FOR PUBLICATION
UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT
FELIPE CRUZ BETANSOS, No. 15-72347 Petitioner, Agency No. v. A077-310-010
WILLIAM P. BARR, Attorney General, Respondent. OPINION
On Petition for Review of an Order of the Board of Immigration Appeals
Argued and Submitted April 12, 2018 Pasadena, California
Filed July 5, 2019
Before: Carlos T. Bea and Mary H. Murguia, Circuit Judges, and Stanley Allen Bastian, * District Judge.
Opinion by Judge Murguia; Concurrence by Judge Murguia
* The Honorable Stanley Allen Bastian, United States District Judge for the Eastern District of Washington, sitting by designation. 2 BETANSOS V. BARR
SUMMARY **
Immigration
Denying Felipe Cruz Betansos’s petition for review of a decision of the Board of Immigration Appeals, the panel deferred to the BIA’s decision in Matter of Cortes Medina that a conviction for indecent exposure under California Penal Code § 314(1) is categorically a crime involving moral turpitude (“CIMT”) and held that Cortes Medina applied retroactively to Betansos’s case such that his § 314(1) conviction was a CIMT that made him ineligible for cancellation of removal.
In concluding that Betansos’s indecent exposure conviction under § 314(1) was a CIMT, the BIA relied on its published decision in Matter of Cortes Medina, 26 I. & N. Dec. 79 (BIA 2013). However, the BIA’s decision in Cortes Medina contradicted this court’s earlier decision, Nunez v. Holder, 594 F.3d 1124 (9th Cir. 2010), in which the court held that indecent exposure under § 314(1) was not categorically a CIMT. In Nunez, the court determined that the BIA’s unpublished decision did not merit deference and adopted a definition of moral turpitude that required the infliction of harm or the involvement of a protected class. In Cortes Medina, the BIA disagreed with Nunez’s generic definition as being too narrow, concluding that the defining characteristic of a CIMT in the indecent exposure context is whether the offense includes “lewd intent.”
** This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. BETANSOS V. BARR 3
The panel concluded that it must defer to Cortes Medina under National Cable & Telecommunications Association v. Brand X Internet Services, 545 U.S. 967 (2005). The panel noted that, unlike in Nunez, the BIA in Cortes Medina presented analysis explaining how it arrived at its generic definition of moral turpitude and explained why violations of § 314(1) are a categorical match to that generic definition. Because Cortes Medina did not misrepresent the authorities it relied on, it relied on published BIA authority, and its analysis was reasoned and thorough, the panel concluded that it could not say that the BIA’s decision was unreasonable.
Applying the five-factor retroactivity framework from Montgomery Ward &. Co., Inc. v. FTC, 691 F.2d 1322 (9th Cir. 1982), the panel also concluded that Cortes Medina applied retroactively to Betansos. The panel concluded that the first factor was not in play in this case, and that the fourth factor—the burden imposed by retroactive application— clearly favored Betansos, but that the fifth factor—the statutory interest in applying a new rule—leaned in the government’s direction. Noting that the second factor— whether the new rule represents an abrupt departure from well established practice—arguably favored Betansos, the panel concluded that overall the factors supported retroactive application because factor three—reliance on the new rule—weighed against Betansos. Specifically, the panel concluded that Betansos did not show that he in fact relied on Nunez prior to the BIA’s decision in Cortes Medina.
Specially concurring, Judge Murguia, joined by Judge Bastian, wrote separately to note a tension between the realities of criminal prosecutions and the tools the court applies in immigration cases involving the categorical 4 BETANSOS V. BARR
approach. Judge Murguia wrote that, because the vast majority—and nearly all—of criminal cases are resolved through plea bargains, a gap remains in the approaches for demonstrating a “realistic probability” of prosecution for conduct that falls outside the generic definition of a crime. Accordingly, Judge Murguia noted that it would be worth developing a mechanism for considering what conduct prosecutors charge and results in defendants accepting pleas.
COUNSEL
Robert Francis Jacobs (argued), Robert F. Jacobs & Associates, Santa Fe Springs, California, for Petitioner.
Erica Miles (argued), Trial Attorney; John W. Blakeley, Assistant Director; Office of Immigration Litigation, Civil Division, United States Department of Justice, Washington, D.C.; for Respondent. BETANSOS V. BARR 5
OPINION
MURGUIA, Circuit Judge:
Felipe Cruz Betansos, a native and citizen of Mexico, appeals the Board of Immigration Appeals’ (“BIA”) dismissal of his application for cancellation of removal. In dismissing his appeal, the BIA affirmed the Immigration Judge’s (“IJ”) conclusion that Betansos’s conviction for indecent exposure under California Penal Code § 314(1) is categorically a crime involving moral turpitude (“CIMT”). Betansos’s criminal record also includes a petty theft conviction, which he does not dispute is a CIMT. Therefore, if Betansos’s indecent exposure conviction is a CIMT, Betansos is ineligible for cancellation of removal because aliens convicted of two CIMTs are ineligible for cancellation of removal. 8 U.S.C. § 1229b(b)(1)(C).
In affirming the IJ’s determination that Betansos’s indecent exposure conviction is a CIMT, the BIA relied on its published decision in Matter of Cortes Medina, 26 I. & N. Dec. 79 (BIA 2013), which held that a conviction under § 314(1) is categorically a CIMT. Cortes Medina contradicts our 2010 decision, Nunez v. Holder, 594 F.3d 1124 (9th Cir. 2010), in which we rejected the BIA’s determination that § 314(1) is categorically a CIMT because the BIA decision we reviewed in Nunez rested entirely on an unproven statement that § 314(1) requires sexual motivation. Nunez, 594 F.3d at 1133. In Nunez, we held that indecent exposure under § 314(1) is not categorically a CIMT. Id. at 1138.
We must now decide whether to defer to the BIA’s more recent determination in Cortes Medina that a violation of § 314(1) categorically constitutes a CIMT. If we defer to Cortes Medina, we must also decide whether we will do so retroactively. For the reasons explained below, we conclude 6 BETANSOS V. BARR
that we must defer to Cortes Medina pursuant to the framework outlined in National Cable & Telecommunications Association v. Brand X Internet Services, 545 U.S. 967, 982 (2005). We also conclude that Cortes Medina applies retroactively in Betansos’s case. We therefore deny Betansos’s petition for review.
I.
Betansos entered the United States in 1989 and has resided here since. He is unmarried, but he lives with his longtime girlfriend with whom he has a 17-year-old daughter who is a United States citizen.
Betansos has five criminal convictions in California. Relevant here are his 1989 petty theft, Cal. Penal Code § 484(a), and 2002 indecent exposure, Cal. Penal Code § 314(1), convictions. 1
On April 22, 2010, the Department of Homeland Security (“DHS”) served Betansos with a Notice to Appear, which initiated his removal proceedings. At that time, Betansos was in custody. On May 11, 2012, Betansos, represented by counsel, admitted that he is not lawfully present in the United States and is a citizen of Mexico and conceded removability. On September 14, 2012, Betansos and his attorney appeared before the IJ to file his application for cancellation of removal. About a year later, in October 2013, the IJ held another hearing and denied Betansos’s application for relief.
1 Betansos was also convicted in 1994 for driving with a blood alcohol level of .08 or more, Cal. Veh. Code § 23152(b), and in 2003 and 2007, for domestic battery, Cal. Penal Code § 243(e)(1). BETANSOS V. BARR 7
A.
Betansos requested cancellation of removal as relief from deportation. To be eligible for cancellation of removal, Betansos had to demonstrate, among other things, that he was not convicted of certain enumerated offenses. 2 8 U.S.C. § 1229b(b)(1)(C).
In denying Betansos’s application for cancellation of removal, the IJ concluded that because Betansos had been convicted of two CIMTs—petty theft and indecent exposure—he was statutorily ineligible for cancellation of removal. 8 U.S.C. § 1229b(b)(1)(B), (C). The IJ relied on Cortes Medina to conclude that indecent exposure constitutes a CIMT, noting that Cortes Medina held that indecent exposure under § 314(1) includes “the element of lewd intent.” Cortes Medina was decided on January 8, 2013, after Betansos applied for cancellation of removal but before the IJ held the October 2013 hearing. Because Betansos sought no other form of relief, the IJ ordered Betansos removed to Mexico. Betansos timely appealed the IJ’s decision to the BIA.
B.
The BIA dismissed Betansos’s appeal on June 29, 2015. In dismissing the appeal, the BIA agreed with the IJ that Betansos was ineligible for cancellation of removal because he was convicted of two CIMTs. The BIA noted that
2 Betansos also had to demonstrate that: (1) he had been continuously physically present in the United States for not less than 10 years immediately before his application was filed, (2) he was a “person of good moral character” during that time, and (3) his removal would “result in exceptional and extremely unusual hardship” to his United States citizen daughter. 8 U.S.C. § 1229b(b)(1)(A)–(D). 8 BETANSOS V. BARR
Betansos did not contest that his petty theft conviction is a CIMT. Then, citing to Cortes Medina, the BIA affirmed the IJ’s conclusion that Betansos’s indecent exposure conviction was categorically a CIMT. In explaining why the BIA affirmed the IJ’s decision, the BIA noted that Betansos bears the burden of demonstrating he is eligible for relief. The BIA found that Betansos had not met his burden of showing that “under current law a realistic probability exists that California would apply the [indecent exposure] statute, either in his case or generically, to conduct that would not involve moral turpitude.” 3 In other words, Betansos failed to show that California would prosecute non-morally turpitudinous conduct under § 314(1). The BIA also highlighted that it found no published or unpublished California cases since Nunez applying § 314(1) to non- morally turpitudinous conduct. Accordingly, the BIA concluded that Cortes Medina applied, that the IJ did not err in relying on Cortes Medina, and that Betansos’s conviction under § 314(1) was a CIMT.
Betansos timely appealed the BIA’s decision.
II.
We lack jurisdiction to review a final order of removal based on a petitioner’s conviction of a CIMT. See Marmolejo-Campos v. Holder, 558 F.3d 903, 907 (9th Cir. 2009) (en banc) (citing 8 U.S.C. § 1252(a)(2)(C)). However, we retain jurisdiction to determine whether a petitioner’s
3 For a state statute of conviction to be categorically broader than the generic definition of a crime, “a realistic probability, not a theoretical possibility, [must exist] that the State would apply its statute to conduct that falls outside the generic definition of a crime.” Gonzales v. Duenas- Alvarez, 549 U.S. 183, 193 (2007). BETANSOS V. BARR 9
conviction is in fact a CIMT as defined in the Immigration and Nationality Act (“INA”). Id.
III.
Generally, when determining whether a petitioner’s conviction is categorically a CIMT, we undertake a two-step process. See Rivera v. Lynch, 816 F.3d 1064, 1070 (9th Cir. 2016) (citing Marmolejo-Campos, 558 F.3d at 907–11). “First, we identify the elements of the statute of conviction, reviewing the BIA’s conclusions on this point de novo.” Vinh Tan Nguyen v. Holder, 763 F.3d 1022, 1027 (9th Cir. 2014); see also Marmolejo-Campos, 558 F.3d at 907.
Second, after identifying the elements of the statute of conviction, we engage in the categorical approach and “compare the elements of the statute of conviction to the generic definition of a [CIMT] and decide whether the conviction meets that definition.” Castrijon-Garcia v. Holder, 704 F.3d 1205, 1208 (9th Cir. 2013). In so doing, “[w]e rely on our own generalized definition of moral turpitude, which divides almost all CIMTs into two basic types: those involving fraud and those involving grave acts of baseness or depravity.” Rivera, 816 F.3d at 1070 (internal quotation marks omitted).
However, our review of the BIA’s conclusion that a statute of conviction is categorically a CIMT is “governed by the same traditional principles of administrative deference we apply to the [BIA’s] interpretation of other ambiguous terms in the INA.” Marmolejo-Campos, 558 F.3d at 911. Accordingly, where “the [BIA] determines that certain conduct is morally turpitudinous in a precedential decision, we apply Chevron deference regardless of whether the order under review is the precedential decision itself or a subsequent unpublished 10 BETANSOS V. BARR
order that relies upon it.” Id. at 911. Under Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984), we defer to an agency’s interpretation of ambiguous language in a statute where Congress has delegated authority to the agency to enforce the statute containing the ambiguous text. Id. at 908–09. Nevertheless, where the BIA issues a precedential decision with “no reasoned explanation for its conclusion,” Chevron deference is unwarranted. Rivera, 816 F.3d at 1070. Instances where we do not defer to the BIA’s interpretation of ambiguous statutory language, however, are rare. Id. at 1071.
We first interpret the statute of conviction to identify its essential elements. In relevant part, Cal. Penal Code § 314(1) states, “[e]very person who willfully and lewdly . . . [e]xposes his person, or the private parts thereof, in any public place, or in any place where there are present other persons to be offended or annoyed thereby . . . is guilty of a misdemeanor.” Cal. Penal Code § 314(1), (2). California Jury Instructions provide that to find a defendant guilty under § 314, the government must prove:
(1) The defendant willfully exposed (his/her) genitals in the presence of another person or persons who might be offended or annoyed by the defendant’s actions; [AND] (2) [w]hen the defendant exposed (himself/herself), (he/she) acted lewdly by intending to direct public attention to (his/her) genitals for the purpose of sexually arousing or gratifying (himself/herself) or another person, or sexually offending another person. BETANSOS V. BARR 11
Judicial Council of Cal. Crim. Jury Instr. No. 1160, Indecent Exposure (2018). Further, under California law,
[A] person does not expose his private parts “lewdly” within the meaning of section 314 unless his conduct is sexually motivated. Accordingly, a conviction of that offense requires proof beyond a reasonable doubt that the actor not only meant to expose himself, but intended by his conduct to direct public attention to his genitals for purposes of sexual arousal, gratification, or affront.
People v. Archer, 119 Cal. Rptr. 2d 783, 785 (Ct. App. 2002) (citing In re Smith, 7 Cal. 3d 362, 366 (1972)); see also People v. Ballard, 16 Cal. Rptr. 2d 624, 630 (Ct. App. 1993) (“This requirement of lewdness, which is needed for a conviction of indecent exposure in California, supplies the assurance that a conviction for indecent exposure is one which necessarily involves moral turpitude.”); People v. Carbajal, 8 Cal. Rptr. 3d 206, 208 (Ct. App. 2003).
Based on the above authorities and reviewing de novo, it is clear that for a person to be convicted under § 314(1), the following three elements are required: (1) the person must have willfully exposed his or her genitals in the presence of others who could be “offended or annoyed”; (2) such exposure must have been sexually motivated, or “lewd”; and (3) the exposure must have been made with the intent to sexually arouse, gratify, or affront the offender or another person.
With this background in mind, we next consider whether the BIA’s conclusion in Cortes Medina that a violation of 12 BETANSOS V. BARR
§ 314(1) categorically constitutes a CIMT is reasonable and therefore entitled to Chevron deference. See Brand X, 545 U.S. at 980; Garfias-Rodriguez v. Holder, 702 F.3d 504, 507 (9th Cir. 2012) (en banc).
Under Chevron’s familiar two-step analysis, we first ask if Congress has directly spoken to the issue; step two asks whether the agency’s interpretation of ambiguous language in the statute the agency is charged with administering is reasonable. Garfias-Rodriguez, 702 F.3d at 508 n.2 (citing Chevron, 467 U.S. at 842–44). We have stated that the term “moral turpitude” “falls well short of clarity” and “is perhaps the quintessential example of an ambiguous phrase.” Marmolejo-Campos, 558 F.3d at 909; see also Nunez, 594 F.3d at 1130. Congress has not clearly defined “moral turpitude,” and we move to Chevron’s second step.
We have hesitated to defer to the BIA’s general understanding of the term “moral turpitude” because the BIA’s “general definition of moral turpitude fails to particularize the term in any meaningful way.” See Marmolejo-Campos, 558 F.3d at 910 (internal quotation marks omitted). Instead, “[w]e [have] rel[ied] on our own generalized definition of moral turpitude, which divides almost all CIMTs into two basic types: those involving fraud and those involving grave acts of baseness or depravity.” Rivera, 816 F.3d at 1071 (internal quotation marks omitted). Our understanding of moral turpitude, we have stated, does not differ materially from the BIA’s understanding. Marmolejo-Campos, 558 F.3d at 910.
In Cortes Medina, the BIA explicitly invoked its authority pursuant to Chevron and reaffirmed in Brand X to interpret ambiguous language in the INA, and re-assessed our definition of moral turpitude in Nunez. Cortes Medina, 26 I. & N. Dec. at 81. We turn to determining if the BIA’s BETANSOS V. BARR 13
reassessment was reasonable. We begin our analysis with Nunez.
1.
In 2010, we considered whether a conviction under the same state statute of conviction at issue here, § 314(1), is categorically a CIMT and concluded that it is not. Nunez, 594 F.3d at 1128. In so deciding, we reviewed the generic definition of moral turpitude in the indecent exposure context and determined that “actual infliction of harm or a protected class of victim” or both is required for a sexual offense to involve moral turpitude. Id. at 1132. We further explained that the conduct at issue needed to be more than just offensive in order for it to be morally turpitudinous. Id. at 1132–33. Indeed, we noted that contemporary sexual attitudes cannot dictate whether conduct is morally turpitudinous. Id. at 1132.
Importantly, because Nunez reviewed an unpublished BIA decision that provided scant analysis, and we defer to such decisions to the extent they have the “power to persuade,” we did not defer to the BIA’s unpersuasive and limited explanation. Id. at 1133 (citing Skidmore v. Swift & Co., 323 U.S. 134, 139 (1944)). The BIA only provided one paragraph of analysis that rested on an unsupported statement that because § 314(1) requires sexual motivation, a conviction under § 314(1) is a CIMT. Id.
Turning to how California has applied § 314(1), 4 Nunez recognized that public exposure is not necessarily lewd, id. 4 This discussion addressed the need to show that there is a “realistic probability, not a theoretical possibility, that the State would apply [the indecent exposure] statute to conduct that falls outside the generic definition of [moral turpitude].” Duenas-Alvarez, 549 U.S. at 193. 14 BETANSOS V. BARR
at 1133, and discussed California state court cases. Nunez divided California state court cases concerning § 314(1) into those dealing with “exposure for sexual gratification” and “exposure for sexual affront.” Id. at 1134–38. Addressing the exposure for sexual gratification cases, we identified a case where California courts upheld a conviction for violating § 314 for nude dancing at bars. Id. at 1135–36 (citing People v. Conway, 162 Cal. Rptr. 877 (Cal. App. Dep’t Super. Ct. 1979)). We concluded that “[w]hatever one’s view of the merits of [nude dancing], it is simply not base, vile, and depraved” and therefore California courts have applied § 314(1) to non-morally turpitudinous conduct. Id. at 1135–36. 5
Nunez also identified exposure-for-sexual-affront cases in California that fell outside the ambit of morally turpitudinous conduct. Id. at 1136–38. We noted two cases in which the conduct did not rise to a CIMT. Id. at 1137 (citing Archer, 119 Cal. Rptr. 2d at 786–87, and People v. Lionel M., No. H031030, 2007 WL 2924052 (Cal. Ct. App. 2007) (unpublished)). In both instances, the male defendant exposed his genitalia to female observers. Id. We determined that the conduct was “crass” and “inappropriate,” but not “inherently base, vile, and depraved.” Id. at 1138.
Based on these two types of cases, we concluded that because nude dancers and people who have made sexual insults have been convicted under § 314(1), there was a realistic probability that California would apply § 314(1) to 5 The dissent, however, challenged the majority’s reliance upon the nude dancing conviction in Conway because that case has been expressly disapproved by subsequent California courts, thereby undermining the conclusion that there was a realistic probability that California would apply § 314(1) to conduct falling outside the generic definition of a CIMT. Nunez, 594 F.3d at 1139 (Bybee, J., dissenting). BETANSOS V. BARR 15
non-morally turpitudinous conduct. Id. (citing Gonzales v. Duenas-Alvarez, 549 U.S. 183, 193 (2007)).
2.
Three years later in 2013, the BIA again addressed whether a conviction under § 314(1) categorically constitutes a CIMT, invoking its authority to revisit the question under Brand X. Cortes Medina, 26 I. & N. Dec. at 80–81. The BIA recognized that determining whether a state statute of conviction categorically is not a CIMT requires demonstrating that a “‘realistic probability, not a theoretical possibility,’ [exists] that the State would apply the statute to prosecute conduct that falls outside the definition of moral turpitude.” Id. at 82 (quoting Duenas-Alvarez, 549 U.S. at 193). While acknowledging that “there is not a single comprehensive definition of [CIMT],” the BIA identified what it considers the two “essential elements” of CIMTs: a culpable mental state and reprehensible conduct. Id. (citations omitted). Citing its own case law, the BIA stated that moral turpitude refers to “‘conduct which is inherently base, vile, or depraved, and contrary to the accepted rules of morality and duties owed between persons or to society in general.’” Id. (quoting Matter of Ajami, 22 I. & N. Dec. 949, 950 (1999)). According to the BIA, “lewd or lascivious intent” is required for indecent exposure to be morally turpitudinous. Id. The BIA, therefore, disagreed with Nunez’s generic definition of a CIMT in the indecent exposure context. Id. at 84 (explaining that the BIA considered Nunez’s definition of moral turpitude—which required the infliction of harm or the involvement of a protected class of victim—too narrow).
The BIA concluded that Nunez’s generic definition of a CIMT is too narrow after analyzing prior BIA case law. Pointing to Matter of P-, 2 I. & N. Dec. 117 (1944), and 16 BETANSOS V. BARR
Matter of Mueller, 11 I. & N. Dec. 268 (1965), Cortes Medina considered cases where conduct did not constitute CIMTs. Id. at 82–83. In Matter of P-, the BIA concluded that the indecent exposure offense at issue—indecent exposure in the presence of minor children in violation of Washington State law—was not a CIMT because the exposure was not intended to arouse sexual desires or made with lewd or lascivious intent. Matter of P-, 2 I. & N. Dec. at 119, 121. Likewise, in Matter of Mueller, the BIA determined the crime there—Mueller’s public exposure of his genitals in violation of Wisconsin law—did not constitute a CIMT because a conviction under the state statute at issue in that case could be secured without any demonstration of intent. Cortes Medina, 26 I. & N. Dec. at 83 (citing Matter of Mueller, 11 I. & N. Dec. at 270).
In contrast, the BIA pointed to Matter of Lambert, 11 I. & N. Dec. 340 (1965), as a case where a violation of a state statute constituted a CIMT. Id. Lambert was convicted under Florida state law for renting rooms knowing that the rooms would be used for lewdness or prostitution. Matter of Lambert, 11 I. & N. Dec. at 340. The BIA concluded that Lambert’s conviction was a CIMT. Id. at 342. Based on its review of these three cases, the BIA also concluded that a person convicted under § 314(1) commits a CIMT because a conviction under this statute requires a finding of “lewdness.” Cortes Medina, 26 I. & N. Dec. at 84.
The BIA, however, agreed with Nunez’s identification of two types of cases prosecuted under § 314(1) that illustrate whether violations of § 314(1) are categorically overbroad. Id. Discussing the “sexual affront” cases first, the BIA disagreed with our description in Nunez of the conduct in Archer as a “tasteless prank.” Id. at 84–85 (citing Archer, 119 Cal. Rptr. 2d 783). The BIA viewed the driver’s BETANSOS V. BARR 17
exposure of himself as a lewd act and therefore a CIMT. Id. As to the sexual gratification, or nude dancing, type of cases, here too the BIA focused on lewdness. Id. at 85. Differentiating between “simple public nudity” and “indecent exposure with a lewd intent,” the BIA determined that California courts require lewd intent to uphold a conviction. Id. (citations omitted).
Importantly, the BIA also considered whether there was a “realistic probability” that a person would be convicted under § 314(1) for “nude dancing or other conduct that does not involve moral turpitude.” Id. at 86. Indeed, the BIA cited to the California Supreme Court’s express disavowal of considering nude dancing a violation of § 314(1), and Cortes Medina’s failure to show that there is a realistic probability that California would apply § 314(1) to conduct not involving moral turpitude, to conclude that a realistic probability did not exist. Id. at 85–86 (citing Morris v. Municipal Court, 652 P.2d 51, 59, n.13 (Cal. 1982)). Therefore, the BIA concluded that a violation of § 314(1) is categorically a CIMT. Id. at 86.
Although Cortes Medina reviewed the same California state cases Nunez considered, the BIA arrived at a different conclusion because its generic definition of a CIMT is broader than the definition Nunez adopted. In other words, because Cortes Medina concluded that “lewd intent” makes indecent exposure a CIMT, the offenses in the California state cases discussed in Nunez as non-categorical matches of § 314(1) fit under the BIA’s broader definition once the BIA identified lewd intent in each of the California cases. Id. at 84–85 (citing Archer, 119 Cal. Rptr. 2d 783, and Conway, 162 Cal. Rptr. 877). 18 BETANSOS V. BARR
3.
Under Brand X, we must defer to the BIA’s interpretation of CIMT in Cortes Medina unless its conclusion is unreasonable. See Brand X, 545 U.S. at 982 (“A court’s prior judicial construction of a statute trumps an agency construction otherwise entitled to Chevron deference only if the prior court decision holds that its construction follows from the unambiguous terms of the statute and thus leaves no room for agency discretion.”). As is clear, the BIA’s conclusion in Cortes Medina contradicts our decision in Nunez. We are now tasked with deciding whether we should defer to Cortes Medina.
In Nunez, we determined that we defer to an unpublished BIA decision with limited reasoning “only to the extent that it has the ‘power to persuade.’” Nunez, 594 F.3d at 1133 (citing Skidmore, 323 U.S. at 139). The BIA’s unpublished decision there did not merit deference because we reviewed a decision that only provided one paragraph of analysis containing an unsupported statement “that because § 314 requires a sexual motivation, it is a crime of moral turpitude.” Id.
Unlike in Nunez, we are presented with Cortes Medina’s analysis explaining how the BIA arrived at its generic definition of moral turpitude. The key difference between Nunez and Cortes Medina is the BIA’s conclusion that the generic definition of moral turpitude in Nunez is too narrow. Cortes Medina, 26 I. & N. Dec. at 84. Instead, according to the BIA, the defining characteristic of a CIMT in the indecent exposure context is whether the offense conduct includes “lewd intent.” Id. at 83. The BIA arrived at this conclusion in Cortes Medina after considering BIA case law and supported it by reviewing California state court cases. Id. at 83–86. The BIA also explained why violations of BETANSOS V. BARR 19
§ 314(1) are a categorical match to its generic definition of a CIMT by reviewing the same cases we considered in Nunez. Cortes Medina does not misrepresent these authorities, its analysis is reasoned and thorough, and it relies on published BIA authority. Accordingly, we cannot say that the BIA’s decision is unreasonable. Cf. Mellouli v. Lynch, 135 S. Ct. 1980, 1989 (2015) (holding that the BIA’s decision was not owed deference because it “makes scant sense”); Rivera, 816 F.3d at 1071 (holding that the BIA’s decision was not entitled to deference where it provided “no reasoning whatsoever”); Coquico v. Lynch, 789 F.3d 1049, 1052–53 (9th Cir. 2015) (holding that the BIA’s decision was not entitled to deference where it misunderstood the elements of a crime under California law); Uppal v. Holder, 605 F.3d 712, 715 (9th Cir. 2010) (“Because the BIA failed to identify the elements of § 268 [aggravated assault] correctly, its CIMT analysis, in which it compares the elements it has identified to the generic definition of moral turpitude, is misdirected and so merits no deference from this Court.”).
Moreover, the fact that the BIA intended to provide an interpretation of moral turpitude in the indecent exposure context in Cortes Medina provides further support for deferring to Cortes Medina. See Lagandaon v. Ashcroft, 383 F.3d 983, 987 (9th Cir. 2004) (“As long as the BIA intended to issue an interpretation of a statute it enforces, its interpretation of ambiguities in that statute is generally accorded deference under [Chevron].” (internal quotation marks omitted)); Cortes Medina, 26 I. & N. Dec. at 81 (invoking Chevron to determine whether a violation of § 314(1) is a CIMT). In reaching this decision, we recognize that we defer to the BIA because it has exercised its delegated policymaking judgment, and not because it is better situated to interpret the INA. Garfias-Rodriguez, 702 F.3d at 515. Reasonable minds can differ when deciding 20 BETANSOS V. BARR
whether certain crimes are morally turpitudinous. Indeed, we did so in Nunez. However, pursuant to Brand X, we must defer to the BIA’s decision in Cortes Medina.
C.
Having concluded that we must defer to the BIA’s decision in Cortes Medina, we now consider whether our holding applies retroactively. In Garfias-Rodriguez, we held that when “we overturn our own precedent following a contrary statutory interpretation by an agency authorized under Brand X, we analyze whether the agency’s statutory interpretation (to which we defer) applies retroactively under the test we adopted in Montgomery Ward [&. Co., Inc. v. FTC, 691 F.2d 1322 (9th Cir. 1982)].” Id. at 520. We conduct this analysis on a case-by-case basis. Id.
The five-factor Montgomery Ward framework asks,
(1) whether the particular case is one of first impression, (2) whether the new rule represents an abrupt departure from well established practice or merely attempts to fill a void in an unsettled area of law, (3) the extent to which the party against whom the new rule is applied relied on the former rule, (4) the degree of the burden which a retroactive order imposes on a party, and (5) the statutory interest in applying a new rule despite the reliance of a party on the old standard.
Id. at 518 (quoting Montgomery Ward, 691 F.2d at 1333). This analysis, however, should be conducted with “the presumption of prospectivity” that accompanies exercises of legislative power because a court’s decision to defer to an BETANSOS V. BARR 21
agency’s decision under Brand X follows from the agency’s “exercise of delegated legislative policymaking authority.” Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1144 (10th Cir. 2016) (explaining that a court’s deference to agency interpretations of ambiguous statutes under Brand X, even after a court declares what the law is, is rooted in the assumption “that Congress had delegated legislative authority to the BIA to make a ‘reasonable’ policy choice in the face of [] statutory ambiguity”); see also Garfias- Rodriguez, 702 F.3d at 519 (“[B]ecause Chevron and Brand X are grounded in the deference we owe to agency policymaking, . . . the presumption in favor of retroactive application” does not apply.).
For the first Montgomery Ward factor, we have stated that whether an issue is one of first impression may not be well suited to the immigration context because this factor was developed in the National Labor Relations Board (“NLRB”) context, which differs significantly from immigration. Garfias-Rodriguez, 702 F.3d at 521 (explaining that because the NLRB is a unique agency that relies on the common-law method to adjudicate cases, immigration petitioners are not similarly situated to litigants in NLRB proceedings, and cases of “first impression” are captured in the second and third Montgomery Ward factors). Accordingly, this factor is inapplicable here.
“The second and the third factors are closely intertwined.” Id. In Garfias-Rodriguez we explained that “these two factors will favor retroactivity if a party could reasonably have anticipated the change in the law such that 22 BETANSOS V. BARR
the new ‘requirement would not be a complete surprise.’” Id. (quoting Montgomery Ward, 691 F.2d at 1333–34).
The second factor favors Betansos because Cortes Medina represents an “abrupt departure” from Nunez. Prior to Nunez, the BIA had no published opinion addressing whether a conviction under § 314(1) constituted a CIMT. See Nunez, 594 F.3d at 1133 (noting that the BIA’s decision deeming the petitioner’s § 314(1) conviction a CIMT was “an unpublished opinion that [did] not rely on prior precedential decisions”). In February 2010, Nunez established, in the first precedential opinion on the issue, that such a conviction does not constitute a CIMT. Id. at 1138. Nunez remained the authoritative, settled policy on this issue for almost three years, until the BIA decided Cortes Medina in January 2013 and took the exact opposite position—that a conviction under § 314(1) categorically does constitute a CIMT. See Cortes Medina, 26 I. & N. Dec. at 81 (explicitly invoking authority under Brand X to disagree with Nunez).
Cortes Medina is fairly characterized as a “complete surprise.” Garfias-Rodriguez, 702 F.3d at 521. The government has identified nothing that would have put Betansos on notice that relying on Nunez was unreasonable or risky. This is not a case where there was an ongoing conversation or a back-and-forth between this Court and the BIA about the proper interpretation. See Lemus v. Lynch, 842 F.3d 641, 649 (9th Cir. 2016) (“[Petitioner] was on notice that our approach was vulnerable based upon repeated contrary decisions, not only from the BIA but from other circuits as well.”). Nor is this a case where the former rule was only in place for a short period of time or the rule was subject to ongoing challenges or revisions. See Garfias- Rodriguez, 702 F.3d at 521–22 (discussing factors that diminish the reasonableness of reliance, including where the BETANSOS V. BARR 23
rule was only in place for six months, there were “multiple changes in the agency’s position regarding the proper rule,” or the rule was subject to “ongoing legal challenges”).
Nunez was well-settled policy in this Circuit from February 10, 2010, the date on which Nunez was decided, until January 8, 2013, the date on which Cortes Medina was decided. See Acosta-Olivarria v. Lynch, 799 F.3d 1271, 1276 (9th Cir. 2015) (holding that it was reasonable for the petitioner to rely on a Ninth Circuit rule because the rule was announced in a published opinion, “there was no contrary BIA decision[,]” and “[p]eople within the Ninth Circuit should be able to rely on our opinions in making decisions”); see also Garfias-Rodriguez, 702 F.3d at 515 (explaining that the former precedential Ninth Circuit rule “was authoritative in this circuit at least until the agency issued a reasonable interpretation to the contrary”).
In sum, Cortes Medina did not “fill a void.” Nunez had already filled the void, years earlier. Instead, Cortes Medina “abruptly departed” from Nunez, announcing a directly contrary interpretation without reasonably clear warning. Therefore, the second factor, analyzed in isolation, weighs in Betansos’s favor.
However, as noted above, the second and third factors are closely intertwined. And, here, the third factor—the extent to which Betansos relied on the former rule—weighs against Betansos.
To demonstrate reliance, Betansos must identify a specific “reliance interest.” See Garfias-Rodriguez, 702 F.3d at 522 (finding that the third factor weighed against the petitioner because he identified “only two specific reliance 24 BETANSOS V. BARR
interests” and neither was sufficient). In the context of a criminal conviction that has immigration consequences, we have held that reliance is presumed if the former, favorable rule was in place at the time the petitioner pleaded guilty or was convicted. Garcia-Martinez v. Sessions, 886 F.3d 1291, 1295 (9th Cir. 2018) (presuming, for purposes of retroactivity analysis, that the petitioner was aware of the relevant BIA interpretation at the time he pleaded guilty to theft crimes); see also I.N.S. v. St. Cyr, 533 U.S. 289, 322 (2001) (“There can be little doubt that, as a general matter, alien defendants considering whether to enter into a plea agreement are acutely aware of the immigration consequences of their convictions.”). But Betansos could not have relied on Nunez when he pleaded guilty in 2002 because Nunez had not yet been decided.
We have also held that a petitioner’s expenditure of fees in reliance on favorable, well-settled precedent may constitute a sufficient reliance interest. See Acosta- Olivarria, 799 F.3d at 1276 (holding that petitioner reasonably relied on the Ninth Circuit rule that made him eligible for adjustment of status because he applied for adjustment and paid the accompanying $1,000 fee, and “[r]etroactive application of the [BIA’s new rule] would cause [petitioner’s] application for adjustment of status to be denied, without any refund of the $1,000 fee”). But Betansos does not assert that he paid fees during his immigration proceedings in reliance on Nunez. Nor does Betansos assert that he made strategic decisions or chose not to apply for other forms of relief because he relied on the availability of cancellation of removal under Nunez. See id. (petitioner argued that he gave up the opportunity to voluntarily depart in reliance on the former rule making him eligible for adjustment of status). BETANSOS V. BARR 25
Indeed, Betansos’s entire argument regarding reliance is that he “clearly relied” on Nunez because “at all stages of proceedings before the IJ, until the October 31, 2013 merits hearing, [Betansos] was statutorily eligible for relief.” In other words, Betansos believes that “reliance” simply means that he cited the rule during his legal proceedings. This is not the type of specific reliance interest we have generally held sufficient. Betansos needed to identify a specific event or action that he took (or failed to take) in the past in reliance on Nunez that now carries new consequences or burdens under Cortes Medina. See Garfias-Rodriguez, 702 F.3d at 522 (“[R]etroactivity law . . . is meant to avoid new burdens imposed on completed acts, not all difficult choices occasioned by new law[.]”) (quoting Fernandez-Vargas v. Gonzales, 548 U.S. 30, 46 (2006)). Betansos has not done so.
In sum, although it would have been reasonable to rely on Nunez between February 2010 and January 2013 (under Montgomery Ward factor two), Betansos has not shown that he in fact relied on Nunez (under Montgomery Ward factor three).
4.
The fourth factor, the degree of the burden that a retroactive order imposes on a party, weighs in Betansos’s favor because “deportation alone is a substantial burden that weighs against retroactive application of an agency adjudication.” Garfias-Rodriguez, 702 F.3d at 523 (internal quotation marks omitted).
5.
Finally, the statutory interest in applying a new rule tips in the government’s favor because “non-retroactivity 26 BETANSOS V. BARR
impairs the uniformity of a statutory scheme, and the importance of uniformity in immigration law is well established.” Id. However, because the new rule announced in Cortes Medina does not follow from the “plain language of the statute,” this factor “only leans in the government’s direction.” Id.
6.
On balance, we find that Cortes Medina should apply to Betansos. The first factor is not in play. The fourth factor clearly favors Betansos. The fifth factor favors the government, but not strongly. And, although the second factor arguably favors Betansos, we have held that factors two and three are “intertwined.” Because factor three weighs against Betansos in this case, we hold that overall the factors support retroactive application against Betansos. See Garfias-Rodriguez, 702 F.3d at 518 (noting that the overall purpose of the Montgomery Ward test is to balance the agency’s interest in changing its rule against the “regulated party’s interest in being able to rely on the terms of a rule”) (emphasis added). 6 Accordingly, the IJ properly applied Cortes Medina and denied Betansos’s application for cancellation of removal because Betansos is statutorily ineligible. See 8 U.S.C. § 1229b(b)(1)(C).
IV.
We defer to the BIA’s decision in Cortes Medina that § 314(1) constitutes a CIMT. And we hold that Cortes
6 We note that the reliance analysis is highly fact dependent and conducted on a case-by-case basis. See Garfias-Rodriguez, 702 F.3d at 519–20. Although Betansos has not identified a specific reliance interest that arose for him during the period that Nunez was well-settled law, another petitioner might do so. BETANSOS V. BARR 27
Medina applies retroactively to Betansos’s case. We therefore deny Betansos’s petition.
PETITION DENIED.
MURGUIA, Circuit Judge, specially concurring, joined by BASTIAN, District Judge:
While Brand X requires us to defer to the BIA’s decision in Cortes Medina in the present case, I write separately to note a tension between the realities of criminal prosecutions and the tools we apply in immigration cases in which we undertake the categorical approach. This tension concerns the requirement that petitioners show a “‘realistic probability’ of prosecution for conduct that falls outside the generic definition” of a crime. Chavez-Solis v. Lynch, 803 F.3d 1004, 1009 (9th Cir. 2015) (quoting Gonzales v. Duenas-Alvarez, 549 U.S. 183, 193 (2007)).
The Supreme Court has clarified that in conducting the categorical inquiry:
[T]o find that a state statute creates a crime outside the generic definition of a listed crime in a federal statute requires more than the application of legal imagination to a state statute’s language. It requires a realistic probability, not a theoretical possibility, that the State would apply its statute to conduct that falls outside the generic definition of a crime.
Duenas-Alvarez, 549 U.S. at 193. We have explained that petitioners can make the requisite “realistic probability” 28 BETANSOS V. BARR
showing by pointing to state court decisions that apply the statute to broader conduct than permitted in the generic definition of a crime or by looking at the text of the state statute itself. See Chavez-Solis, 803 F.3d at 1009–10.
However, a gap remains in the two approaches we have so far endorsed for demonstrating that a “realistic probability” of prosecution exists. The vast majority—and nearly all—of criminal cases are resolved through plea bargains. 1 These agreements between prosecutors and defendants are not published, nor are they readily accessible for review, yet they would illuminate the possibly broader conduct for which individuals are prosecuted pursuant to various state statutes. If we are to determine accurately whether there is a “realistic probability” that a state would charge an offense for broader conduct than that permitted by the generic definitions of crimes, accounting for the vast majority of criminal prosecutions makes sense. We currently lack a mechanism for considering criminal plea bargains when conducting a categorical analysis. However, it is worth exploring how courts can account for plea deals. Developing a mechanism for considering what conduct prosecutors charge and results in defendants accepting pleas may be
1 See, e.g., United States v. Booker, 543 U.S. 220, 273–74 (2005) (Stevens, J., dissenting in part) (noting that over 95 percent of criminal prosecutions end in a plea bargain); U.S. Sentencing Comm’n, Overview of Federal Criminal Cases, Fiscal Year 2017, at 5 (2018), https://www.ussc.gov/sites/default/files/pdf/research-and-publications/r esearch-publications/2018/FY17_Overview_Federal_Criminal_Cases.p df (“In fiscal year 2017, the vast majority of offenders (97.2 [percent]) pleaded guilty. This high rate has been consistent for more than 15 years.”); Judicial Council of Cal., Court Statistics Report, at 114 (2016), http://www.courts.ca.gov/documents/2016-Court-Statistics- Report.pdf (in California, 96.8 percent of state criminal felony cases were resolved before trial, including 79.9 percent guilty pleas). BETANSOS V. BARR 29
particularly helpful in cases such as this one where the BIA relies on decades-old cases to assess whether present-day conduct is morally turpitudinous. See Matter of Cortes Medina, 26 I. & N. Dec. 79, 82–83 (2013) (discussing BIA decisions from 1944, 1956, and 1965 in analyzing what makes indecent exposure a crime of moral turpitude). Therefore, we should be careful to consider all information that could help us develop a full picture of what conduct states prosecute under particular statutes.