Bibi Alli & Matthew Blaisdell

The categorical analysis seeks to determine if a state offense (that the noncitizen was convicted of)  matches a federal offense (providing immigration consequences); each element of the state offense is compared to the generic definition of the federal offense. If the two do not match, the consequences do not apply. This strict approach has been long affirmed by the U.S. Supreme Court as a means to ensure ensures fairness as well as “judicial and administrative efficiency.” 

In our last two posts, we described the approach taken to determine the immigration consequences of criminal activity, and walked through a case where we recently applied this analysis. Today, we look at a decision handed down from Attorney General William Barr applying a mix-and-match approach, revamping the analysis in a manner that would make it far easier to apply these consequences.

The immigration judge and BIA affirm that the categorical analysis matches a state offense to a single federal offense

Matter of Reyes, 28 I&N Dec. 52 (A.G. 2020), examined a conviction of New York Penal Law (NYPL) § 155.20(1), grand larceny in the second degree, which criminalizes both theft and fraud. (We previously applies the categorical analysis to compare NY larceny statutes to a federal offense––crimes involving moral turpitude––here.)

A person is guilty under NYPL § 155.40(1) when he steals property and the value of the property exceeds $50,000. Under NYPL § 155.05, larceny is defined as “a person steals property and commits larceny when, with intent to deprive another of property or to appropriate the same to himself or to a third person, he wrongfully takes, obtains or withholds such property from an owner thereof.” There are five ways to commit larceny:

a) By conduct heretofore defined or known as common law larceny by trespassory taking, common law larceny by trick, embezzlement, or obtaining property by false pretenses; 

(b) By acquiring lost property; 

(c) By committing the crime of issuing a bad check; 

(d) By false promise; or

(e) By extortion. 

Following Ms. Reyes’ conviction, the Department of Homeland Security (DHS) initiated removal proceedings charging her with having committed  an aggravated felony. “The term [aggravated felony] applies to an offense described in this paragraph whether in violation of Federal or State law and applies to such an offense in violation of the law of a foreign country for which the term of imprisonment was completed within the previous 15 years.” INA § 101(a)(43). “An offense” has always been understood to refer to one individual ground of the 20 listed under § 101(a)(43).

Unusually, rather than pointing to a specific ground and alleging that it matched NYPL § 155.20(1), DHS instead alleged that NYPL § 155.20(1) matched two different grounds collectively: the theft ground at INA § 101(a)(43)(G) and the fraud ground at INA § 101(a)(43)(M)(i)­­––but not either one individually. We examined these grounds in more depth in a prior post here; for now it is sufficient to explain that the theft ground requires a taking without consent, and the fraud ground requires a taking with consent (see Matter of Garcia-Madruga).

Reyes argued that the NY offense does not categorically match either ground because it criminalizes takings both with and without consent, and is thus broader than both either ground on its own. DHS essentially argued that it was not necessary to categorically match either particular ground, as the conviction necessarily triggered one or the other. Because the Immigration Judge could not determine which ground the DHS was charging, proceedings were terminated. 

The DHS appealed to the Board of Immigration Appeals (BIA), which sought to address the issue of “[w]hether DHS can establish removability by charging an alien as an aggravated felon under two separate aggravated felony definitions, neither of which would independently be a categorical match to the statute of conviction, if all means of violating the statute fall within at least one of the charged aggravated felony definitions.” Meaning: if the state offense is so broad that it could include several different aggravated felonies, without categorically matching any one in particular, is it actually “an offense described” in INA § 101(a)(43)? The BIA determined that it may compare elements of the state offense only to one of the aggravated felony grounds, affirming the immigration judge and dismissing the DHS appeal.

Barr applies a brand new ‘mix and match’ approach

8 C.F.R. § 1003.1(h) gives the Attorney General the authority to certify BIA decisions for review. Past Attorney Generals have used this authority sparingly. However, within the past four years, the Attorney Generals appointed by the current administration has used this power frequently: each time in a manner dramatically restricting the relief available to noncitizens. 

Barr acknowledges the distinction between aggravated felony grounds listing “familiar crimes” (such as theft or fraud) versus others defining “categories of offenses.” He even acknowledges that, per Garcia Madrudga, the BIA “has therefore concluded that fraud and theft are mutually exclusive.” 

From here, though, Barr decides to walk off the reservation. 

For over 20 years, the categorical analysis has always been understood as matching the elements of a state offense to “one particular prong or another” of § 101(a)(43). Every Supreme Court case analyzing the categorical analysis have done so by matching the state offense to a single federal ground, and each time that the BIA has considered state offenses that might implicate multiple grounds, the BIA proceeded to match the offenses individually, one a one-to-one basis. 

In Reyes, Barr held instead that DHS could establish removability for commission of an aggravated felony by charging an individual with two separate aggravated felony grounds, neither of which would independently be a categorical match to the state offense, so long as all means of violating the state offense fall within “at least one” of the charged grounds. “[E]ven when a statute is not a categorical match to a particular aggravated felony offense, a conviction is still an aggravated felony if every means of violating the statute fall within multiple means of aggravated felony offenses . . . what matters is that a serious crime necessarily falls within the definition somewhere, not whether it falls within one particular prong or another.” 

In other words, rather than strictly matching a state offense with the generic definition of a federal offense, Barr is re-interpreting the analysis so that it is not restricted to determining whether there is a one-to-one match, but is instead opened up to allow a mix-and-match approach; the state offense can be broken up into different elements, with each one matched to a different aggravated felony ground.  

Barr essentially reasoned that, if the elements of the two aggravated felony fraud grounds were combined, they would cover all the means of committing larceny in NY [NYPL § 155.05 (a)–(e), described above].

One can immediately see how Barr’s rationale would contradict the Supreme Court’s objectives of fairness and of efficiency in justice and administration. DHS could find a categorical match by simply matching the individual elements of the state offense to not one just one federal identified aggravated felony, but anywhere across the 20 separate offenses identified as aggravated felonies under INA § 101(a)(43)(A)––effectively creating new ‘hybrid’ offenses identified not by the statutes written by Congress but by the imagination of DHS or the immigration judge in a manner that maximizes enforcement at the expense of fairness. (Former Chairman of the BIA, Paul Wickham Schmidt, makes this argument in more forthright terms.) 

Barr disputes this characterization: 

“DHS does not contend here that New York larceny should be viewed as a hybrid offense . . . that must satisfy the elements of both theft and fraud. Nor does DHS seek to fuse the criteria of section 101(a)(43)(G) with that of section 101(a)(43)(M)(i) to create a novel aggravated felony. What DHS does contend is that the respondent was convicted of an aggravated felony because, under New York law, her larceny conviction necessarily constitutes one of two listed offenses . . . “

Barr rewrites the physics of the categorical analysis

Yet, the NY conviction constitutes one of the two offenses only because it does “fuse” them––or rather, it superimposes them.

Imagine that the NY larceny law addresses the ‘stealing’ element as a qbit––a superimposition of the two binary states of 1 and 0. Congress, when drafting the 20 disparate offenses under INA § 101(a)(43), could have included a larceny ground that does the same. Instead, Congress sought to address the separate binary digits through separate grounds (say, for reasons of voltage or measurement): theft requires a 1, and fraud requires a 0.  (Or, extending the metaphor further, by drafting one ground that addresses entities that behave as particles and drafting another ground that addresses entities that behave as waves.) 

Said another way: imagine the NY offense had an element of ‘green.’ None of the aggravated felony grounds have an element of green. But one has an element of ‘blue,’ and another has an element of ‘yellow.’ Barr’s solution is to mix these 20 disparate offenses together until he’s able to come up with something that looks green. 

We can easily imagine how this approach would massively boost the ability of DHS to bring and sustain charges that in many respects would deport and permanently banish noncitizens, including lawful permanent residents, from the U.S. Presumably this dramatic result is exactly why Congress did not write broad ‘green’ laws as aggravated felonies––but rather numerous charges designed to capture more specific activity. 

Barr uses the certification process to circumvent Congress and the courts

Neither the fraud nor theft ground matches the NY larceny definition. Congress could have utilized broader language by including larceny among the “categories of offenses” Barr alluded to. Alternatively, Congress could have easily copied the language utilized by the NY larceny statutes to capture a wider swath of activity. It did neither, deciding instead to address larceny through separate grounds, distinguished by specific language to addressing different, specific activity. Barr’s approach would seem to treat these distinct grounds as mere suggestions or hypotheticals, rather than as defined, distinct grounds. 

It is not just Congress that Barr seeks to elide. The U.S. Supreme court could have adopted Barr’s reasoning in the numerous times that it has examined the categorical analysis over the years––instead each time emphasizing the restrictive nature of the analysis and holding the DHS to the burden that Congress intended. 

Barr criticizes the Supreme Court’s application of the categorical analysis as producing “outcomes that seem random or disconnected from reality” that “undermine the rule of law . . . But the Supreme Court has spoken.” Without irony, he would seek to reduce randomness be departing from the clarity of matching a state offense to “one” aggravated felony ground to matching a state offense to the rather more vague standard of mixing and matching the elements of a state offense to “a” ground, “at least one such offense,” or “one or more” of the 20 grounds––essentially, any hybrid that he would create to achieve his objectives in any particular case. 

He further claims that the application of the categorical analysis (following the Supreme Court’s decisions in Mathis and Descamps) leads to results in which “an alien who indisputably was convicted of an aggravated felony may evade that conclusion simply because there is a lack of clarity about which aggravated felony the conviction reflects.” Barr does not point to a single instance in which a court found a noncitizen to have been “indisputably” convicted of an aggravated felony but in which the charge could not be sustained due to lack of clarity about which aggravated felony. Moreover, Barr implies that any lack of clarity in charging is the result of flaws in the categorical analysis: not in the drafting of the laws by Congress or, much more appropriately, by the formulation of the charges by DHS.

Trying to have it both ways

Barr is at least aware that he cannot interpret statutes in a manner that openly conflicts with that of the U.S. Supreme Court. Throughout Reyes he continually criticizes and departs from the clear thrust of the Court’s rulings on the categorical analysis, while yet framing his holding as consistent with the same precedent. Barr paraphrases Reyes’ argument as requiring her conviction to “categorically map” on to either one or another grounds. Indeed, that’s exactly the interpretation of the categorical analysis repeatedly set forth by the Supreme Court. Rather than acknowledging this conflict, he minimizes the Court’s precedent as having something to do with the restrictions of the trial court and the fact that the decisions don’t directly address the issue as he has framed it: “a lack of clarity about which aggravated felony the conviction reflects.”  

To thread this needle, Barr relies (for perhaps the only time) on precedent from the Ninth Circuit, U.S. v. Becerril-Lopez, which held that a California conviction was categorically a “crime of violence” because it fell under either “robbery” or “extortion.” 

All well and good, except that it predates both Mathis and Decamps, as well as other rulings from the Supreme Court on the categorical analysis (Moncrieffe v. Holder and Mellouli v. Lynch) which, while not expressly addressing and overruling the rationale of Becerril-Lopez, consistently require a more restrictive approach to the categorical analysis than what Barr advocates. 

Even more confounding is the extent to which Barr extends this reasoning to past convictions as well, arguing that he is not announcing a new rule (in which event it could only be applied to convictions entered after his decision), but rather clarifying the manner in which the rule has always been understood. After criticizing the Supreme Court’s understanding and application of the categorical analysis, Barr abruptly notes that he has nevertheless not “departed from settled precedent or practice” simply because the Supreme Court has not explicitly ruled on this question. 

This finding is not without consequence, as it permits DHS to apply his rule retroactively to convictions pre-dating Reyes, in which defendants may have relied on the advice of counsel that was premised upon the prior, longstanding understanding of the categorical analysis: thus unwittingly pleading guilty to convictions that entail immigration consequences far more severe than they estimated when making the decision to plead.

Regarding Barr’s authority in interpreting the perceived ambiguity of a statutory term of both criminal and civil application, as well as his cursory retroactivity analysis, much more could be said. However, for our purposes here, we would direct you to this joint advisory issued by the Immigrant Defense Project, the Immigrant Legal Resource Center, and the National Immigration Project of the National Lawyers Guild.

Barr’s target is policy, not clarity

All of that said, Barr ultimately advances an argument for sustaining the DHS’ charge in a manner that essentially ignores all of the above by simply identifying a single charge––the theft ground. “The respondent further asserts that larceny by acquiring lost property “can be performed in the absence of consent, or by mistaken consent,” and therefore it does not satisfy the definition of theft as stated in Garcia-Madruga . . . Larceny by acquiring lost property categorically fits within aggravated-felony theft.” But issuing a decision essentially limited to NY larceny convictions is not Barr’s aim. (For a fuller discussion on these topics, see our prior posts cited above.)

In our opinion, Reyes is exemplary of the Trump administration’s many attempts to bypass the other branches of government to enforce impose its own form of immigration law––particularly, Barr’s attempts to utilize the courts within the Department of Justice, not as arbiters of fairness and justice, but as an instruments of ICE for maximum enforcement. 

To Barr’s credit, he’s consistently transparent in motive, if not subject. “[T]he categorical approach has proven to be particularly disruptive to the immigration system, because in a number of cases it has prevented the removal of aliens who have unquestionably been convicted of serious crimes, all because the manner in which a state legislature has phrased a particular criminal statute or a state court has interpreted its elements.” 

Barr is well aware of the many and varied classifications of immigration offenses that attach to criminal convictions and the severity of the penalties they impose. In line with the instincts of this administration, he instead seeks to impose his own simple definition of ‘serious crimes’ and to apply it as broadly as possible. Misleading is his attempt to pin the blame upon the states, which do not draft their criminal codes for immigration purposes. He knows well that this is the province of Congress; it is the states that make the easier target. 

They’re going to lose this one, too

While the Supreme Court has sided with the administration on cases related to its ability to regulate the entry of noncitizens into the U.S., it has tended to push back on attempts to circumvent the regulatory process, and it has been consistent in its formulation of the categorical analysis. We suspect it will do so again here.