castillo-Perez, 27 I. & N. Dec. 495, 495 (Att’y Gen. 2018) (same); In re Thomas & Thompson, 27 I. & N. Dec. 556, 556 (Att’y Gen. 2019) (same); In re Reyes, 27 I. & N. Dec. 708, 708 (Att’y Gen. 2019) (same); In re A-M-R-C-, 28 I. & N. Dec. 7, 7 (Att’y Gen. 2020) (same).

[59] See Wadhia & Walker, supra note 1, at 1215–24.

[60] Although not a central focus of their argument, the Professors also argue that no “legal or policy expertise in immigration or foreign relations” is required to become a member of the Board. See Wadhia & Walker, supra note 1, at 1217–19. That is true in one sense, as a lack of such knowledge is not disqualifying, but untrue in another, as subject-matter expertise is an important part of the application process—a fact borne out by the complete job announcement that the Professors refer to only in part. See Immigration Judge, USAJobs, https://www.usajobs.gov/Get/Job?viewDetails/570894500 [https://perma.cc/ZU5H-49D7] (noting the requirement that applicants respond to the Quality Rating Factors, including “knowledge of immigration laws and procedures”). It is also worth noting that new hires are tested prior to the commencement of adjudicatory activities, ensuring there is a baseline of subject-matter specific expertise. See 8 C.F.R. § 1003.0(b)(1)(vi) (2021). Both Board members and immigration judges are required to undertake ongoing training “to promote the quality and consistency of adjudications.” See 8 C.F.R. §§ 1003.0(b)(1)(vii), 1003.1(a)(2)(i)(B) (2021). And throughout their time as adjudicators, Board members are assessed and evaluated to ensure they are properly discharging the functions of the office. See 8 C.F.R. §§ 1003.0(b)(1)(vi), 1003.1(a)(2)(i)(D) (2021).

[61] See Wadhia & Walker, supra note 1, at 1221.

[62] Id. at 1222.

[63] See H.R. Rep. No. 109-72, at 161–69 (2005) (Conf. Rep.).

[64] See, e.g., Matter of L-A-C-, 26 I. & N. Dec. 516, 518–24 (BIA 2015) (interpreting the REAL ID Act’s corroboration requirement in light of its professed purpose of rejecting contrary court of appeals’ precedent and adopting the Board’s rule); Matter of C-T-L-, 25 I. & N. Dec. 341, 343–48 (BIA 2010) (applying a similar reasoning in regard to the REAL ID Act’s amendments to the statutory burden-of-proof provisions for asylum and withholding of removal).

[65] Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, Div. C, 110 Stat. 3009-546 (1996) (codified as amended at 8 U.S.C. ch. 12).

[66] See H.R. Rep. No. 104-828, at 223–24 (1996) (Conf. Rep.) (citing Matter of Esposito, 21 I. & N. Dec. 1 (BIA 1995); Matter of Castro, 19 I. & N. Dec. 692 (BIA 1988); Matter of Ozkok, 19 I. & N. Dec. 546 (BIA 1988)).

[67] H.R. Rep. No. 104-828, at 213–14 (citing Matter of O-J-O-, 21 I. & N. Dec. 381 (BIA 1996)); see Patrick J. Glen & Alanna R. Kennedy, The Strange and Unexpected Afterlife of Pereira v. Sessions, 34 Geo. Immigr. L.J. 1, 10–12 (2019) (recounting changes to the discretionary-relief regime made by IIRIRA).

[68] H.R. Rep. No. 104-828, at 228 (citing Matter of Farias, 21 I. & N. Dec. 269 (BIA 1996)); see Glen, supra note 51, at 750–51 (noting IIRIRA’s reversal of Matter of Farias); Gonzales & Glen, supra note 53, at 890 (same).

[69] See, e.g., Matter of Cardenas-Abreu, 24 I. & N. Dec. 795, 796–802 (BIA 2009) (en banc) (interpreting IIRIRA’s statutory enactment of a definition for the term “convicted” in light of pre-1996 Board and federal court precedent, as well as legislative intent); Matter of Monreal-Aguinaga, 23 I. & N. Dec. 56, 58–63 (BIA 2001) (en banc) (interpreting the new “exceptional and extremely unusual hardship” requirement for cancellation of removal in light of the prior suspension-of-deportation standard and evident congressional intent in heightening the various eligibility criteria for discretionary relief from removal in IIRIRA).

[70] See Wadhia & Walker, supra note 1, at 1217, 1223–24.

[71] Sidney A. Shapiro, The Failure to Understand Expertise in Administrative Law: The Problem and the Consequences, 50 Wake Forest L. Rev. 1097, 1099 (2015).

[72] See Wadhia & Walker, supra note 1, at 1223 (noting that BIA members become specialists on immigration law by interpreting INA and immigration policies. Compared to federal judges, the agency developed “expertise in reconciling and accounting for conflicting evidence and arguments, disciplinary perspectives, political demands, and legal commands.” (citation omitted)).

[73] Id. at 1223–24.

[74] Convention Relating to the Status of Refugees, Art. 1(A)(2), July 28, 1951, 189 U.N.T.S. 137.

[75] See Wadhia & Walker, supra note 1, at 1223.

[76] See id. at 1202–03.

[77] See, e.g., SEC v. Chenery Corp., 332 U.S. 194, 203 (1947) (“[T]he choice made between proceeding by general rule or by individual, ad hoc litigation is one that lies primarily in the informed discretion of the administrative agency.”); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 808 n.29 (1978) (“The Commission has substantial discretion as to whether to proceed by rulemaking or adjudication.” (citation omitted)).

[78] See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”) (emphasis added).

[79] 8 U.S.C. § 1158(b)(2)(B)(ii) (emphasis added).

[80] 8 U.S.C. § 1158(b)(2)(C).

[81] 8 U.S.C. § 1158(d)(5)(B).

[82] 8 U.S.C. § 1158(d)(1).

[83] See M-E-V-G-, 26 I. & N. Dec. 227, 227 (BIA 2014); W-G-R-, 26 I. & N. Dec. 208, 208 (BIA 2014).

[84] See generally Asylum and Withholding Definitions, 65 Fed. Reg. 76,588, 76,593–95 (Dec. 7, 2000) (proposing standards to govern when an alleged particular social group is cognoscible under the INA).

[85] See cases cited in supra note 83 (listing the Board’s most recent decisions on how and when a “particular social group” may be established, including historical overviews of the Board’s development of the “particular social group” analysis).

[86] See SEC v. Chenery Corp., 332 U.S. 194, 202 (1947) (explaining that not every principle can be immediately formed as a general rule because “[s]ome principles must await their own development, while others must be adjusted to meet particular, unforeseeable situations[,]” and therefore, concluding that “an administrative agency must be equipped to act either by general rule or by individual order. To insist upon one form of action to the exclusion of the other is to exalt form over necessity”).

[87] See Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421, 448 (1987) (“There is obviously some ambiguity in a term like ‘well-founded fear’ which can only be given concrete meaning through a process of case-by-case adjudication. In that process . . . the courts must respect the interpretation of the agency . . . .” (internal quotation marks and citations omitted)); Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (adopting a similar reasoning).

[88] See, e.g., Wadhia & Walker, supra note 1, at 1241.

[89] See id. at 1221–24 (focusing on the expertise of agency personnel directly involved with the drafting and consideration of legislation on behalf of the Executive Branch).

[90] See id. at 1222.

[91] See supra Part II.C.

[92] Wadhia & Walker, supra note 1, at 1222.

[93] Id.

[94] Id. at 1222–23.

[95] See supra Part II.B.

[96] E. Donald Elliot, Re-inventing Rulemaking, 41 Duke L.J. 1490, 1492 (1992); see David J. Barron & Elena Kagan, Chevron’s Nondelegation Doctrine, 2001 Sup. Ct. Rev. 201, 231–32 (2001) (“The more courts have required agencies to give detailed notice of proposed regulatory action to interest groups, the more pressure agencies have felt to complete the bulk of their work prior to the onset of the rulemaking process . . . the less flexibility they show during rulemaking.”); see also Wendy Wagner, Katherine Barnes & Lisa Peters, Rulemaking in the Shade: An Empirical Study of EPA’s Air Toxic Emission Standards, 63 Admin. L. Rev. 99, 123 (2011) (“The findings of this study [on the EPA’s adoption of the air toxic emissions standards] underscore both the accuracy and the importance of Professor Elliott’s remarks . . . [it] reinforce[s] the possibility that a great deal of interest group influence occurs outside of the glass box of notice and comment.”).

[97] Elliot, supra note 96, at 1492; see Stephanie Stern, Cognitive Consistency: Theory Maintenance and Administrative Rulemaking, 63 Pitt. L. Rev. 589, 597 (2002) (“The timing of rulemaking encourages agency lock-in by concentrating the bulk of decisionmaking in the pre-notice period. Notice occurs after the agency has completed substantial amounts of development, analysis, and review.”); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1669, 1775 (1975) (“Indeed, the content of rulemaking decisions is often largely determined in advance through a process of informal consultation in which organized interests may enjoy a preponderant influence.”).

[98] Elliot, supra note 96, at 1492–93 (listing several methods to secure public participation: “[F]rom informal meetings with trade associations and other constituency groups, to roundtables, to floating ‘trial balloons’ in speeches or leaks to the trade press, to the more formal techniques of advisory committees and negotiated rulemaking”).

[99] Id. at 1493; see Jud Mathews, Minimally Democratic Administrative Law, 68 Admin. L. Rev. 605, 636 (2016) (noting divergence between “the justifications that they will believe satisfy courts” and “the agency’s real reasons for acting” and concluding that “demanding judicial review standards contribute to the ossification of the rulemaking process, generating delays and reducing agencies’ functional rulemaking capacity”).

[100] See, e.g., Shoba Sivaprasad Wadhia, Symposium: Reflections on the Travel Ban Decision, SCOTUSblog (June 26, 2018, 5:02 PM), https://www.scotusblog.com/2018/06/symposium-reflections-on-the-travel-ban-decision [https://perma.cc/WL5S-VS9W] (noting that Professor Wadhia co-authored an amicus brief for scholars of immigration law on the history of the Immigration and Nationality Act in support of Hawaii in Trump v. Hawaii, 138 S.Ct. 2392 (2018)).

[101] See Wadhia & Walker, supra note 1, at 1236.

[102] Id. at 1241.

[103] Kisor v. Wilkie, 139 S. Ct. 2400 (2019).

[104] Wadhia & Walker, supra note 1, at 1241 (citing Processes and Procedures for Issuance and Use of Guidance Documents, 85 Fed. Reg. 63,200, 63200–04 (Oct. 7, 2020) (to be codified at 28 C.F.R. pt. 50)).

[105] See, e.g., Berger v. United States, 295 U.S. 78, 88 (1935) (“The United States Attorney is the representative not of an ordinary party to a controversy, but of a sovereignty . . . and whose interest, therefore, in a criminal prosecution is not that it shall win a case, but that justice shall be done.”); United States v. Young, 470 U.S. 1, 25 (1985) (Brennan, J., concurring in part and dissenting in part) (noting that the Supreme Court has “long emphasized that a representative of the United States Government is held to a higher standard of behavior” than private counsel) (alteration in original).

[106] This obviously happens infrequently, and when it does happen the issue is usually not a substantive interpretation of the agency but a jurisdictional determination by the court of appeals. See Mata v. Lynch, 576 U.S. 143, 147 (2015) (noting government’s agreement with the petitioner contra the holding of the court of appeals); Kucana v. Holder, 558 U.S. 233, 241–42 (2010) (same); see also Abdisalan v. Holder, 774 F.3d 517, 523 (9th Cir. 2014) (en banc) (noting the confluence of the petitioner’s and government’s arguments on rehearing).

[107] See generally Mark B. Stern & Alisa B. Klein, The Government’s Litigator: Taking Clients Seriously, 52 Admin. L. Rev. 1409 (2000) (noting that in representing agencies in the federal courts, Justice Department litigators are effectively representing “clients”).

[108] See, e.g., Wadhia & Walker, supra note 1, at 1241–42.

[109] See Pereira v. Sessions, 138 S. Ct. 2105, 2120–21 (2018) (Kennedy, J., concurring). Justice Alito stated in his dissent:

Here, a straightforward application of Chevron requires us to accept the Government’s construction of the provision at issue. But the Court rejects the Government’s interpretation in favor of one that it regards as the best reading of the statute. I can only conclude that the Court, for whatever reason, is simply ignoring Chevron.

Id. at 2121 (Alito, J., dissenting).

[110] See Patrick J. Glen & Kate E. Stillman, Chevron Deference or the Rule of Lenity? Dual-Use Statutes and Judge Sutton’s Lonely Lament, 77 Ohio St. L.J. Furthermore 129, 133–40 (2016).

[111] See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56–64 (2014) (applying deference to uphold the Board’s restrictive interpretation of the Child Status Protection Act); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (applying deference to uphold the Board’s interpretation of the continuous residence provision of the cancellation-of-removal statute).

[112] See Johnson v. Guzman Chavez, 141 S. Ct. 2271, 2291 (2021); Sanchez v. Mayorkas, 141 S. Ct. 1809, 1812 (2021); Niz-Chavez v. Garland, 141 S. Ct. 1474, 1480 (2021).

[113] Of the cases presenting some form of deference, the government prevailed in Guzman Chavez, 141 S. Ct. 2271 and Sanchez, 141 S. Ct. 1809, and lost in Niz-Chavez, 141 S. Ct. 1474. The government additionally prevailed in Garland v. Dai, 141 S. Ct. 1669 (2021), and Pereida v. Wilkinson, 141 S. Ct. 754 (2021).

[114] See United States v. Palomar-Santiago, 141 S. Ct. 1615, 1618 (2021).

[115] See Niz-Chavez, 141 S. Ct. at 1485.

The Case For Chevron Deference To Immigration Adjudications

Volume 71 October 2021
THE CASE FOR CHEVRON DEFERENCE
TO IMMIGRATION ADJUDICATIONS

Patrick J. Glen Adjunct Professor of Law, Georgetown University Law Center; Senior Litigation Counsel, United States Department of Justice, Civil Division, Office of Immigration Litigation. The author would like to thank Chris Walker and Smita Ghosh for their helpful comments on an initial draft of this article. The views and opinions expressed herein are the author’s own, and do not represent those of the Department of Justice or any component thereof.

PDF
PDF

Abstract

Chevron skepticism is in vogue in legal academia, as Professors Shoba Wadhia and Christopher Walker’s recent entry in the genre demonstrates. They place their project within the broader academic trend of arguing for limitations on the application of deference to various administrative decisions, but their aim is ultimately narrower—to show that “this case against Chevron has . . . its greatest force when it comes to immigration.”

The Professors are incorrect. Immigration adjudication presents one of the strongest cases for deference to administrative adjudication. This case is founded in the text of the statute itself and its myriad general and specific delegations of authority to the Attorney General, the expertise of the agency which has honed its interpretive enterprise through adjudicating tens of thousands of cases annually, and the ultimate political accountability of the agency head in immigration adjudication. For these reasons, the Supreme Court has applied Chevron deference to immigration adjudications since the very foundation of that framework. Although they advance an interesting contrarian thesis, the Professors ultimately provide no sound basis for retreating from four decades of established jurisprudence.

Introduction

In a recent article, published as part of Duke Law Journal’s annual Administrative Law Symposium, Professors Shoba Sivaprasad Wadhia and Christopher Walker argue that Chevron deference should not be applied to immigration adjudications, specifically, the decisions of the Board of Immigration Appeals. [1][1]. Shoba Sivaprasad Wadhia & Christopher J. Walker, The Case Against Chevron Deference in Immigration Adjudication, 70 Duke L.J. 1197 (2021). The removal proceeding is the proceeding held before an immigration judge to determine a noncitizen’s removability from the United States, as well as any eligibility for relief or protection from removal. See 8 U.S.C. § 1229a (2018) (providing the statutory authority and framework for the conduct of removal proceedings); see also Patrick J. Glen, Judulang v. Holder and the Future of 212(c) Relief, 27 Geo. Immigr. L.J. 1, 4 (2012) (providing a historical overview of removal proceedings). The immigration judge conducts the removal proceeding in the first instance, see 8 U.S.C. § 1229a(a)(1), with the Board of Immigration Appeals exercising appellate jurisdiction over the decisions of immigration judges, see 8 C.F.R. § 1003.1(b)(3) (2021) (providing the bases for appellate jurisdiction). This is a provocative thesis, contrary as it is to nearly four decades of federal court decisions, including repeated affirmations of Chevron’s applicability to immigration adjudications by the Supreme Court. [2][2]. See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56 (2014) (applying Chevron deference to uphold a decision of the Board); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (same). It is sure to prompt further debate in academia, as well as the federal courts, and to the extent it forces all interested parties to reconsider the justifications for and application of deference principles, it will be an important article. [3][3]. See, e.g., Amaya v. Rosen, 986 F.3d 424, 438 n.1 (4th Cir. 2021) (Richardson, J., dissenting) (noting the article and its argument that Chevron deference should not apply to immigration adjudications, while opining that Chevron must still be applied as a matter of precedent).

If the article has a fundamental shortcoming, however, it is perspectival: the Professors write largely from outside the system they are critiquing. That does not undercut the strength that many of their points carry, but it does provide an opening for a riposte from within the system. Having litigated the Chevron issue for the government in the immigration context before both the federal courts of appeals and the Supreme Court, my views differ significantly both in how the justifications for deference are framed and with the bottom-line conclusion that adjudications should categorically be denied deference. On my account, I would place significant weight on Congress’s delegations to the Attorney General in the Immigration and Nationality Act and the background principle of plenary power, which may be weakened in its contemporary form but is far from dead. Even assuming deference must be justified with resort to principles beyond Congress’s explicit delegations, immigration adjudications pass that bar: such adjudications are the poster child for political accountability, as the Attorney General sits at the apex of the adjudicatory bureaucracy, and the agency possesses both the craft and legislative expertise that the Professors view as integral to the agency’s decision-making process. Finally, it is important to note that rulemaking—the Professors’ preferred avenue for immigration policy-making—and adjudication often serve different purposes, a point with special importance under the INA, a statute that makes clear those circumstances where Congress mandated rulemaking and those where the agency has a freer hand to choose the method for implementing policy. The agency should be entitled to deference regardless of the path it chooses, so long as its decision is otherwise reasonable.

Part I begins by placing the Chevron issue in context. Although the immigration bureaucracy is a sprawling system of adjudication encompassing hundreds of thousands of adjudications each year before both immigration judges and the Board of Immigration Appeals, Chevron deference is concerned only with the vanishingly small minority of cases in which the Board issues a precedential decision interpreting an ambiguous provision of the INA. And in that small minority of cases, the decision-making of the Board is robust and comprehensive. Part II then proceeds to the meat of the issue, arguing that Chevron deference to immigration adjudications is appropriate. This section addresses the main points already highlighted above: the explicit general and specific statutory delegations to the Attorney General and the background principle of the Executive Branch’s plenary power in immigration; the political accountability in immigration adjudication via the Attorney General’s referral authority; and the significant institutional expertise the Board brings to bear in resolving cases through precedent-setting adjudications. Part III argues that there is no substantial reason to prefer rulemaking to adjudication when it comes to affording deference to the agency’s determinations, while highlighting the complementary roles each often plays in the development and promulgation of policy. Finally, Part IV turns to the Professors’ recommendations for reform, while posing the question of what effect, if any, would Chevron’s recalibration have in immigration cases?

I. Placing Chevron in Context: Immigration Adjudications and the Board of Immigration Appeals’ Decisional Process

Professors Wadhia and Walker lay out at some length the structure of the immigration adjudicatory system and its operational realities. [4][4]. See Wadhia & Walker, supra note 1, at 1204–13. Rather than retread that ground, it is worthwhile to address two additional points that are especially relevant for addressing the Chevron question: the scope of Chevron’s potential applicability to immigration adjudications, and the decisional process the Board engages in when issuing decisions entitled to Chevron deference.

First, Chevron’s potential scope. The immigration adjudication system is massive, encompassing 65 immigration courts and 529 immigration judges, [5][5]. See Executive Office for Immigration Review (“EOIR”) Immigration Court Listing, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/immigration-court-administrative-control-list [https://perma.cc/3LD5-KVPW] (last updated Aug. 2, 2021); U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Immigration Judge (IJ) Hiring (2021), https://www.justice.gov/eoir/page/file/1242156/download [https://perma.cc/8HQB-B57N]. as well as the Board of Immigration Appeals, an appellate body that now has 23 permanent members. [6][6]. See Board of Immigration Appeals, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/board-of-immigration-appeals-bios [https://perma.cc/5HPB-FZX2] (last updated July 19, 2021). In 2020, immigration courts received 367,038 new cases, completed 231,435, and ended the year with a backlog of 1,256,954 cases. [7][7]. See U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Pending Cases, New Cases, and Total Completions (2021), https://www.justice.gov/eoir/page/file/1242166/download [https://perma.cc/9QMQ-4N3X]. In the same year, the Board received 51,300 new appeals, resolved 33,974 appeals, and ended the year with 84,769 appeals pending. [8][8]. EOIR, Adjudication Statistics, Case Appeals Filed, Completed, and Pending, U.S. Dep’t Just. Exec. Off. For Immigr. Rev., https://www.justice.gov/eoir/page/file/1248501/download [https://perma.cc/3QRT-7WYZ] (last updated July 8, 2021). Such a high workload may, from a 30,000-foot view, lend credence at the threshold to an argument that deference is not appropriate to immigration adjudications. Professor Wadhia and Walker hint at this argument throughout their article, noting not only high caseloads but also certain variances in case outcomes based on which immigration court is conducting the proceeding and whether the alien is represented. [9][9]. See Wadhia & Walker, supra note 1, at 1228–30. These factors may undercut the contention that deference is warranted because they point to a system that is incapable of engaging in the type of reasoned adjudication that should undergird a decision claiming entitlement to deference. [10][10]. See id.

For that reason, it is worthwhile drilling down through the system to focus on that narrow class of decisions—out of the hundreds of thousands of cases that will be resolved by immigration judges and the Board in any given year—for which the question of Chevron deference would be a live issue. Deference applies only when the adjudicator “acts in its lawmaking capacity and, in the case of the BIA’s adjudications, that means only when the BIA’s decision is binding precedent within the agency.” [11][11]. Flores-Molina v. Sessions, 850 F.3d 1150, 1157 (10th Cir. 2017) (quoting Rangel-Perez v. Lynch, 816 F.3d 591, 597 (10th Cir. 2017)). The Board’s decision is binding as precedent only in circumstances where the Board issues a published decision, [12][12]. See 8 C.F.R. § 1003.1(g) (2021). and that occurs only rarely in any given year. In 2020, the Board issued 25 precedential decisions, a number on par with its average yearly output over the last eight years: 16 (2019), 23 (2018), 27 (2017), 26 (2016), 28 (2015), 29 (2014), 19 (2013). [13][13]. See Agency Decisions, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/ag-bia-decisions [https://perma.cc/BJ8H-JXYQ] (listing the Board’s precedential decisions for each volume) (last updated Sept. 8, 2021). It is this class of cases—representing .07% of the total decisions issued in 2020—that is relevant for Chevron purposes. But even this calculation overstates the universe of cases to which the government may ultimately seek deference. Although the Board may issue a precedential decision on any issue, and although that decision will bind agency adjudicators in the discharge of their responsibilities, many of these cases will not be entitled to deference before the courts of appeals. For instance, in cases where the Board decides on the divisibility of state criminal statutes or whether such a statute is a categorical match to the generic federal offense, [14][14]. See, e.g., Matter of Nemis, 28 I. & N. Dec. 250 (BIA 2021) (adjudicating application of the categorical approach to the federal conspiracy statute, 18 U.S.C. § 371, and a visa-fraud provision, 18 U.S.C. § 1546(a)); Matter of Dikhtyar, 28 I. & N. Dec. 214, 214–16 (BIA 2021) (addressing the divisibility of the Utah state drug schedules by specific controlled substance); Matter of P-B-B-, 28 I. & N. Dec. 43 (BIA 2020) (addressing the divisibility with regard to Arizona’s criminalization of possession of a “dangerous drug”). the decision will not be entitled to deference before the courts of appeals even though the decision is a precedential decision that binds agency adjudicators. This is because the Board is held to have no particular expertise on state criminal laws, and because the categorical approach implicates a legal analysis to which deference is not relevant. [15][15]. See, e.g., Omargharib v. Holder, 775 F.3d 192, 196 (4th Cir. 2014) (“Although we generally defer to the BIA’s interpretations of the INA, where, as here, the BIA construes statutes [and state law] over which it has no particular expertise, its interpretations are not entitled to deference.” (alteration in original) (quoting Karimi v. Holder, 715 F.3d 561, 566 (4th Cir. 2013))); Ramos v. U.S. Att’y Gen., 709 F.3d 1066, 1069 n.2 (11th Cir. 2013) (“We owe no Chevron deference to the Board’s interpretation of the Georgia [criminal] statute, which the Board has no power to administer.”); Denis v. U.S. Att’y Gen., 633 F.3d 201, 208 (3d Cir. 2011) (“[I]f the issue turns on the meaning of a federal statute other than the INA, we possess the requisite expertise to interpret a federal criminal statute such that no deference is due.” (citation omitted)); see also Almeida v. Holder, 588 F.3d 778, 785 (2d Cir. 2009) (deferring to the BIA’s interpretation of “theft offense,” but reviewing questions of a particular crime defined by state law de novo). For present purposes, it is enough to note that the universe of Chevron-eligible decisions in the immigration context is exceedingly narrow, and implicates less than one-tenth of one percent of all Board decisions issued in any given year.

Second, a corollary of the first point, the decisional-process undergirding the issuance of a precedential decision is more intensive and deliberative than issuance of single-member decisions. Precedential decisions are only issued in cases where the Board has sat as a three-member panel or en banc, and the decision may only be issued as a precedent where the entire Board membership discusses the case and votes in the affirmative to issue it as such. [16][16]. See 8 C.F.R. § 1003.1(g) (2021). In this sense, every Board decision is essentially en banc—no precedent issues unless a majority of the Board members believes it should be so issued. In this way, the work of the Board parallels that of the federal courts of appeals. The Ninth Circuit may issue only a handful of precedential decisions in any given week, while issuing 100 unpublished memorandum dispositions. [17][17]. For March 15, 2021, for instance, the Ninth Circuit issued 21 unpublished decisions and 3 published decisions. See Log of Unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021); Opinions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/opinions [https://perma.cc/8ST8-G668] (last visited Oct. 7, 2021). On March 12, 2021, when it issued no published decisions, it issued 13 unpublished decisions. See Log of unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021). Nobody would allege that a published opinion issued by the Ninth Circuit is somehow lacking in relative deliberative processes just because the court also issued 20 or more unpublished decisions the same day. What matters for assessing the adequacy of deliberation is the case at issue, and for the Board and Chevron purposes, the vanishingly small number of precedential decisions it issues in a year, not the thousands of other orders and decisions it will reach.

II. The Case for Chevron Deference

When a court is confronted with an issue resolved by the Board in a precedential decision, it should afford deference to that resolution under familiar principles of Chevron deference. This deference stems from the statute’s own explicit delegations to the Attorney General, coupled with background principles of judicial deference to the Executive Branch in immigration matters, the political accountability the agency enjoys in the course of discharging its adjudicatory functions, and its comparative expertise in the immigration issues subject to adjudication. Although Professors Wadhia and Walker attempt to cast doubt on the legitimacy of these justifications for deference in the immigration context, their arguments are ultimately misguided.

A. Explicit Delegations and Structural Cues

The “statutory” case for Chevron deference in immigration adjudications is simple and straightforward. The INA broadly provides the Attorney General with authority to “establish . . . regulations, prescribe such forms of bond, reports, entries, and other papers, issue such instructions, review such administrative determinations in immigration proceedings, delegate such authority, and perform such other acts as [he] determines to be necessary for carrying out this section.” [18][18]. 8 U.S.C. § 1103(g)(2). The Act also provides that the Attorney General’s “determination and ruling . . . with respect to all questions of law shall be controlling.” [19][19]. 8 U.S.C. § 1103(a)(1). These grants of authority are important markers for Chevron purposes, as they represent “express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.” [20][20]. United States v. Mead Corp., 533 U.S. 218, 229 (2001). Beyond the broad delegations framing the Attorney General’s authority under the INA as a whole, the statute contains many more specific, express delegations of authority. For instance, the INA delegates to the Attorney General the authority for establishing procedures for consideration of asylum applications, provides authority to establish mandatory bars to eligibility and other conditions for consideration of such applications, and uses open-ended terminology that allows for interpretation by Executive Branch adjudicators and rule-makers. [21][21]. See 8 U.S.C. § 1158(d)(1) (“The Attorney General shall establish a procedure for the consideration of asylum applications filed under [this section].”); 8 U.S.C. § 1158(d)(5)(B) (“The Attorney General may provide by regulation for any other conditions or limitations on the consideration of an application for asylum not inconsistent with this chapter.”); 8 U.S.C. § 1158(b)(2)(B)(ii) (“The Attorney General may designate by regulation offenses that will be considered to be a [disqualifying crime for purposes of asylum eligibility].”); 8 U.S.C. § 1158(b)(2)(C) (“The Attorney General may by regulation establish additional limitations and conditions, consistent with this section, under which an alien shall be ineligible for asylum under paragraph (1).”). See generally Patrick J. Glen, In re L-A-C-: A Pragmatic Approach to the Burden of Proof and Corroborating Evidence in Asylum Proceedings, 35 Geo. Immigr. L.J. 1, 34 (2020) (“The asylum statute . . . does not generally mandate specific procedures for the consideration of asylum applications. Rather, that section is composed almost entirely of explicit and implicit delegations.”). And the INA otherwise tends towards broad and general language for key points that necessitates interpretation by the Attorney General, whether that pertains to what is encompassed by the term “public charge,” [22][22]. See 8 U.S.C. §§ 1182(a)(4), 1227(a)(5) (stating public charge ground of inadmissibility and deportability, respectively); see also Field Guidance on Deportability and Inadmissibility on Public Charge Grounds, 64 Fed. Reg. 28,689, 28,691 (Mar. 26, 1999) (providing public guidance as to how the Immigration and Naturalization Service would apply the public charge grounds of removability); Inadmissibility on Public Charge Grounds, 84 Fed. Reg. 41,292, 41,315 (Aug. 14, 2019) (revising the Executive Branch’s interpretation of the public charge inadmissibility ground). how “good moral character” should be assessed, [23][23]. See 8 U.S.C. § 1101(f) (listing per se categories that would involve a finding that the alien lacked good moral character, but allow the Attorney General discretion to conclude that aliens outside those categories also lack good moral character); see, e.g., 8 U.S.C. § 1229b(b)(1)(B) (requiring good moral character in order to establish eligibility for cancellation of removal). or what offenses fall within the parameters of crimes of “domestic violence, stalking, and child abuse,” [24][24]. See 8 U.S.C. § 1227(a)(2)(E)(i); see also Matter of Velazquez-Herrera, 24 I. & N. Dec. 503, 504 (BIA 2008) (interpreting this provision). to offer only a few examples. The point is that the statutory framework established by Congress not only explicitly delegates broad authority to the Attorney General to resolve both general and specific interpretive issues, but almost invariably uses language that invites resolution through interpretation.

The Supreme Court has itself based its decisions extending Chevron deference to immigration adjudications on this statutory framework. In Immigration and Naturalization Service v. Cardoza-Fonseca, [25][25]. Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421 (1987). the Court recognized that although there will still be strictly legal questions to which deference will not apply, there are also statutory ambiguities in the INA “which can only be given concrete meaning through a process of case-by-case adjudication,” and in the process of filling the gaps, the courts “must respect the interpretation of the agency to which Congress has delegated the responsibility for administering the statutory program.” [26][26]. Id. at 448 (internal quotation marks and citations omitted). The Court clarified its rationale for applying Chevron to immigration adjudications in INS v. Aguirre-Aguirre. [27][27]. Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 424–25 (1999). Given that the INA explicitly provided that the “determination and ruling by the Attorney General with respect to all questions of law shall be controlling,” [28][28]. 8 U.S.C. § 1103(a)(1). the Court deemed it “clear that principles of Chevron deference are applicable” to immigration adjudications. [29][29]. Aguirre-Aguirre, 526 U.S. at 424. In Aguirre-Aguirre, the Court found further support in the statute’s specific conferral of “decisionmaking authority on the Attorney General” in determining an alien’s eligibility for withholding of removal, [30][30]. Id. at 424–25. as well as its prior recognition “that judicial deference to the Executive Branch is especially appropriate in the immigration context where officials ‘exercise especially sensitive political functions that implicate questions of foreign relations.’” [31][31]. Id. at 425 (quoting Immigr. & Naturalization Serv. v. Abudu, 485 U.S. 94, 110 (1988)).

These latter two points should be emphasized. First, the statute broadly provides authority to the Attorney General to interpret the statute he is charged with administering, while containing numerous other delegations of decisional authority akin to what the Supreme Court found compelling in Aguirre-Aguirre, including provisions relating to asylum, cancellation of removal, adjustment of status, and numerous other waivers and additional forms of relief or protection from removal. [32][32]. See Patrick J. Glen, Interring the Immigration Rule of Lenity, 99 Neb. L. Rev. 533, 560–61 n.191 (2021) (noting that “[t]he statute is . . . replete with express delegations of decisional authority to the Attorney General,” and citing examples relating to asylum, cancellation of removal, and other forms of relief). In other words, the INA is not a statute where courts must discern whether there is an implied or implicit delegation of authority to resolve statutory ambiguity; Congress has provided its explicit instruction that the Attorney General’s determinations, made in the course of discharging his responsibilities and decision-making authority under the INA, should be given “controlling” weight. [33][33]. See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”).

Second, judicial deference in the immigration context must be placed within the history of Executive authority over immigration; the Executive’s authority stems not only from the statutory scheme, but constitutes a background principle inherent in the nature of immigration law. In 1950, the Supreme Court noted that “[w]hen Congress prescribes a procedure concerning the admissibility of aliens, it is not dealing alone with a legislative power. It is implementing an inherent executive power.” [34][34]. United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 542 (1950) (emphasis added). In Fiallo v. Bell, [35][35]. Fiallo v. Bell, 430 U.S. 787 (1977). the Supreme Court opined that its cases “have long recognized the power to expel or exclude aliens as a fundamental sovereign attribute exercised by the Government’s political department largely immune from judicial control.” [36][36]. Id. at 792 (citation omitted). This simply restated what it had said a year earlier in Hampton v. Mow Sun Wong [37][37]. Hampton v. Mow Sun Wong, 426 U.S. 88 (1976).: “the power over aliens is of a political character and therefore subject only to narrow judicial review.” [38][38]. Id. at 101 n.21 (citation omitted). For this reason, the Supreme Court has traditionally, and even still, extended broad deference to the Executive’s immigration-related decisions because of the specific context of immigration law.

Beyond these principles, subsequent decisions of the Supreme Court have focused on the specific expertise of the agency in resolving complicated issues of immigration law, as well as the policy-oriented disposition of certain issues that permissibly rests with the agency. For instance, Justice Kagan recently described an immigration case as

[T]he kind of case Chevron was built for. Whatever Congress might have meant . . . it failed to speak clearly. Confronted with a self-contradictory, ambiguous provision in a complex statutory scheme, the Board chose a textually reasonable construction consonant with its view of the purposes and policies underlying immigration law. [39][39]. Scialabba v. Cuellar de Osorio, 573 U.S. 41, 75 (2014) (plurality opinion).

And in the related context of the ordinary remand-rule, the Court used agency expertise as a rationale for remanding to the agency rather than a court of appeals to address legal or statutory interpretation questions in the first instance. [40][40]. See Negusie v. Holder, 555 U.S. 511, 523 (2009) (explaining that remand rules exist, in part, “because ‘ambiguities in statutes within an agency’s jurisdiction to administer are delegations of authority to the agency to fill the statutory gap in reasonable fashion. Filling these gaps . . . involves difficult policy choices that agencies are better equipped to make than courts.’” (quoting Nat’l Cable & Telecomm. Ass’n. v. Brand X Int. Serv., 545 U.S. 967, 980 (2005)). Justice Kennedy also noted that on remand,
[t]he agency’s interpretation of the statutory meaning of “persecution” may be explained by a more comprehensive definition, one designed to elaborate on the term in anticipation of a wide range of potential conduct; and that expanded definition in turn may be influenced by how practical, or impractical, the standard would be in terms of its application to specific cases. These matters may have relevance in determining whether its statutory interpretation is a permissible one.
Id. at 524; see also Gonzales v. Thomas, 547 U.S. 183, 186–87 (2006) (per curiam) (noting that on remand, “[t]he agency can bring its expertise to bear upon the matter; it can evaluate the evidence; it can make an initial determination . . . it can, through informed discussion and analysis, help a court later determine whether its decision exceeds the leeway that the law provides” (quoting Immigr. & Naturalization Serv. v. Ventura, 537 U.S. 12, 17 (2002) (per curiam))).

This history provides a firm foundation for Chevron’s application. Congress has explicitly delegated authority to the Attorney General, including to conduct adjudications under the INA, and this delegation bolsters an inherent Executive authority in the immigration context. Add to that the expertise the agency has in its sole subject matter focus, and it is little wonder that deference has been applied to immigration decisions continually since the advent of Chevron.

B. Political Accountability and Attorney General Referral

Immigration adjudication, perhaps more than any other form of contemporary agency adjudication, offers exactly the type of political accountability that should underpin Chevron deference. The Professors argue that two types of political accountability should be relevant for administrative purposes, an “elections matter” conception that advances the “electoral accountability in the administrative state,” and a “deliberative accountability” that is centered on participatory goals, including obtaining the views of affected or interested parties. [41][41]. See Wadhia & Walker, supra note 1, at 1231–32 (citations omitted). The Professors only make the conclusory assertion that “policymaking through adjudication may not be an adequate substitute for rulemaking under an ‘elections matter’ accountability theory,” seemingly basing this contention on a comparison of adjudication versus rulemaking rather than the inherent lack of political accountability in immigration adjudication. [42][42]. See id. Regarding deliberative shortcomings, they rely on purported shortcomings in the decisional process of the Attorney General on review. The Professors’ concerns are overblown, and that political accountability is certainly present in the immigration system.

The Attorney General is, under the text of the statute, the chief administrator of the INA. [43][43]. See 8 U.S.C. § 1103(a), (g). The Board acts only as his delegate, and only within the context in which he has delegated his authority to the Board to act. [44][44]. See 8 C.F.R. § 1003.1(d)(1) (1958) (“The Board shall function as an appellate body charged with the review of those administrative adjudications under the Act that the Attorney General may by regulation assign to it.”); see also Delegation of Powers and Definition of Duties, Immigration and Naturalization Service, 5 Fed. Reg. 2454, 2454 (July 1, 1940) (“[T]he Board of Review of the Immigration and Naturalization Service shall have authority to exercise the powers of the Attorney General [in certain delineated cases].”). As the Attorney General’s delegate, the Board must still exercise its own independent judgment, [45][45]. See 8 C.F.R. § 1003.1(d)(1)(ii) (“Board members shall exercise their independent judgment and discretion in considering and determining the cases coming before the Board.”). and its decisions are deemed its own and not attributable to the Attorney General. [46][46]. See Tefel v. Reno, 972 F. Supp. 608, 613 n.1 (S.D. Fl. 1997) (“[T]he decision of the BIA is not factually, nor legally, the decision of the Attorney General.”). Moreover, a corollary to the regulatory requirement that the Board’s decisions are a reflection of its own independent judgment, the Attorney General may not attempt to influence or dictate the decisions of the Board. [47][47]. See United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260, 266–67 (1954) (“[T]he word ‘discretion’ means that the recipient must exercise his authority according to his own understanding and conscience. In short . . . the Attorney General denies himself the right to sidestep the Board or dictate its decision in any manner.”). But the regulations do safeguard the Attorney General’s fundamental role in immigration adjudication, by permitting him to decide cases he opts to refer to himself, or in which the Board or the Department of Homeland Security requests his review. [48][48]. See 8 C.F.R. § 1003.1(h). This mechanism allows the Attorney General to decide both discrete issues and broader questions of law through adjudication, and means that “the Attorney General is the final arbiter of the immigration agency’s interpretation of a statute.” [49][49]. Sanchez-Penunuri v. Longshore, 7 F. Supp. 2d 1136, 1149 (D. Colo. 2013).

Exercise of this authority fits squarely with the “elections matter” conception of Executive Branch accountability. [50][50]. See Wadhia & Walker, supra note 1, at 1232. The Trump Administration Attorneys General utilized the authority to implement an immigration policy that tracked with what the campaign promised it would do, a policy implementation that followed the election returns. [51][51]. See Patrick J. Glen, Robert F. Kennedy and the Attorney General’s Referral Authority: A Blueprint for the Biden Administration, 49 Hofstra L. Rev. 719, 721 (2021) (noting how the Trump Administration utilized Attorney General referral to advance an immigration policy consonant with its campaign promises, including restricting eligibility for asylum, streamlining enforcement and adjudications to limit the possibilities of delay in removal proceedings, and focusing on detention of aliens both during and after the conclusion of removal proceedings). Likewise, I have recently argued that the Biden Administration should utilize the authority to implement its preferred immigration policy within the limits the authority permits. [52][52]. See generally id. (discussing Biden administration’s policy).

Other Attorneys General have also utilized the referral authority in a manner that squares with this aspect of the political accountability theory. A few examples: Attorneys General in the Reagan and Bush Administrations referred the deportation case of a United Kingdom citizen based on foreign policy concerns and the impact on the U.S.-U.K. relationship; [53][53]. See Deportation Proceedings for Joseph Patrick Thomas Doherty, 13 Op. O.L.C. 1, 8 (1989); Deportation Proceedings of Joseph Patrick Thomas Doherty, 12 Op. O.L.C. 1, 8 (1988); see also Alberto R. Gonzales & Patrick Glen, Advancing Executive Branch Immigration Policy Through the Attorney General’s Review Authority, 101 Iowa L. Rev. 841, 882–86 (2016) (providing background and context to the Attorney Generals’ decisions in Matter of Doherty). Attorney General Reno referred the Board’s decision in Matter of R-A-, rejecting the claim of an asylum applicant that victims of domestic violence constitute a “particular social group,” based on the concerns of advocates, who had also led the Clinton Administration to propose rulemaking on the same issue; [54][54]. See R-A-, 22 I. & N. Dec. 906, 906 (Att’y Gen. 2001); see also Gonzales & Glen, supra note 53, at 886–89 (providing background and context to the Attorney General’s decision in Matter of R-A-). Attorney General Mukasey referred a decision on female genital mutilation and reversed the Board’s restrictive opinion after an outcry from advocates and direct pleas by lawmakers and advocates to intervene; [55][55]. See A-T-, 24 I. & N. Dec. 617, 617 (Att’y Gen. 2008); see also Gonzales & Glen, supra note 53, at 861–63 (providing background and context to the Attorney General’s decision in Matter of A-T-). and Attorney General Holder remanded a same-sex civil union case for further proceedings amid the Obama Administration’s deliberations over whether to continue defending the Defense of Marriage Act. [56][56]. See Dorman, 25 I. & N. Dec. 485, 485 (Att’y Gen. 2011); see also Glen, supra note 51, at 760–71 (providing background and context to the Attorney General’s decision in Matter of Dorman). This list is not exhaustive, [57][57]. For more on the history of Attorney General’s decisions, see generally Gonzales & Glen, supra note 53 (focusing on decisions issued in the Bush II and Obama administrations, with reference to additional decisions in the Reagan, Bush I, and Clinton administrations); Glen, supra note 51, at 726–63 (addressing the decisions issued during the Kennedy, Clinton, and Obama administrations). but it does establish the Attorney General as the final arbiter in immigration proceedings when necessary to take into account potential political ramifications of the decision. That it does not happen more often does not point to a lack of political accountability in immigration adjudication, but to the fact that that adjudication—unlike other areas of administrative law—rarely implicates questions that would place the need for political accountability front and center.

In the course of conducting these adjudications, the Attorney General also far more often than not (almost invariably, in fact) meets the conditions under the “deliberative democracy” conception of political accountability. When the Attorney General accepts a case on referral, briefing from the parties is normally contemplated, and immigration advocates have usually been invited to participate as amici, as well. [58][58]. See In re Castro-Tum, 27 I. & N. Dec. 187, 187 (Att’y Gen. 2018) (referring the case, directing briefing on specific issues, and inviting “interested amici to submit briefs”); In re A-B-, 27 I. & N. Dec. 227, 227 (Att’y Gen. 2018) (same); In re L-A-B-R-, 27 I. & N. Dec. 245, 245 (Att’y Gen. 2018) (same); In re M-G-G-, 27 I. & N. Dec. 469, 469 (Att’y Gen. 2018) (same); M-S-, 27 I. & N. Dec. 476, 476 (Att’y Gen. 2018) (same); In re Negusie, 27 I. & N. Dec. 481, 481 (Att’y Gen. 2018) (same); In re L-E-A-, 27 I. & N. Dec. 494, 494 (Att’y Gen. 2018) (same); In re Castillo-Perez, 27 I. & N. Dec. 495, 495 (Att’y Gen. 2018) (same); In re Thomas & Thompson, 27 I. & N. Dec. 556, 556 (Att’y Gen. 2019) (same); In re Reyes, 27 I. & N. Dec. 708, 708 (Att’y Gen. 2019) (same); In re A-M-R-C-, 28 I. & N. Dec. 7, 7 (Att’y Gen. 2020) (same). This is not a regulatory requirement, nor does every case elicit a request for responses from the parties. But in those cases where such briefs are solicited (the overwhelming majority), it is difficult to see why that public invitation and the briefing it may prompt is substantially less important from a participatory point of view than the publication of a proposed rule with its own invitation for comments. This process also has sufficient transparency. The public is not privy to the decision-making process of the Attorney General in the course of rendering his decision on a referred case, but neither is it privy to the consideration of submitted comments and what responses they may trigger from the interested agencies engaged in a rulemaking. In both cases, the public does see the final result—the Attorney General’s decision with arguments accepted or rejected and the reasoning for the relevant determinations, and the Final Rule with comments noted and the agencies’ response memorialized—but in neither is there any significant transparency concerning how the agency arrived at that final determination.

C. The Expertise of the Board of Immigration Appeals

The Professors also fault the agency for a lack of relevant “expertise” for Chevron purposes. [59][59]. See Wadhia & Walker, supra note 1, at 1215–24. Although there is some force to aspects of their argument—scientific knowledge and expertise may be entirely lacking in the immigration context, and technical expertise, too, may be less important than in other areas, other aspects seem incomplete or wrong. Here, I deal with two issues: first, the concept of “legislative expertise,” the expertise the agency obtains through its interactions with the legislative branch in the course of revising the statute, and which, in my (perhaps idiosyncratic) conception, the Board does possess; and second, “craft expertise,” the expertise the agency obtains through its focus on resolving issues under a particular statute, and which, given the case flows through the agency, the Board could be expected to have in spades. [60][60]. Although not a central focus of their argument, the Professors also argue that no “legal or policy expertise in immigration or foreign relations” is required to become a member of the Board. See Wadhia & Walker, supra note 1, at 1217–19. That is true in one sense, as a lack of such knowledge is not disqualifying, but untrue in another, as subject-matter expertise is an important part of the application process—a fact borne out by the complete job announcement that the Professors refer to only in part. See Immigration Judge, USAJobs, https://www.usajobs.gov/Get/Job?viewDetails/570894500 [https://perma.cc/ZU5H-49D7] (noting the requirement that applicants respond to the Quality Rating Factors, including “knowledge of immigration laws and procedures”). It is also worth noting that new hires are tested prior to the commencement of adjudicatory activities, ensuring there is a baseline of subject-matter specific expertise. See 8 C.F.R. § 1003.0(b)(1)(vi) (2021). Both Board members and immigration judges are required to undertake ongoing training “to promote the quality and consistency of adjudications.” See 8 C.F.R. §§ 1003.0(b)(1)(vii), 1003.1(a)(2)(i)(B) (2021). And throughout their time as adjudicators, Board members are assessed and evaluated to ensure they are properly discharging the functions of the office. See 8 C.F.R. §§ 1003.0(b)(1)(vi), 1003.1(a)(2)(i)(D) (2021).

First, the Professors argue that the Board lacks “[l]egislative [e]xpertise,” defined as “the expertise derived from the principal-agent relationship between Congress and the agency.” [61][61]. See Wadhia & Walker, supra note 1, at 1221. The Professors contend that “[t]his specialized knowledge of legislative purpose and process should only matter, from a Chevron-expertise perspective, if the agency statutory interpreter possesses that expertise—either directly because the interpreter helped draft the statute or indirectly because the interpreter interacts with the agency personnel who possess that expertise.” [62][62]. Id. at 1222. There is no question that Board members do not themselves assist with the drafting or review of legislation, and no reason to believe they have any interactions with agency personnel that do, so in one sense there is force to the Professors’ argument.

But the concept of “legislative expertise” is given too-narrow a definition in the Professors’ argument. In a more fundamental sense, the Board does have important insights into relevant legislative policies and purposes, because immigration enactments have largely unfolded in the past three to four decades as a complicated case of action and reaction, with the Board and courts of appeals constituting the “action” and Congress providing the “reaction.” The major amendments to the asylum statute made by the REAL ID Act of 2005, for instance, including new provisions relating to the burden of proof, credibility assessments, and corroboration requirements, were meant to codify existing Board standards as against contrary precedent in the Ninth Circuit. [63][63]. See H.R. Rep. No. 109-72, at 161–69 (2005) (Conf. Rep.). Now that Congress has looked to the Board itself for the statement of the proper standards, who is better than the Board to interpret any gaps or ambiguities left in those provisions? And, in fact, knowledge of this legislative background—the rules Congress wanted to reject and the policies it was interested in advancing—has informed the Board’s interpretation of these provisions, and produced exactly the type of decision that should qualify for deference under the terms of Chevron. [64][64]. See, e.g., Matter of L-A-C-, 26 I. & N. Dec. 516, 518–24 (BIA 2015) (interpreting the REAL ID Act’s corroboration requirement in light of its professed purpose of rejecting contrary court of appeals’ precedent and adopting the Board’s rule); Matter of C-T-L-, 25 I. & N. Dec. 341, 343–48 (BIA 2010) (applying a similar reasoning in regard to the REAL ID Act’s amendments to the statutory burden-of-proof provisions for asylum and withholding of removal).

The REAL ID Act is not an isolated instance, either. In the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA), [65][65]. Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, Div. C, 110 Stat. 3009-546 (1996) (codified as amended at 8 U.S.C. ch. 12). Congress made numerous changes to the INA premised on issues that arose through administrative and federal litigation, including: amending the definition of conviction to adopt certain aspects of the Board’s prevailing definition, while rejecting others; [66][66]. See H.R. Rep. No. 104-828, at 223–24 (1996) (Conf. Rep.) (citing Matter of Esposito, 21 I. & N. Dec. 1 (BIA 1995); Matter of Castro, 19 I. & N. Dec. 692 (BIA 1988); Matter of Ozkok, 19 I. & N. Dec. 546 (BIA 1988)). eliminating suspension of deportation and replacing it with cancellation of removal, including a heightened hardship standard deemed necessary because of lax application of the prior “extreme hardship” standard; [67][67]. H.R. Rep. No. 104-828, at 213–14 (citing Matter of O-J-O-, 21 I. & N. Dec. 381 (BIA 1996)); see Patrick J. Glen & Alanna R. Kennedy, The Strange and Unexpected Afterlife of Pereira v. Sessions, 34 Geo. Immigr. L.J. 1, 10–12 (2019) (recounting changes to the discretionary-relief regime made by IIRIRA). and clarifying certain issues relating to when a qualifying relationship must exist for purposes of certain waivers of inadmissibility. [68][68]. H.R. Rep. No. 104-828, at 228 (citing Matter of Farias, 21 I. & N. Dec. 269 (BIA 1996)); see Glen, supra note 51, at 750–51 (noting IIRIRA’s reversal of Matter of Farias); Gonzales & Glen, supra note 53, at 890 (same). In other words, the base-line for many of the shifts in IIRIRA was Board precedent, and thus the Board is well-placed to understand: 1) what the prior rule was; 2) how Congress altered that rule; and 3) what that alteration means for purposes of a permissible interpretation going forward. [69][69]. See, e.g., Matter of Cardenas-Abreu, 24 I. & N. Dec. 795, 796–802 (BIA 2009) (en banc) (interpreting IIRIRA’s statutory enactment of a definition for the term “convicted” in light of pre-1996 Board and federal court precedent, as well as legislative intent); Matter of Monreal-Aguinaga, 23 I. & N. Dec. 56, 58–63 (BIA 2001) (en banc) (interpreting the new “exceptional and extremely unusual hardship” requirement for cancellation of removal in light of the prior suspension-of-deportation standard and evident congressional intent in heightening the various eligibility criteria for discretionary relief from removal in IIRIRA). This of course tracks quite nicely with the concept of “legislative experience” offered by the authors. Their error, it seems to me, was reading that concept too narrowly to exclude the institutional knowledge the agency has by virtue of its central interpretive role in the relevant statutory scheme.

Second, the Professors posit the possibility that the Board may possess “craft expertise,” relying on the work of Professor Sidney Shapiro. [70][70]. See Wadhia & Walker, supra note 1, at 1217, 1223–24. Professor Shapiro advances a conception of “institutional expertise,” tied to the “unique wisdom of [the] regulatory agency,” and ultimately premised on “experience” rather than “formal knowledge.” [71][71]. Sidney A. Shapiro, The Failure to Understand Expertise in Administrative Law: The Problem and the Consequences, 50 Wake Forest L. Rev. 1097, 1099 (2015). With tens of thousands of appeals resolved each year, the Board could be expected to possess “craft expertise” in spades, and in fact, Professor Wadhia and Walker feint in that direction initially. [72][72]. See Wadhia & Walker, supra note 1, at 1223 (noting that BIA members become specialists on immigration law by interpreting INA and immigration policies. Compared to federal judges, the agency developed “expertise in reconciling and accounting for conflicting evidence and arguments, disciplinary perspectives, political demands, and legal commands.” (citation omitted)). Ultimately, they conclude that the issue of craft expertise does not weigh in the Board’s favor for two reasons: 1) statutory ambiguities in the INA rarely implicate the foundational issues of expertise relevant to the concept; and 2) even if the Board does possess some craft expertise, it is a lesser form of expertise compared to that of other agency actors in the rulemaking process. [73][73]. Id. at 1223–24.

Fairly construed, the calculus seems to weigh in the other direction. By dint of resolving tens of thousands of cases annually, presenting variations on a relatively firm set of themes, the Board does augment its understanding of the immigration laws generally, as well as how they specifically apply to certain circumstances. Considering the concept of “particular social group” in the refugee definition, [74][74]. Convention Relating to the Status of Refugees, Art. 1(A)(2), July 28, 1951, 189 U.N.T.S. 137. for instance, the Board may hone its understanding of the general concept as the issue is raised across a variety of circumstances, while also sharpening its application to specifics that may be presented over and over again. By having to resolve the question repeatedly, the Board’s expertise as a general and specific matter begins to far outstrip other actors in the bureaucracy. It may be true that some questions do not implicate technical or scientific expertise, but many, if not most, that are resolved through precedential decision-making will involve matters peculiar to the agency’s institutional mission, like the “particular social group” interpretation noted above. In cases where the Board issues a precedential decision, that decision is likely the end product of having considered the issue in hundreds, thousands, or even tens of thousands of other cases, with the precedent-setting decision marking the culmination of this expansive decisional process. Not only does the Board possess the institutional expertise posited by Professor Shapiro, because of the sheer breadth of its workload, but it is also difficult to imagine an agency that could claim greater “craft expertise.”

The comparative expertise argument also seems weak. The Professors do note that the focus of the Board on immigration issues and the stream of cases means that the agency possesses greater comparative expertise than the federal courts. [75][75]. See Wadhia & Walker, supra note 1, at 1223. But there is little reason to believe that the Board does not also possess greater “craft expertise” than other agency experts. Again tying this to work-flow, the Board will have seen a particular issue raised in virtually every possible circumstance in which it can be raised, giving it a more expansive view of the legal playing field than other experts tied solely to the language of the statute or legal imagination. Given the number of cases decided by immigration judges, and the possibilities of the full panoply of issues under the INA being appealed to the Board, it is the Board that has the most comprehensive view of immigration law of virtually any actor in the field. The number of appeals considered and resolved each year, and the range of issues presented therein, dwarf the scope of rulemaking. In other words, if there is institutional expertise borne of practice, the Board can comfortably claim that mantle in the immigration context.

III. Rulemaking versus Adjudication

The Professors’ arguments against Chevron deference to immigration adjudications are not free-standing attacks against the concept, but rather the framework they utilize to advance their main normative argument: that at least so far as immigration policy-making is concerned, the agencies should categorically prefer rulemaking to adjudication. [76][76]. See id. at 1202–03. For me, because I do not believe in the strength of their arguments against the applicability of Chevron deference to immigration adjudications, I remain at best agnostic about the choice between adjudication and rulemaking. But I also believe that, even on its own terms, the Professors’ argument relies on an idealized conception of rulemaking when, in reality, the purported gulf between the desirability of these options is substantially narrower (or non-existent).

At the threshold, there is no question—and the Professors do not argue to the contrary—that when given the choice to proceed via rulemaking or adjudication, the choice is entirely within the discretion of the agency. [77][77]. See, e.g., SEC v. Chenery Corp., 332 U.S. 194, 203 (1947) (“[T]he choice made between proceeding by general rule or by individual, ad hoc litigation is one that lies primarily in the informed discretion of the administrative agency.”); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 808 n.29 (1978) (“The Commission has substantial discretion as to whether to proceed by rulemaking or adjudication.” (citation omitted)). Likewise, under controlling precedent, deference is warranted in either case, so long as the agency is fulfilling its statutory mandate and Congress has not unambiguously foreclosed the interpretation or decision made by the agency. [78][78]. See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”) (emphasis added). The provision of this choice makes sense, as some issues may be amenable to either rulemaking or adjudication, for others, there are considerations that may weigh more heavily in favor of the one rather than the other, and for still others, the choice may be directed by statute or the nature of the policy question being resolved.

Using just the asylum statute as an example, all these dynamics can be seen at work. The Attorney General may, “by regulation,” “designate . . . offenses that will be considered to be” disqualifying particularly serious crimes or serious nonpolitical crimes for purposes of asylum eligibility; [79][79]. 8 U.S.C. § 1158(b)(2)(B)(ii) (emphasis added). he “may . . . establish additional limitations and conditions . . . under which an alien shall be ineligible for asylum”; [80][80]. 8 U.S.C. § 1158(b)(2)(C). and he “may provide . . . other conditions or limitations on the consideration of an application for asylum.” [81][81]. 8 U.S.C. § 1158(d)(5)(B). He also “shall establish a procedure for the consideration of asylum applications filed under” the statute. [82][82]. 8 U.S.C. § 1158(d)(1). These are issues that by their nature or by specific direction of the statute must be resolved via rulemaking. In contrast, the definition of “particular social group,” one of the statutorily protected grounds on which basis an alien may assert eligibility for asylum, is an issue that would be amenable to either rulemaking or adjudication. The Board has established broad criteria to consider in assessing whether a particular social group qualifies as such, [83][83]. See M-E-V-G-, 26 I. & N. Dec. 227, 227 (BIA 2014); W-G-R-, 26 I. & N. Dec. 208, 208 (BIA 2014). but it is possible that such criteria could have been promulgated via regulation, as well. No final rule on “particular social groups” has ever been promulgated, but the Clinton Administration did issue a draft rule that would have addressed some of these questions. [84][84]. See generally Asylum and Withholding Definitions, 65 Fed. Reg. 76,588, 76,593–95 (Dec. 7, 2000) (proposing standards to govern when an alleged particular social group is cognoscible under the INA). In contrast with the initial examples in this paragraph, however, there is nothing about resolving the ambiguity inherent in the term “particular social group” that makes rulemaking or adjudication the preferred course of action. Over the better part of four decades, the Board has, through adjudication, provided an interpretation of that term to guide determinations of when a “particular social group” may qualify under the statute, and has applied that definition in precedential decisions to render broadly applicable holdings regarding frequently raised “particular social groups.” [85][85]. See cases cited in supra note 83 (listing the Board’s most recent decisions on how and when a “particular social group” may be established, including historical overviews of the Board’s development of the “particular social group” analysis). The same goals could have been achieved through the rulemaking process, but that process would not have provided any better avenue for issuing Executive Branch policy on the question. Either forum allows complete consideration of the important issues and final resolution of the questions so that the law may be applied to future cases. In short, policy could be established through either mechanism. But then consider application of that generally promulgated framework to specific proposed social groups, whether it is victims of domestic abuse, the nuclear family, former gang members, or wealthy returning deportees. Rulemaking cannot exhaustively address each and every conceivable proposed social group, even if it could establish the parameters for considering when such a group could be recognized. [86][86]. See SEC v. Chenery Corp., 332 U.S. 194, 202 (1947) (explaining that not every principle can be immediately formed as a general rule because “[s]ome principles must await their own development, while others must be adjusted to meet particular, unforeseeable situations[,]” and therefore, concluding that “an administrative agency must be equipped to act either by general rule or by individual order. To insist upon one form of action to the exclusion of the other is to exalt form over necessity”). Instead, adjudication, by considering each question on a case-by-case basis, can provide greater form and context for interpretation of the ambiguous term, and in so doing should be entitled to deference consistent with the Supreme Court’s direction in cases such as Cardoza-Fonseca and Aguirre-Aguirre. [87][87]. See Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421, 448 (1987) (“There is obviously some ambiguity in a term like ‘well-founded fear’ which can only be given concrete meaning through a process of case-by-case adjudication. In that process . . . the courts must respect the interpretation of the agency . . . .” (internal quotation marks and citations omitted)); Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (adopting a similar reasoning).

Presenting rulemaking as an all-or-nothing option if an agency wants to obtain deference on review fails to take into account these differences inherent in the multitude of issues raised before and considered by agencies. At the very least, failing to meaningfully grapple with these questions undercuts the force of the Professors adamant turn to rulemaking in the immigration context. To be sure, on one level the question is not the choice itself but the result of that choice, and even there the Professors’ argument resorts to the extension of some deference (just not Chevron). [88][88]. See, e.g., Wadhia & Walker, supra note 1, at 1241. Regardless of these points, however, the end result is the same: for the agencies’ policy choices to receive the deference due to them under Chevron, they must proceed through rulemaking to the exclusion of adjudication.

Despite the foregoing, the Professors’ Manichean approach might make sense if rulemaking were some sort of panacea for all the ills they raise regarding immigration adjudication. On a clear-eyed view, though, it is not. The shortcomings of rulemaking may be different in kind or scope than those of adjudication, but they are shortcomings nonetheless. The concerns noted by the Professors ultimately provide no support for their aggressive argument in favor of rulemaking.

Returning, first, to the question of expertise; in arguing against relevant expertise by agency adjudicators, the Professors framed that perceived lack against the greater benefits of expertise offered by other individuals within the subject agencies. [89][89]. See id. at 1221–24 (focusing on the expertise of agency personnel directly involved with the drafting and consideration of legislation on behalf of the Executive Branch). The question of comparative expertise seems closer to a wash, however. There are undoubtedly other experts at EOIR, DHS, and DOJ, who have meaningful contributions to make in the course of rulemaking (as well as the antecedent step of statutory revision and drafting), but there is little reason to view any of these discrete experts as in possession of substantially more knowledge than the adjudicators and their adjutants. Regarding the aggregation of expertise, in the course of rulemaking the agencies may benefit from multiple different viewpoints being brought to bear on a single problem—EOIR may be able to explain adjudicative issues posed by the rule, DHS operational impact, and DOJ potential litigation fall-out. [90][90]. See id. at 1222. Here, the benefits of rulemaking in the abstract may seem formidable, but with any precedential decision the Board will have the views of its own expert members, the staff of EOIR who work for the Board, and the views of DHS via briefing and possibly argument. The slate of experts available to the Board is certainly less than what would be available to the administration via rulemaking, but this should also be considered within the scope of the decisions being rendered. As explained above, issues that are peculiarly subject to adjudication are also substantially less likely to benefit from the kind of aggregation of expertise encouraged by the rulemaking process. Interpreting the term “particular social group” and applying that definition in specific cases as aliens proffer a range of purported social groups is of a different kind of exercise than promulgating procedures to guide the filing and consideration of asylum applications. The former issues present questions of adjudicatory expertise that are dependent on the Board’s consistent engagement with both the general statute and specific questions that arise under the statute, whereas the latter issue presents more purely institutional questions where the aggregation of agency expertise may likely contribute to the development of an optimal regulation. The aggregation issue arises only through posing a false equivalency between issues considered in rulemaking and those considered in adjudication.

The nature of expertise is also a tricky concept in this area. As I argue above, I believe that the Board has substantial institutional expertise that arises because of the impetus behind much of Congress’s reforms over the past decades. [91][91]. See supra Part II.C. Those reforms have occurred with Board precedent in mind, and thus the Board is peculiarly well-placed to understand what Congress was doing and why. I also agree with the Professors, however, that others within the immigration bureaucracy may have important contributions to make regarding the same question. [92][92]. Wadhia & Walker, supra note 1, at 1222. The Professors note in this regard the Board’s inability to access this expertise because of the ethical and practical barriers that often exist between the agency adjudicator and other government policy-makers, [93][93]. Id. and use that fact as an argument against deference to the Board itself. [94][94]. Id. at 1222–23. But if that wall is blocking a flow of information, it is blocking that flow from both directions. The Board may not have access to other experts’ views, but then those same experts do not have access to the Board’s own expert views and what could be its singular insights into the motivating factors of statutory amendments. Where the Professors see only a shortcoming insofar as the Board itself is concerned, I see a two-way street whereby experts engaged in rulemaking have only a limited universe of expertise to consider, while the Board itself has only a slightly differently constituted limited universe of expertise to consider. Both procedures lack something, and given that—along with my more generally supportive views of Board expertise in the course of adjudication—I can find no expertise-based reason to categorically prefer rulemaking while withdrawing adjudication from the scope of Chevron deference.

Political accountability also seems more or less a wash. Given the possibilities of Attorney General referral, adjudication via that avenue provides as much political accountability as does Executive agency rulemaking. [95][95]. See supra Part II.B. In both cases, voters know where the buck has stopped. I also disagree that rulemaking is inherently preferable from the perspective of public participation. Here, the Professors are quite bullish on the benefits of rulemaking, which seem superficially clear: a rule is proposed, the public is invited to review and comment, and the administration then considers the comments (sometimes in the hundreds of thousands) individually (!) and drafts a final rule taking into consideration all these concerns and comments. The notion that this process is actually engaged in soliciting and considering public views to the ends of altering the agencies’ preferred rules is at best in tension with reality. Writing in an earlier version of the Duke Administrative Law Symposium, Professor E. Donald Elliott, formerly a Public Member of the Administrative Conference of the United States and Assistant Administrator and General Counsel of the Environmental Protection Agency, remarked that “[n]o administrator in Washington turns to full-scale notice-and-comment rulemaking when she is genuinely interested in obtaining input from interested parties.” [96][96]. E. Donald Elliot, Re-inventing Rulemaking, 41 Duke L.J. 1490, 1492 (1992); see David J. Barron & Elena Kagan, Chevron’s Nondelegation Doctrine, 2001 Sup. Ct. Rev. 201, 231–32 (2001) (“The more courts have required agencies to give detailed notice of proposed regulatory action to interest groups, the more pressure agencies have felt to complete the bulk of their work prior to the onset of the rulemaking process . . . the less flexibility they show during rulemaking.”); see also Wendy Wagner, Katherine Barnes & Lisa Peters, Rulemaking in the Shade: An Empirical Study of EPA’s Air Toxic Emission Standards, 63 Admin. L. Rev. 99, 123 (2011) (“The findings of this study [on the EPA’s adoption of the air toxic emissions standards] underscore both the accuracy and the importance of Professor Elliott’s remarks . . . [it] reinforce[s] the possibility that a great deal of interest group influence occurs outside of the glass box of notice and comment.”). No doubt informed by his own personal experience, Professor Elliott offered a telling simile: “Notice-and-comment rulemaking is to public participation as Japanese Kabuki theater is to human passions—a highly stylized process for displaying in a formal way the essence of something which in real life takes place in other venues.” [97][97]. Elliot, supra note 96, at 1492; see Stephanie Stern, Cognitive Consistency: Theory Maintenance and Administrative Rulemaking, 63 Pitt. L. Rev. 589, 597 (2002) (“The timing of rulemaking encourages agency lock-in by concentrating the bulk of decisionmaking in the pre-notice period. Notice occurs after the agency has completed substantial amounts of development, analysis, and review.”); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1669, 1775 (1975) (“Indeed, the content of rulemaking decisions is often largely determined in advance through a process of informal consultation in which organized interests may enjoy a preponderant influence.”). This is not to say that rulemaking is not informed by public concerns as well as interested constituents, but only that that process is not likely to occur within the confines of rulemaking itself. [98][98]. Elliot, supra note 96, at 1492–93 (listing several methods to secure public participation: “[F]rom informal meetings with trade associations and other constituency groups, to roundtables, to floating ‘trial balloons’ in speeches or leaks to the trade press, to the more formal techniques of advisory committees and negotiated rulemaking”). The notice-and-comment process is now quite formal and driven by compliance with statutory standards, and the bare-fact of meeting those standards (while also compiling the record on which the rule will be defended in the federal courts)—not “provid[ing] . . . public input into government thinking—is the “primary[]” function of the process. [99][99]. Id. at 1493; see Jud Mathews, Minimally Democratic Administrative Law, 68 Admin. L. Rev. 605, 636 (2016) (noting divergence between “the justifications that they will believe satisfy courts” and “the agency’s real reasons for acting” and concluding that “demanding judicial review standards contribute to the ossification of the rulemaking process, generating delays and reducing agencies’ functional rulemaking capacity”).

A comparison of those parties submitting comments to the prior administration in the immigration rulemaking context would, I would wager, align highly with the interested amici submitting briefs before the Attorney General in referred cases. Professor Wadhia specifically has been an active amicus participant in immigration cases before the Supreme Court, federal courts of appeals, and district courts. [100][100]. See, e.g., Shoba Sivaprasad Wadhia, Symposium: Reflections on the Travel Ban Decision, SCOTUSblog (June 26, 2018, 5:02 PM), https://www.scotusblog.com/2018/06/symposium-reflections-on-the-travel-ban-decision [https://perma.cc/WL5S-VS9W] (noting that Professor Wadhia co-authored an amicus brief for scholars of immigration law on the history of the Immigration and Nationality Act in support of Hawaii in Trump v. Hawaii, 138 S.Ct. 2392 (2018)). Why is the latter form of participation dramatically different from the former? The Professors do not pose this question and thus do not answer it, but it is worth considering. If the essence of the public-participation argument is the ability to be heard, the opportunities are similar as between rulemaking and adjudication in that narrow class of case that will prompt a decision entitled to deference on review. The opportunities are not identical, and I do not mean to argue to the contrary—the Board and Attorney General may decide cases without amicus participation or additional briefing from the parties. But fairly considered, the opportunities to participate in those adjudications that will lead to a precedential decision are more robust and important than the Professors acknowledge, while their own preferred course of rulemaking contains only a formalized mechanism of public participation whose actual substantive importance is open to debate.

All this is to say only that rulemaking has its own warts. It is a necessary mechanism for adopting certain policies, and the route that should be preferred in other classes of cases, as well. But it is not without its own shortcomings, including as to expertise, accountability, and participation, and the Professors fail to advance any compelling reason for categorically preferring rulemaking to adjudication in the advancement of administration policy, much less an argument for why we should continue to confer deference on rulemakings while withdrawing it from adjudications.

IV. Responsibility for and the Likely Outcome of Chevron’s Recalibration

This leaves possibilities for reform—assuming Chevron should be recalibrated for purposes of immigration adjudication, how should that recalibration be accomplished? First, if Congress wants to eliminate deference for all administrative cases or any subset thereof, there is no obvious impediment to that action, although I share the Professors’ skepticism that any substantial reform to deference principles will come through legislation. [101][101]. See Wadhia & Walker, supra note 1, at 1236. The Professors do note the possibility of comprehensive immigration reform in the Biden Administration, and this well may provide an avenue to action on the specific question of deference. But it additionally opens the door to other ways to cut back on the circumstances where such deference is relevant—revising the INA or enacting new provisions that more particularly and explicitly address the relevant questions rather than more open-ended provisions leaving interpretive discretion with the agency. Eliminating or cutting back on the chances an agency would have to render an interpretation ultimately entitled to deference before the federal courts is as wise a move towards “mitigating” deference’s reign as would be a wholesale revocation of deference.

Second, perhaps most provocatively, the Professors argue for the Executive Branch to take the lead by waiving deference in immigration adjudications and “shifting major immigration policymaking away from adjudication and into the realm of notice-and-comment rulemaking.” [102][102]. Id. at 1241. For support, they note a recent regulation limiting the circumstances where the Department of Justice will seek deference in the wake of Kisor v. Wilkie’s [103][103]. Kisor v. Wilkie, 139 S. Ct. 2400 (2019). reconceptualization of Auer deference. [104][104]. Wadhia & Walker, supra note 1, at 1241 (citing Processes and Procedures for Issuance and Use of Guidance Documents, 85 Fed. Reg. 63,200, 63200–04 (Oct. 7, 2020) (to be codified at 28 C.F.R. pt. 50)). I would assume there is no barrier to the Department also doing so in the context of Chevron deference and defense of immigration adjudications. But the Professors miss an important point in citing this regulation—the Department sought to ensure that its litigation unfolded consistent with governing Supreme Court law, and did not push the boundaries by asking for deference to guidance and internal documents that otherwise would not warrant deference. That argument is entirely lacking here—the Supreme Court has consistently and emphatically described immigration adjudication as entitled to deference on judicial review. In promulgating any contrary Departmental policy or rule, the Department would not be acting to ensure compliance with Supreme Court law, but would be itself catalyzing a shift away from that law.

This also implicates attorney-client obligations. The Department’s lawyers have higher obligations and duties regarding the rule of law than private practitioners. [105][105]. See, e.g., Berger v. United States, 295 U.S. 78, 88 (1935) (“The United States Attorney is the representative not of an ordinary party to a controversy, but of a sovereignty . . . and whose interest, therefore, in a criminal prosecution is not that it shall win a case, but that justice shall be done.”); United States v. Young, 470 U.S. 1, 25 (1985) (Brennan, J., concurring in part and dissenting in part) (noting that the Supreme Court has “long emphasized that a representative of the United States Government is held to a higher standard of behavior” than private counsel) (alteration in original). This means that in certain circumstances, the Department will decline to defend otherwise favorable precedent or may confess error. [106][106]. This obviously happens infrequently, and when it does happen the issue is usually not a substantive interpretation of the agency but a jurisdictional determination by the court of appeals. See Mata v. Lynch, 576 U.S. 143, 147 (2015) (noting government’s agreement with the petitioner contra the holding of the court of appeals); Kucana v. Holder, 558 U.S. 233, 241–42 (2010) (same); see also Abdisalan v. Holder, 774 F.3d 517, 523 (9th Cir. 2014) (en banc) (noting the confluence of the petitioner’s and government’s arguments on rehearing). Along with this obligation, however, is the Department’s obligation to defend its client-agencies within the limits permitted by the law. [107][107]. See generally Mark B. Stern & Alisa B. Klein, The Government’s Litigator: Taking Clients Seriously, 52 Admin. L. Rev. 1409 (2000) (noting that in representing agencies in the federal courts, Justice Department litigators are effectively representing “clients”). The limits of the law obviously entail not stretching the text of statutes, regulations, or judicial precedent beyond their reasonable import, but there can be no colorable argument that those limits also entail voluntarily declining to seek deference in a class of adjudications that the Supreme Court has repeatedly told us warrant deference. If Chevron continues to exist, and its application to immigration adjudications continues to be sanctioned, there is no legal rationale to support the Department’s waiver of deference, and any institutional interests are purely hypothetical. [108][108]. See, e.g., Wadhia & Walker, supra note 1, at 1241–42.

Most importantly, however, voluntary abdication of deference takes the burden off the entity that should be charged with rethinking the scope of Chevron: the Supreme Court itself. Deference, as currently conceptualized, stems from Supreme Court precedent and is a canon applicable to judicial review of agency action. It is thus with the judiciary that the mantle of reform must ultimately rest. Unless or until the Court seeks to alter the deference calculus, the Executive is entitled to continue to argue for deference consistent with that precedent. And, of course, this may already be happening in immigration cases. Former Justice Kennedy criticized the courts of appeals reflexive recourse to deference in a recent immigration case, where Justice Alito, in his solo dissent, argued that the majority’s decision was an implicit repudiation of Chevron deference. [109][109]. See Pereira v. Sessions, 138 S. Ct. 2105, 2120–21 (2018) (Kennedy, J., concurring). Justice Alito stated in his dissent:
Here, a straightforward application of Chevron requires us to accept the Government’s construction of the provision at issue. But the Court rejects the Government’s interpretation in favor of one that it regards as the best reading of the statute. I can only conclude that the Court, for whatever reason, is simply ignoring Chevron.
Id. at 2121 (Alito, J., dissenting).
The Court has also declined to take a position on whether deference is warranted to immigration adjudications that touch on so-called dual-use statutes, statutes that may entail both civil and criminal consequences, instead of conducting de novo review of the question presented. [110][110]. See Patrick J. Glen & Kate E. Stillman, Chevron Deference or the Rule of Lenity? Dual-Use Statutes and Judge Sutton’s Lonely Lament, 77 Ohio St. L.J. Furthermore 129, 133–40 (2016). Of course, in the midst of these cases, the Court has continued to apply Chevron deference. [111][111]. See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56–64 (2014) (applying deference to uphold the Board’s restrictive interpretation of the Child Status Protection Act); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (applying deference to uphold the Board’s interpretation of the continuous residence provision of the cancellation-of-removal statute). But to the extent there is an increasing number of cases where the Court balks at that application, any necessary recalibration is likely already underway.

It is also important to address a question the Professors do not raise—what is the practical effect if immigration adjudications no longer qualify for Chevron deference? Presumably, the Professors would argue that the elimination of deference would be of significant importance, otherwise there would be little reason to spend 47 pages arguing for that result in a prestigious law journal. I am skeptical. In the Supreme Court itself, I see little reason to believe that a lack of deference will negatively affect the rate at which the government prevails. First, again, many cases are already resolved without recourse to Chevron. Second, the cases where Chevron was applied were not likely to have come out the other way absent a robust conception of deference. The recently-concluded Supreme Court Term offers some support for these points. The Court decided five civil immigration cases, three of which presented some claim to Chevron deference. [112][112]. See Johnson v. Guzman Chavez, 141 S. Ct. 2271, 2291 (2021); Sanchez v. Mayorkas, 141 S. Ct. 1809, 1812 (2021); Niz-Chavez v. Garland, 141 S. Ct. 1474, 1480 (2021). The Court resolved all five on plain-text grounds, and the government prevailed in four of the five cases, including two of the three that presented the deference question. [113][113]. Of the cases presenting some form of deference, the government prevailed in Guzman Chavez, 141 S. Ct. 2271 and Sanchez, 141 S. Ct. 1809, and lost in Niz-Chavez, 141 S. Ct. 1474. The government additionally prevailed in Garland v. Dai, 141 S. Ct. 1669 (2021), and Pereida v. Wilkinson, 141 S. Ct. 754 (2021). Folding in an additional criminal-immigration case decided by the Court, and the government had a 5-1 record in immigration cases in the October 2020 Term. [114][114]. See United States v. Palomar-Santiago, 141 S. Ct. 1615, 1618 (2021). A more robust form of deference may well have led to a win for the government in the sole case it lost this Term. [115][115]. See Niz-Chavez, 141 S. Ct. at 1485. But that fact does not undermine the point that plain-text resolutions of immigration issues are often, even if not invariably, going to coincide with the disposition of the agency.

The issue may be more complicated in the courts of appeals, but the Professors certainly do not make any substantial case for a sea-change in judicial review of immigration adjudications in a hypothetical post-Chevron world. The points regarding the Supreme Court are likely to be just as important in the courts of appeals, i.e., many cases are already disposed of on non-deference-related grounds, while a “better-reading” argument will often favor the position advanced by the government even in the absence of Chevron deference (whether on de novo review or on application of Skidmore “deference”). Of course, this may just provide fuel for their argument—if the outcome will not have catastrophic consequences on the adjudicatory system, why not jettison deference and give a freer hand to the courts of appeals? That would be a rational counterpunch. For me, however, it argues for maintenance of the status quo, especially in these circumstances where I think most of the criticisms of deference lack strength and the proposed alternative of rulemaking offers no net benefits.

Conclusion

Professors Wadhia and Walker have written a thoughtful critique of the application of Chevron deference to immigration adjudications, but for me it ultimately misses the mark. In some sense, I have a dog in this fight, although my interests are more institutional than substantive. The actual practice of deference in immigration litigation has not engendered any significant problems in the four decades since Chevron, and unreasonable decisions or those prohibited by the plain language of the statute will be vacated on review. But where there is ambiguity or room for the agency to make a policy choice, it is entitled to pursue that decision through adjudication. And having done so, the courts should defer under the long-accepted principles of Chevron.


Copyright © 2021 Patrick J. Glen.

Adjunct Professor of Law, Georgetown University Law Center; Senior Litigation Counsel, United States Department of Justice, Civil Division, Office of Immigration Litigation. The author would like to thank Chris Walker and Smita Ghosh for their helpful comments on an initial draft of this Article. The views and opinions expressed herein are the author’s own, and do not represent those of the Department of Justice or any component thereof.

[1] Shoba Sivaprasad Wadhia & Christopher J. Walker, The Case Against Chevron Deference in Immigration Adjudication, 70 Duke L.J. 1197 (2021). The removal proceeding is the proceeding held before an immigration judge to determine a noncitizen’s removability from the United States, as well as any eligibility for relief or protection from removal. See 8 U.S.C. § 1229a (2018) (providing the statutory authority and framework for the conduct of removal proceedings); see also Patrick J. Glen, Judulang v. Holder and the Future of 212(c) Relief, 27 Geo. Immigr. L.J. 1, 4 (2012) (providing a historical overview of removal proceedings). The immigration judge conducts the removal proceeding in the first instance, see 8 U.S.C. § 1229a(a)(1), with the Board of Immigration Appeals exercising appellate jurisdiction over the decisions of immigration judges, see 8 C.F.R. § 1003.1(b)(3) (2021) (providing the bases for appellate jurisdiction).

[2] See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56 (2014) (applying Chevron deference to uphold a decision of the Board); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (same).

[3] See, e.g., Amaya v. Rosen, 986 F.3d 424, 438 n.1 (4th Cir. 2021) (Richardson, J., dissenting) (noting the article and its argument that Chevron deference should not apply to immigration adjudications, while opining that Chevron must still be applied as a matter of precedent).

[4] See Wadhia & Walker, supra note 1, at 1204–13.

[5] See Executive Office for Immigration Review (“EOIR”) Immigration Court Listing, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/immigration-court-administrative-control-list [https://perma.cc/3LD5-KVPW] (last updated Aug. 2, 2021); U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Immigration Judge (IJ) Hiring (2021), https://www.justice.gov/eoir/page/file/1242156/download [https://perma.cc/8HQB-B57N].

[6] See Board of Immigration Appeals, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/board-of-immigration-appeals-bios [https://perma.cc/5HPB-FZX2] (last updated July 19, 2021).

[7] See U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., Adjudication Statistics: Pending Cases, New Cases, and Total Completions (2021), https://www.justice.gov/eoir/page/file/1242166/download [https://perma.cc/9QMQ-4N3X].

[8] EOIR, Adjudication Statistics, Case Appeals Filed, Completed, and Pending, U.S. Dep’t Just. Exec. Off. For Immigr. Rev., https://www.justice.gov/eoir/page/file/1248501/download [https://perma.cc/3QRT-7WYZ] (last updated July 8, 2021).

[9] See Wadhia & Walker, supra note 1, at 1228–30.

[10] See id.

[11] Flores-Molina v. Sessions, 850 F.3d 1150, 1157 (10th Cir. 2017) (quoting Rangel-Perez v. Lynch, 816 F.3d 591, 597 (10th Cir. 2017)).

[12] See 8 C.F.R. § 1003.1(g) (2021).

[13] See Agency Decisions, U.S. Dep’t of Just., Exec. Off. for Immigr. Rev., https://www.justice.gov/eoir/ag-bia-decisions [https://perma.cc/BJ8H-JXYQ] (listing the Board’s precedential decisions for each volume) (last updated Sept. 8, 2021).

[14] See, e.g., Matter of Nemis, 28 I. & N. Dec. 250 (BIA 2021) (adjudicating application of the categorical approach to the federal conspiracy statute, 18 U.S.C. § 371, and a visa-fraud provision, 18 U.S.C. § 1546(a)); Matter of Dikhtyar, 28 I. & N. Dec. 214, 214–16 (BIA 2021) (addressing the divisibility of the Utah state drug schedules by specific controlled substance); Matter of P-B-B-, 28 I. & N. Dec. 43 (BIA 2020) (addressing the divisibility with regard to Arizona’s criminalization of possession of a “dangerous drug”).

[15] See, e.g., Omargharib v. Holder, 775 F.3d 192, 196 (4th Cir. 2014) (“Although we generally defer to the BIA’s interpretations of the INA, where, as here, the BIA construes statutes [and state law] over which it has no particular expertise, its interpretations are not entitled to deference.” (alteration in original) (quoting Karimi v. Holder, 715 F.3d 561, 566 (4th Cir. 2013))); Ramos v. U.S. Att’y Gen., 709 F.3d 1066, 1069 n.2 (11th Cir. 2013) (“We owe no Chevron deference to the Board’s interpretation of the Georgia [criminal] statute, which the Board has no power to administer.”); Denis v. U.S. Att’y Gen., 633 F.3d 201, 208 (3d Cir. 2011) (“[I]f the issue turns on the meaning of a federal statute other than the INA, we possess the requisite expertise to interpret a federal criminal statute such that no deference is due.” (citation omitted)); see also Almeida v. Holder, 588 F.3d 778, 785 (2d Cir. 2009) (deferring to the BIA’s interpretation of “theft offense,” but reviewing questions of a particular crime defined by state law de novo).

[16] See 8 C.F.R. § 1003.1(g) (2021).

[17] For March 15, 2021, for instance, the Ninth Circuit issued 21 unpublished decisions and 3 published decisions. See Log of Unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021); Opinions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/opinions [https://perma.cc/8ST8-G668] (last visited Oct. 7, 2021). On March 12, 2021, when it issued no published decisions, it issued 13 unpublished decisions. See Log of unpublished Dispositions, U.S. Ct. for Ninth Cir., https://www.ca9.uscourts.gov/memoranda [https://perma.cc/B23P-8Z9X] (last visited Oct. 7, 2021).

[18] 8 U.S.C. § 1103(g)(2).

[19] 8 U.S.C. § 1103(a)(1).

[20] United States v. Mead Corp., 533 U.S. 218, 229 (2001).

[21] See 8 U.S.C. § 1158(d)(1) (“The Attorney General shall establish a procedure for the consideration of asylum applications filed under [this section].”); 8 U.S.C. § 1158(d)(5)(B) (“The Attorney General may provide by regulation for any other conditions or limitations on the consideration of an application for asylum not inconsistent with this chapter.”); 8 U.S.C. § 1158(b)(2)(B)(ii) (“The Attorney General may designate by regulation offenses that will be considered to be a [disqualifying crime for purposes of asylum eligibility].”); 8 U.S.C. § 1158(b)(2)(C) (“The Attorney General may by regulation establish additional limitations and conditions, consistent with this section, under which an alien shall be ineligible for asylum under paragraph (1).”). See generally Patrick J. Glen, In re L-A-C-: A Pragmatic Approach to the Burden of Proof and Corroborating Evidence in Asylum Proceedings, 35 Geo. Immigr. L.J. 1, 34 (2020) (“The asylum statute . . . does not generally mandate specific procedures for the consideration of asylum applications. Rather, that section is composed almost entirely of explicit and implicit delegations.”).

[22] See 8 U.S.C. §§ 1182(a)(4), 1227(a)(5) (stating public charge ground of inadmissibility and deportability, respectively); see also Field Guidance on Deportability and Inadmissibility on Public Charge Grounds, 64 Fed. Reg. 28,689, 28,691 (Mar. 26, 1999) (providing public guidance as to how the Immigration and Naturalization Service would apply the public charge grounds of removability); Inadmissibility on Public Charge Grounds, 84 Fed. Reg. 41,292, 41,315 (Aug. 14, 2019) (revising the Executive Branch’s interpretation of the public charge inadmissibility ground).

[23] See 8 U.S.C. § 1101(f) (listing per se categories that would involve a finding that the alien lacked good moral character, but allow the Attorney General discretion to conclude that aliens outside those categories also lack good moral character); see, e.g., 8 U.S.C. § 1229b(b)(1)(B) (requiring good moral character in order to establish eligibility for cancellation of removal).

[24] See 8 U.S.C. § 1227(a)(2)(E)(i); see also Matter of Velazquez-Herrera, 24 I. & N. Dec. 503, 504 (BIA 2008) (interpreting this provision).

[25] Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421 (1987).

[26] Id. at 448 (internal quotation marks and citations omitted).

[27] Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 424–25 (1999).

[28] 8 U.S.C. § 1103(a)(1).

[29] Aguirre-Aguirre, 526 U.S. at 424.

[30] Id. at 424–25.

[31] Id. at 425 (quoting Immigr. & Naturalization Serv. v. Abudu, 485 U.S. 94, 110 (1988)).

[32] See Patrick J. Glen, Interring the Immigration Rule of Lenity, 99 Neb. L. Rev. 533, 560–61 n.191 (2021) (noting that “[t]he statute is . . . replete with express delegations of decisional authority to the Attorney General,” and citing examples relating to asylum, cancellation of removal, and other forms of relief).

[33] See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”).

[34] United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 542 (1950) (emphasis added).

[35] Fiallo v. Bell, 430 U.S. 787 (1977).

[36] Id. at 792 (citation omitted).

[37] Hampton v. Mow Sun Wong, 426 U.S. 88 (1976).

[38] Id. at 101 n.21 (citation omitted).

[39] Scialabba v. Cuellar de Osorio, 573 U.S. 41, 75 (2014) (plurality opinion).

[40] See Negusie v. Holder, 555 U.S. 511, 523 (2009) (explaining that remand rules exist, in part, “because ‘ambiguities in statutes within an agency’s jurisdiction to administer are delegations of authority to the agency to fill the statutory gap in reasonable fashion. Filling these gaps . . . involves difficult policy choices that agencies are better equipped to make than courts.’” (quoting Nat’l Cable & Telecomm. Ass’n. v. Brand X Int. Serv., 545 U.S. 967, 980 (2005)). Justice Kennedy also noted that on remand,

[t]he agency’s interpretation of the statutory meaning of “persecution” may be explained by a more comprehensive definition, one designed to elaborate on the term in anticipation of a wide range of potential conduct; and that expanded definition in turn may be influenced by how practical, or impractical, the standard would be in terms of its application to specific cases. These matters may have relevance in determining whether its statutory interpretation is a permissible one.

Id. at 524; see also Gonzales v. Thomas, 547 U.S. 183, 186–87 (2006) (per curiam) (noting that on remand, “[t]he agency can bring its expertise to bear upon the matter; it can evaluate the evidence; it can make an initial determination . . . it can, through informed discussion and analysis, help a court later determine whether its decision exceeds the leeway that the law provides” (quoting Immigr. & Naturalization Serv. v. Ventura, 537 U.S. 12, 17 (2002) (per curiam))).

[41] See Wadhia & Walker, supra note 1, at 1231–32 (citations omitted).

[42] See id.

[43] See 8 U.S.C. § 1103(a), (g).

[44] See 8 C.F.R. § 1003.1(d)(1) (1958) (“The Board shall function as an appellate body charged with the review of those administrative adjudications under the Act that the Attorney General may by regulation assign to it.”); see also Delegation of Powers and Definition of Duties, Immigration and Naturalization Service, 5 Fed. Reg. 2454, 2454 (July 1, 1940) (“[T]he Board of Review of the Immigration and Naturalization Service shall have authority to exercise the powers of the Attorney General [in certain delineated cases].”).

[45] See 8 C.F.R. § 1003.1(d)(1)(ii) (“Board members shall exercise their independent judgment and discretion in considering and determining the cases coming before the Board.”).

[46] See Tefel v. Reno, 972 F. Supp. 608, 613 n.1 (S.D. Fl. 1997) (“[T]he decision of the BIA is not factually, nor legally, the decision of the Attorney General.”).

[47] See United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260, 266–67 (1954) (“[T]he word ‘discretion’ means that the recipient must exercise his authority according to his own understanding and conscience. In short . . . the Attorney General denies himself the right to sidestep the Board or dictate its decision in any manner.”).

[48] See 8 C.F.R. § 1003.1(h).

[49] Sanchez-Penunuri v. Longshore, 7 F. Supp. 2d 1136, 1149 (D. Colo. 2013).

[50] See Wadhia & Walker, supra note 1, at 1232.

[51] See Patrick J. Glen, Robert F. Kennedy and the Attorney General’s Referral Authority: A Blueprint for the Biden Administration, 49 Hofstra L. Rev. 719, 721 (2021) (noting how the Trump Administration utilized Attorney General referral to advance an immigration policy consonant with its campaign promises, including restricting eligibility for asylum, streamlining enforcement and adjudications to limit the possibilities of delay in removal proceedings, and focusing on detention of aliens both during and after the conclusion of removal proceedings).

[52] See generally id. (discussing Biden administration’s policy).

[53] See Deportation Proceedings for Joseph Patrick Thomas Doherty, 13 Op. O.L.C. 1, 8 (1989); Deportation Proceedings of Joseph Patrick Thomas Doherty, 12 Op. O.L.C. 1, 8 (1988); see also Alberto R. Gonzales & Patrick Glen, Advancing Executive Branch Immigration Policy Through the Attorney General’s Review Authority, 101 Iowa L. Rev. 841, 882–86 (2016) (providing background and context to the Attorney Generals’ decisions in Matter of Doherty).

[54] See R-A-, 22 I. & N. Dec. 906, 906 (Att’y Gen. 2001); see also Gonzales & Glen, supra note 53, at 886–89 (providing background and context to the Attorney General’s decision in Matter of R-A-).

[55] See A-T-, 24 I. & N. Dec. 617, 617 (Att’y Gen. 2008); see also Gonzales & Glen, supra note 53, at 861–63 (providing background and context to the Attorney General’s decision in Matter of A-T-).

[56] See Dorman, 25 I. & N. Dec. 485, 485 (Att’y Gen. 2011); see also Glen, supra note 51, at 760–71 (providing background and context to the Attorney General’s decision in Matter of Dorman).

[57] For more on the history of Attorney General’s decisions, see generally Gonzales & Glen, supra note 53 (focusing on decisions issued in the Bush II and Obama administrations, with reference to additional decisions in the Reagan, Bush I, and Clinton administrations); Glen, supra note 51, at 726–63 (addressing the decisions issued during the Kennedy, Clinton, and Obama administrations).

[58] See In re Castro-Tum, 27 I. & N. Dec. 187, 187 (Att’y Gen. 2018) (referring the case, directing briefing on specific issues, and inviting “interested amici to submit briefs”); In re A-B-, 27 I. & N. Dec. 227, 227 (Att’y Gen. 2018) (same); In re L-A-B-R-, 27 I. & N. Dec. 245, 245 (Att’y Gen. 2018) (same); In re M-G-G-, 27 I. & N. Dec. 469, 469 (Att’y Gen. 2018) (same); M-S-, 27 I. & N. Dec. 476, 476 (Att’y Gen. 2018) (same); In re Negusie, 27 I. & N. Dec. 481, 481 (Att’y Gen. 2018) (same); In re L-E-A-, 27 I. & N. Dec. 494, 494 (Att’y Gen. 2018) (same); In re Castillo-Perez, 27 I. & N. Dec. 495, 495 (Att’y Gen. 2018) (same); In re Thomas & Thompson, 27 I. & N. Dec. 556, 556 (Att’y Gen. 2019) (same); In re Reyes, 27 I. & N. Dec. 708, 708 (Att’y Gen. 2019) (same); In re A-M-R-C-, 28 I. & N. Dec. 7, 7 (Att’y Gen. 2020) (same).

[59] See Wadhia & Walker, supra note 1, at 1215–24.

[60] Although not a central focus of their argument, the Professors also argue that no “legal or policy expertise in immigration or foreign relations” is required to become a member of the Board. See Wadhia & Walker, supra note 1, at 1217–19. That is true in one sense, as a lack of such knowledge is not disqualifying, but untrue in another, as subject-matter expertise is an important part of the application process—a fact borne out by the complete job announcement that the Professors refer to only in part. See Immigration Judge, USAJobs, https://www.usajobs.gov/Get/Job?viewDetails/570894500 [https://perma.cc/ZU5H-49D7] (noting the requirement that applicants respond to the Quality Rating Factors, including “knowledge of immigration laws and procedures”). It is also worth noting that new hires are tested prior to the commencement of adjudicatory activities, ensuring there is a baseline of subject-matter specific expertise. See 8 C.F.R. § 1003.0(b)(1)(vi) (2021). Both Board members and immigration judges are required to undertake ongoing training “to promote the quality and consistency of adjudications.” See 8 C.F.R. §§ 1003.0(b)(1)(vii), 1003.1(a)(2)(i)(B) (2021). And throughout their time as adjudicators, Board members are assessed and evaluated to ensure they are properly discharging the functions of the office. See 8 C.F.R. §§ 1003.0(b)(1)(vi), 1003.1(a)(2)(i)(D) (2021).

[61] See Wadhia & Walker, supra note 1, at 1221.

[62] Id. at 1222.

[63] See H.R. Rep. No. 109-72, at 161–69 (2005) (Conf. Rep.).

[64] See, e.g., Matter of L-A-C-, 26 I. & N. Dec. 516, 518–24 (BIA 2015) (interpreting the REAL ID Act’s corroboration requirement in light of its professed purpose of rejecting contrary court of appeals’ precedent and adopting the Board’s rule); Matter of C-T-L-, 25 I. & N. Dec. 341, 343–48 (BIA 2010) (applying a similar reasoning in regard to the REAL ID Act’s amendments to the statutory burden-of-proof provisions for asylum and withholding of removal).

[65] Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, Div. C, 110 Stat. 3009-546 (1996) (codified as amended at 8 U.S.C. ch. 12).

[66] See H.R. Rep. No. 104-828, at 223–24 (1996) (Conf. Rep.) (citing Matter of Esposito, 21 I. & N. Dec. 1 (BIA 1995); Matter of Castro, 19 I. & N. Dec. 692 (BIA 1988); Matter of Ozkok, 19 I. & N. Dec. 546 (BIA 1988)).

[67] H.R. Rep. No. 104-828, at 213–14 (citing Matter of O-J-O-, 21 I. & N. Dec. 381 (BIA 1996)); see Patrick J. Glen & Alanna R. Kennedy, The Strange and Unexpected Afterlife of Pereira v. Sessions, 34 Geo. Immigr. L.J. 1, 10–12 (2019) (recounting changes to the discretionary-relief regime made by IIRIRA).

[68] H.R. Rep. No. 104-828, at 228 (citing Matter of Farias, 21 I. & N. Dec. 269 (BIA 1996)); see Glen, supra note 51, at 750–51 (noting IIRIRA’s reversal of Matter of Farias); Gonzales & Glen, supra note 53, at 890 (same).

[69] See, e.g., Matter of Cardenas-Abreu, 24 I. & N. Dec. 795, 796–802 (BIA 2009) (en banc) (interpreting IIRIRA’s statutory enactment of a definition for the term “convicted” in light of pre-1996 Board and federal court precedent, as well as legislative intent); Matter of Monreal-Aguinaga, 23 I. & N. Dec. 56, 58–63 (BIA 2001) (en banc) (interpreting the new “exceptional and extremely unusual hardship” requirement for cancellation of removal in light of the prior suspension-of-deportation standard and evident congressional intent in heightening the various eligibility criteria for discretionary relief from removal in IIRIRA).

[70] See Wadhia & Walker, supra note 1, at 1217, 1223–24.

[71] Sidney A. Shapiro, The Failure to Understand Expertise in Administrative Law: The Problem and the Consequences, 50 Wake Forest L. Rev. 1097, 1099 (2015).

[72] See Wadhia & Walker, supra note 1, at 1223 (noting that BIA members become specialists on immigration law by interpreting INA and immigration policies. Compared to federal judges, the agency developed “expertise in reconciling and accounting for conflicting evidence and arguments, disciplinary perspectives, political demands, and legal commands.” (citation omitted)).

[73] Id. at 1223–24.

[74] Convention Relating to the Status of Refugees, Art. 1(A)(2), July 28, 1951, 189 U.N.T.S. 137.

[75] See Wadhia & Walker, supra note 1, at 1223.

[76] See id. at 1202–03.

[77] See, e.g., SEC v. Chenery Corp., 332 U.S. 194, 203 (1947) (“[T]he choice made between proceeding by general rule or by individual, ad hoc litigation is one that lies primarily in the informed discretion of the administrative agency.”); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 808 n.29 (1978) (“The Commission has substantial discretion as to whether to proceed by rulemaking or adjudication.” (citation omitted)).

[78] See, e.g., United States v. Mead Corp., 533 U.S. 218, 229 (2001) (“We have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”) (emphasis added).

[79] 8 U.S.C. § 1158(b)(2)(B)(ii) (emphasis added).

[80] 8 U.S.C. § 1158(b)(2)(C).

[81] 8 U.S.C. § 1158(d)(5)(B).

[82] 8 U.S.C. § 1158(d)(1).

[83] See M-E-V-G-, 26 I. & N. Dec. 227, 227 (BIA 2014); W-G-R-, 26 I. & N. Dec. 208, 208 (BIA 2014).

[84] See generally Asylum and Withholding Definitions, 65 Fed. Reg. 76,588, 76,593–95 (Dec. 7, 2000) (proposing standards to govern when an alleged particular social group is cognoscible under the INA).

[85] See cases cited in supra note 83 (listing the Board’s most recent decisions on how and when a “particular social group” may be established, including historical overviews of the Board’s development of the “particular social group” analysis).

[86] See SEC v. Chenery Corp., 332 U.S. 194, 202 (1947) (explaining that not every principle can be immediately formed as a general rule because “[s]ome principles must await their own development, while others must be adjusted to meet particular, unforeseeable situations[,]” and therefore, concluding that “an administrative agency must be equipped to act either by general rule or by individual order. To insist upon one form of action to the exclusion of the other is to exalt form over necessity”).

[87] See Immigr. & Naturalization Serv. v. Cardoza-Fonseca, 480 U.S. 421, 448 (1987) (“There is obviously some ambiguity in a term like ‘well-founded fear’ which can only be given concrete meaning through a process of case-by-case adjudication. In that process . . . the courts must respect the interpretation of the agency . . . .” (internal quotation marks and citations omitted)); Immigr. & Naturalization Serv. v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (adopting a similar reasoning).

[88] See, e.g., Wadhia & Walker, supra note 1, at 1241.

[89] See id. at 1221–24 (focusing on the expertise of agency personnel directly involved with the drafting and consideration of legislation on behalf of the Executive Branch).

[90] See id. at 1222.

[91] See supra Part II.C.

[92] Wadhia & Walker, supra note 1, at 1222.

[93] Id.

[94] Id. at 1222–23.

[95] See supra Part II.B.

[96] E. Donald Elliot, Re-inventing Rulemaking, 41 Duke L.J. 1490, 1492 (1992); see David J. Barron & Elena Kagan, Chevron’s Nondelegation Doctrine, 2001 Sup. Ct. Rev. 201, 231–32 (2001) (“The more courts have required agencies to give detailed notice of proposed regulatory action to interest groups, the more pressure agencies have felt to complete the bulk of their work prior to the onset of the rulemaking process . . . the less flexibility they show during rulemaking.”); see also Wendy Wagner, Katherine Barnes & Lisa Peters, Rulemaking in the Shade: An Empirical Study of EPA’s Air Toxic Emission Standards, 63 Admin. L. Rev. 99, 123 (2011) (“The findings of this study [on the EPA’s adoption of the air toxic emissions standards] underscore both the accuracy and the importance of Professor Elliott’s remarks . . . [it] reinforce[s] the possibility that a great deal of interest group influence occurs outside of the glass box of notice and comment.”).

[97] Elliot, supra note 96, at 1492; see Stephanie Stern, Cognitive Consistency: Theory Maintenance and Administrative Rulemaking, 63 Pitt. L. Rev. 589, 597 (2002) (“The timing of rulemaking encourages agency lock-in by concentrating the bulk of decisionmaking in the pre-notice period. Notice occurs after the agency has completed substantial amounts of development, analysis, and review.”); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1669, 1775 (1975) (“Indeed, the content of rulemaking decisions is often largely determined in advance through a process of informal consultation in which organized interests may enjoy a preponderant influence.”).

[98] Elliot, supra note 96, at 1492–93 (listing several methods to secure public participation: “[F]rom informal meetings with trade associations and other constituency groups, to roundtables, to floating ‘trial balloons’ in speeches or leaks to the trade press, to the more formal techniques of advisory committees and negotiated rulemaking”).

[99] Id. at 1493; see Jud Mathews, Minimally Democratic Administrative Law, 68 Admin. L. Rev. 605, 636 (2016) (noting divergence between “the justifications that they will believe satisfy courts” and “the agency’s real reasons for acting” and concluding that “demanding judicial review standards contribute to the ossification of the rulemaking process, generating delays and reducing agencies’ functional rulemaking capacity”).

[100] See, e.g., Shoba Sivaprasad Wadhia, Symposium: Reflections on the Travel Ban Decision, SCOTUSblog (June 26, 2018, 5:02 PM), https://www.scotusblog.com/2018/06/symposium-reflections-on-the-travel-ban-decision [https://perma.cc/WL5S-VS9W] (noting that Professor Wadhia co-authored an amicus brief for scholars of immigration law on the history of the Immigration and Nationality Act in support of Hawaii in Trump v. Hawaii, 138 S.Ct. 2392 (2018)).

[101] See Wadhia & Walker, supra note 1, at 1236.

[102] Id. at 1241.

[103] Kisor v. Wilkie, 139 S. Ct. 2400 (2019).

[104] Wadhia & Walker, supra note 1, at 1241 (citing Processes and Procedures for Issuance and Use of Guidance Documents, 85 Fed. Reg. 63,200, 63200–04 (Oct. 7, 2020) (to be codified at 28 C.F.R. pt. 50)).

[105] See, e.g., Berger v. United States, 295 U.S. 78, 88 (1935) (“The United States Attorney is the representative not of an ordinary party to a controversy, but of a sovereignty . . . and whose interest, therefore, in a criminal prosecution is not that it shall win a case, but that justice shall be done.”); United States v. Young, 470 U.S. 1, 25 (1985) (Brennan, J., concurring in part and dissenting in part) (noting that the Supreme Court has “long emphasized that a representative of the United States Government is held to a higher standard of behavior” than private counsel) (alteration in original).

[106] This obviously happens infrequently, and when it does happen the issue is usually not a substantive interpretation of the agency but a jurisdictional determination by the court of appeals. See Mata v. Lynch, 576 U.S. 143, 147 (2015) (noting government’s agreement with the petitioner contra the holding of the court of appeals); Kucana v. Holder, 558 U.S. 233, 241–42 (2010) (same); see also Abdisalan v. Holder, 774 F.3d 517, 523 (9th Cir. 2014) (en banc) (noting the confluence of the petitioner’s and government’s arguments on rehearing).

[107] See generally Mark B. Stern & Alisa B. Klein, The Government’s Litigator: Taking Clients Seriously, 52 Admin. L. Rev. 1409 (2000) (noting that in representing agencies in the federal courts, Justice Department litigators are effectively representing “clients”).

[108] See, e.g., Wadhia & Walker, supra note 1, at 1241–42.

[109] See Pereira v. Sessions, 138 S. Ct. 2105, 2120–21 (2018) (Kennedy, J., concurring). Justice Alito stated in his dissent:

Here, a straightforward application of Chevron requires us to accept the Government’s construction of the provision at issue. But the Court rejects the Government’s interpretation in favor of one that it regards as the best reading of the statute. I can only conclude that the Court, for whatever reason, is simply ignoring Chevron.

Id. at 2121 (Alito, J., dissenting).

[110] See Patrick J. Glen & Kate E. Stillman, Chevron Deference or the Rule of Lenity? Dual-Use Statutes and Judge Sutton’s Lonely Lament, 77 Ohio St. L.J. Furthermore 129, 133–40 (2016).

[111] See Scialabba v. Cuellar de Osorio, 573 U.S. 41, 56–64 (2014) (applying deference to uphold the Board’s restrictive interpretation of the Child Status Protection Act); Holder v. Martinez Gutierrez, 566 U.S. 583, 591 (2012) (applying deference to uphold the Board’s interpretation of the continuous residence provision of the cancellation-of-removal statute).

[112] See Johnson v. Guzman Chavez, 141 S. Ct. 2271, 2291 (2021); Sanchez v. Mayorkas, 141 S. Ct. 1809, 1812 (2021); Niz-Chavez v. Garland, 141 S. Ct. 1474, 1480 (2021).

[113] Of the cases presenting some form of deference, the government prevailed in Guzman Chavez, 141 S. Ct. 2271 and Sanchez, 141 S. Ct. 1809, and lost in Niz-Chavez, 141 S. Ct. 1474. The government additionally prevailed in Garland v. Dai, 141 S. Ct. 1669 (2021), and Pereida v. Wilkinson, 141 S. Ct. 754 (2021).

[114] See United States v. Palomar-Santiago, 141 S. Ct. 1615, 1618 (2021).

[115] See Niz-Chavez, 141 S. Ct. at 1485.