Since I published this post, the Ninth Circuit has expressly endorsed the Fifth Circuit's reasoning and conclusion in Ali v. Barr. See Bare v. Barr, 975 F.3d 952, 968 (9th Cir. 2020). I'm using that news as an excuse to re-up this post because having reviewed my handiwork below with fresh eyes, I think I did a decent job of explaining why Judge Oldham's opinion is important and interesting. In my humble view, this case merits more attention from admin-law geeks, judges, and law clerks.
In Ali v. Barr, No. 17-60604 (5th Cir. Feb. 24, 2020) (Ali II) (Jones, Ho, Oldham), the Fifth Circuit held that an immigrant’s asylee status ends when he voluntarily adjusts to lawful permanent resident (LPR) status under 8 U.S.C. § 1159(b). The Court first confronted this issue in 2016, so we'll start our analysis there. See Ali v. Lynch (Ali I), 814 F.3d 306, 309 (5th Cir. 2016).
In Ali I, both parties argued that the unambiguous language of the statute supported their position. The BIA pointed to § 1159(b), which says the Attorney General “may adjust” an asylee “to the status of an alien lawfully admitted for permanent residence” if the asylee meets certain requirements. 8 U.S.C. § 1159(b). According to the BIA, adjusting “to” LPR status necessarily meant adjusting “away from” whatever underlying status an immigrant may have had.
For his part, Ali pointed to § 1158(c), which permits the Attorney General to terminate an immigrant’s asylee status for five specified reasons. 8 U.S.C. § 1158(c). Noting that adjustment to LPR status isn’t among them, Ali argued that the statute unambiguously foreclosed the BIA’s interpretation.
Emphasizing the “inconsistencies” between these two provisions, the Court held that the statute was ambiguous and that both parties had presented permissible interpretations. Ali I, 814 F.3d at 311. Rather than affirm the BIA’s interpretation at Chevron Step Two, however, the Court remanded for the BIA to consider the statutory interpretation question in greater depth. Id. at 312. Because the BIA had not recognized the statutory ambiguity, the Court explained, it hadn’t exercised its Chevron discretion in resolving the ambiguity in the first instance. Id.
On remand, the BIA addressed the issues Ali I had identified before reaching the same conclusion it had the first time. In a precedential opinion, the Board concluded that Ali’s asylee status ended the moment he voluntarily adjusted to LPR status. See Matter of N-A-I, 27 I. & N. Dec. 72 (BIA 2017). Ali then petitioned the Fifth Circuit for review once again.
A different panel (Jones, Ho, Oldham) addressed the issue in Ali II. The Court, per Judge Oldham, agreed with the BIA that § 1159(b) unambiguously provided that an immigrant’s aslyee status ends when he voluntarily adjusts to LPR status. In the Ali II panel’s view, Ali I was “best understood” as a Chevron Step Zero case. Step Zero, the Court explained, is “the initial inquiry into whether the Chevron framework applies at all.” But where, as here, “the agency has not yet offered its interpretation of the statute,” it reasoned, “Chevron Step Zero cannot be completed.” In that circumstance, “a remand is essential.”
In the discussion that follows, I’ll push back on Ali II's characterization of Ali I. First, I’ll lay out the Court’s rationale for viewing Ali I as a Step Zero case. Second, I’ll explain that while the Court may be right that Ali I is "best understood" as a Step Zero decision, it doesn't fit the traditional Step Zero mold. Third, I'll argue that Ali I is better understood as a Step One-and-a-Half case. Fourth, I’ll venture some guesses as to why the Court went with the Step Zero thesis anyway. And fifth, I’ll conclude with a brief discussion of a couple of other important aspects of Ali II.
I. Ali II's Rationale For Treating Ali I As A Step Zero Case
Judge Oldham’s opinion for the Court in Ali II concluded that Ali I is best understood as a Step Zero decision” because Ali I concluded that the BIA had not “interpret[ed] the relevant INA provisions in the first instance.” That, he argued, made it impossible for Ali I to “complete” the Step Zero analysis. As a result, the Ali I Court correctly concluded that “a remand [wa]s essential.”
Ali II also offered a process-of-elimination-style argument for its Step Zero thesis. Ali I must be a Step Zero case, the argument goes, because it can’t be a Step One or Step Two case. Ali I couldn’t have decided whether the statute was ambiguous (Step One), the Court explained, for two reasons. First, “Step One cannot be performed before Step Zero,” and as already mentioned, the BIA’s failure to offer its own interpretation of the Act prevented Ali I from completing Step Zero. Second, proceeding to Step One would have required the Court to interpret the Act in the first instance, which Ali I expressly refused to do because that was a task properly reserved for the BIA.
For similar reasons, Ali II refused to accept that Ali I found the statute ambiguous:
In a footnote, the Court added that “even if Ali I intended to find the INA provisions ambiguous,” Brand X “would leave the Board free to find those provisions unambiguously support its position.”
II. Problems With Ali II's Step Zero Thesis
Perhaps the Court is right that Ali I is “best understood” as a Step Zero decision, but I’m not convinced it actually is one. For one thing, as far as I can tell, Ali I didn’t cite United State v. Mead or otherwise discuss the two-part Step Zero standard, which asks whether (1) it appears that Congress delegated authority to the agency generally to make rules carrying the force of law, and (2) the agency interpretation claiming deference was promulgated in the exercise of that authority. Indeed, to the extent Ali I addressed Step Zero indirectly, it seems the panel was satisfied that no Step Zero problem existed:
Although Ali’s case was not designated as precedential by the BIA, the BIA relied on C–J–H–, which is a precedential BIA decision…. Because C–J–H– is precedential, we apply Chevron to the BIA’s interpretation of the INA that adjustment to LPR status terminates asylum status. Thus, we first determine if the INA leaves open whether an asylee’s adjustment to LPR status terminates his asylum status. See Chevron, 467 U.S. at 842-43, 104 S.Ct. 2778.
814 F.3d at 310. For the same reasons, I’m not convinced that the BIA’s “failure to interpret the relevant INA provision” prevented Ali I from completing the Step Zero inquiry. In my view, Ali I's emphasis on the BIA’s failure to exercise its “Chevron discretion” is grounded squarely in the Court’s finding that the alleged “inconsistency” between §§ 1158(c) and 1159(b) rendered the statute ambiguous. After all, if, as Ali II concluded, the first panel had not found the statute ambiguous, the BIA would have had no Chevron discretion to exercise. Had that been the case, the Ali I remand would have been nonsensical–not “essential."
III. Ali I As a Step One-and-a-Half Case
I strongly suspect that none of this was lost on the Ali II Court. Hence its qualified conclusion that Ali I is merely “best understood” as a Step Zero decision. But why would Ali II go to such lengths to force the “square peg” of Ali I into the “round hole” of Chevron Step Zero?
The Court certainly had other options. It might, for example, have concluded that Ali I is best understood as a Chevron Step-One-and-a-Half case. See Daniel J. Hemel and Aaron L. Nielson, Chevron Step One-and-a-Half, 84 U. Chi. L. Rev. 757 (2017). The Step One-and-a-Half doctrine recommends remand where an agency’s explanation for a statutory interpretation demonstrates that the agency failed to recognize a statutory ambiguity. Id. This makes sense, the argument goes, because “[i]f agencies are entrusted with discretionary power on the grounds that they are more accountable than courts, then judicial review should encourage agencies to take responsibility for their decisions.” Id. at 764.
Ali I appears to be a good example. As already explained, Ali I recognized an ambiguity in the statute’s text that the BIA had apparently overlooked. Thus, while the BIA’s interpretation might be a reasonable construction of the Act, the Court felt compelled to reject it because the agency hadn’t adopted it in the exercise of its Chevron discretion but instead based on its mistaken view that the statute was unambiguous and its hands were tied.
IV. Why Not Step One-And-a-Half?
So why didn’t the Court treat Ali I as a Step-One-and-a-Half decision? As I see it, there are two possible explanations. First, doing so would have meant acknowledging Ali I's determination that the statute is ambiguous. The Ali II panel obviously didn’t share that view and might have wanted to avoid attributing error to a prior panel unnecessarily. Or perhaps the Court wanted to shield its (and the BIA’s) view of the statute from any future Brand X vulnerability.
Second, it is also possible that the Ali II panel questions the legitimacy of the Step-One-and-a-Half doctrine. Consider, for example, this quote, which I discussed earlier:
Of course, it seems to me that’s precisely what Ali I did. Even if I’m wrong about that, though, it still strikes me as a fair description of most Step One-and-a-Half decisions. According to the Ali II panel, however, "[t]hat wouldn't be Chevron at all." Can this line of reasoning be read as anything other than hostile to the Step-One-an-a-Half doctrine?
V. The Best of the Rest
Ali also noted that the BIA’s interpretation conflicted with Department of Homeland Security regulations. Relying on Encino Motorcars, he insisted that the BIA should have acknowledged the apparent change in position and provided a reasoned basis for the shift in policy.
The Court rejected that argument, holding that Encino Motorcars only applies when an agency changes its own previously expressed view of the law. In a footnote, the Court added this interesting coda to its discussion of the BIA-DHS relationship:
And finally, the Court rejected Ali’s argument that the BIA’s interpretation misunderstood the INA’s legislative history. Along the way, the Court cited not one but two articles by some guy named “Christopher J. Walker.” It’s well-worth a read.