Interim Decision Number 3585
File A71 416 447. York.Board of Immigration Appeals Executive Office for Immigration Review U.S. Department of Justice
Decided September 28, 2007.
(1) A single ground for removal may require proof of a conviction tied to the statutory elements of a criminal offense, as well as proof of an additional fact or facts that are not tied to the statutory elements of any such offense.
(2) When a removal charge depends on proof of both the elements leading to a conviction and some nonelement facts, the nonelement facts may be determined by means of evidence beyond the limited “record of conviction” that may be considered by courts employing the “categorical approach,” the “modified categorical approach,” or a comparable “divisibility analysis,” although the record of conviction may also be a suitable source of proof, depending on the circumstances.
(3) Section 101(a)(43)(M)(i) of the Immigration and Nationality Act, 8 U.S.C. § 1101(a)(43)(M)(i) (2000), which defines the term “aggravated felony” to mean “an offense that involves fraud or deceit in which the loss to the victim or victims exceeds $10,000,” depends on proof of both a conviction having an element of fraud or deceit and the nonelement fact of a loss exceeding $10,000 that is tied to the conviction.
(4) Because the phrase “in which the loss to the victim or victims exceeds $10,000” is not tied to an element of the fraud or deceit offense, the loss determination is not subject to the limitations of the categorical approach, the modified categorical approach, or a divisibility analysis and may be proved by evidence outside the record of conviction, provided that the loss is still shown to relate to the conduct of which the person was convicted and, for removal purposes, is proven by clear and convincing evidence.
(5) The Immigration Judge erred in declining to consider a presentence investigation report as proof of victim loss because of his mistaken belief that he was restricted to consideration of the respondent’s record of conviction.
FOR RESPONDENT: Thomas M. Griffin, Esquire, Philadelphia, Pennsylvania.
FOR THE DEPARTMENT OF HOMELAND SECURITY: Rhonda M. Dent, Appellate Counsel.
BEFORE: Board Panel: OSUNA, Acting Chairman; FILPPU and PAULEY, Board Members.
FILPPU, Board Member:
This case concerns the scope of the evidence that an Immigration Judge may consider in removal proceedings to determine whether a conviction for an offense involving fraud or deceit was one “in which the loss to the victim or victims exceeds $10,000,” as required by section 101(a)(43)(M)(i) of the Immigration and Nationality Act, 8 U.S.C. § 1101(a)(43)(M)(i) (2000). We hold that an Immigration Judge is not restricted to “record of conviction” evidence but may consider any evidence admissible in removal proceedings bearing on the loss to the victim. Consequently, we will sustain the appeal of the Department of Homeland Security (“DHS”) from the Immigration Judge’s March 6, 2006, decision dismissing the aggravated felony charge against the respondent and terminating the removal proceedings.
I. FACTUAL AND PROCEDURAL HISTORY
The respondent, a native and citizen of Uzbekistan, has been a lawful permanent resident since 1998. In 2003, the respondent and various codefendants were indicted in the United States District Court for the Southern District of New York on four criminal counts arising from a scheme to defraud insurance companies, between July 2002 and January 2003, by staging automobile accidents, receiving unnecessary medical treatment, and submitting false and fraudulent claims for insurance benefits. On January 21, 2005, the respondent was convicted on a guilty plea of the offenses set forth in counts one and three of the indictment, charging him with conspiracy and mail fraud, respectively, under 18 U.S.C. §§ 371 and 1341 (2000). He was sentenced to a term of imprisonment of 10 months, followed by 3 years of supervised release, and was ordered to pay $19,850 in restitution pursuant to 18 U.S.C. § 3663 et seq.
The DHS has pursued charges of deportability against the respondent under section 237(a)(2)(A)(iii) of the Act, 8 U.S.C. § 1227(a)(2)(A)(iii) (2000), as an alien convicted of an “aggravated felony.” Specifically, the DHS alleged that the respondent was convicted of an offense that “involves fraud or deceit in which the loss to the victim or victims exceeds $10,000” under section 101(a)(43)(M)(i) of the Act, and of an attempt or conspiracy to commit such an offense under section 101(a)(43)(U). In support of the charge, the DHS submitted the criminal judgment, indictment, statement of reasons (“SOR”) by the sentencing judge, and presentence investigation report (“PSR”) into evidence.
The Immigration Judge dismissed the charge and terminated proceedings. There is no dispute that the respondent’s offenses involved “fraud or deceit.” See Valansi v. Ashcroft, 278 F.3d 203, 210 (3d Cir. 2002). The Immigration Judge found, however, that the DHS had failed to offer clear and convincing evidence that the offense involved more than $10,000 in loss to the victim. The Immigration Judge noted that neither the criminal statutes under which the
respondent was convicted nor the counts of the indictment made any loss a prerequisite to a finding of guilt.
The criminal judgment does identify the “total loss” arising from the respondent’s offenses as $19,850, the amount of restitution ordered, but it does not reflect how the trial court calculated this loss amount. However, the same figure appears in the PSR as the amount allegedly paid out in fraudulent claims, and, as indicated in the SOR, the trial court expressly adopted the PSR without change. Nevertheless, the Immigration Judge observed that the amount of restitution ordered is a “post-guilt calculus” aimed at making a victim whole and is not necessarily representative of actual loss arising from the offenses of conviction. The Immigration Judge further found that consulting a PSR to determine the amount of loss would not comport with the “categorical approach” to statutory interpretation set forth in the Supreme Court’s decisions i Taylor v. United States, 495 U.S. 575 (1990), and Shepard v. United States, 544 U.S. 13 (2005). See also Gonzales v. Duenas-Alvarez, 127 S. Ct. 815 (2007) (discussing the categorical approach as applied to aggravated felony determinations under the immigration law). Finally, the Immigration Judge emphasized that the record contains no evidence, such as a plea agreement or plea colloquy transcript, reflecting that the respondent admitted that his criminal conduct resulted in any particular amount of loss.
In its appellate brief, the DHS argues that the award of $19,850 in restitution is clear and convincing evidence that the offenses of which the respondent was convicted occasioned losses of more than $10,000 to his victims. The DHS also argues that the PSR is admissible to prove the amount of loss, even under Taylor and Shepard, because the district court judge affirmatively adopted it when imposing the sentence. At oral argument, the DHS further contended that the amount of loss arising from a fraud or deceit offense need not be established categorically, by reference to documents conventionally understood as constituting the “record of conviction,” but may instead be proven by reference to any evidence that is otherwise admissible, including the testimony of witnesses or the admissions of the respondent taken by the Immigration Judge during the removal proceedings. Noting that it has been unable to identify any State or Federal fraud statute that includes as an element the requirement that a victim sustain losses exceeding $10,000, the DHS claims that the categorical approach embodied in Taylor an Shepard is simply unworkable as applied to the loss calculation required by section 101(a)(43)(M)(i) of the Act.
The respondent, in both his appellate brief and at oral argument, claimed that the restitution award is not evidence of the actual loss that he was convicted of causing to his victims because that award included losses that arose from other conduct as well. Moreover, the respondent contends that the district court’s findings as to victim loss in the sentencing context are not sufficiently reliable to establish deportability because those findings need only have been supported
by a preponderance of the evidence, rather than the “clear and convincing” evidence needed for deportability.
II. ANALYSIS A. Introduction and Summary
The questions before us turn in large measure on the nature of the determinations that section 101(a)(43)(M)(i) requires and on whether an Immigration Judge must apply the so-called “categorical” approach when seeking to ascertain the amount of loss arising from a conviction for an offense that “involves fraud or deceit.” The court of appeals case law addressing the “categorical” and “modified categorical” approaches contains conflicting views over the scope and applicability of these concepts. In the end, however, we believe a very basic principle governs.
Simply put, the categorical and modified categorical approaches, as we understand Taylor and Shepard, properly apply only when the statute currently being implemented or administered demands a focus exclusively on the elements of a prior conviction. Further, neither Taylor no Shepard demands the use of the categorical or the modified categorical approach to any currently required determination that is not tied to an element of a prior conviction.
Here, the removal provision demands a prior conviction for fraud or deceit. But, as we explain, the statute also requires a separate finding as to loss that is not tied to the elements of any State or Federal criminal statute. The categorical and modified categorical approaches properly govern the assessment as to whether the elements of the conviction for fraud or deceit are present, but they do not apply when assessing the additional “nonelement” factor of victim loss.
The proper focus is on the nature of the findings that must be made to resolve a removal charge. Reliance exclusively on the categorical and modified categorical approaches is both necessary and proper if the removal charge is tied solely to the elements of an earlier conviction. But if something
more is demanded by the terms of the removal provision at issue, then the categorical and modified categorical approaches will not apply to this additional aspect of the removal statute. Substantial confusion has arisen, however, from both the case law and the multiple evidentiary purposes that can be served by conviction records. For example, some nonelement determinations are easily and appropriately made from records of conviction, such as the length of sentence the defendant receive after being found guilty of all the elements of the crime. The proper use of the conviction record to assess nonelement factors, such as length of sentence, mimics a categorical or modified categorical approach as set forth in Taylor and Shepard. However, it is really a look at the conviction record as a piece of ordinary evidence that discloses the length of sentence and is not an inquiry into what the trier of fact was required to find to determine guilt or innocence.
B. Supreme Court’s Categorical Approach
Under the Supreme Court’s categorical approach, an adjudicator seeking to determine whether a defendant’s prior “convictions” justify an enhanced sentence on a new conviction “must look only to the statutory definitions of the prior offenses,” and may not “consider other evidence concerning the defendant’s prior crimes,” including “the particular facts underlying [the] convictions.” Taylor v. United States supra, at 600 (addressing the meaning of “burglary” as defined in 18 U.S.C. § 924(e)(2)(B)(ii)). Thus, the categorical approach focuses on “whether the elements of the offense are of the type that would justify its inclusion” within the scope of a federally listed term, “without inquiring into the specific conduct of this particular offender.”James v. United States, 127 S. Ct. 1586, 1594 (2007); see also Shepard v. United States, supra,
at 19 (stating that “the categorical approach . . . refers to predicate offenses in terms not of prior conduct but of prior `convictions’ and the `element[s]’ of crimes” (quoting Taylor v. United States supra, at 600-01)).
The Supreme Court nonetheless indicated that the “categorical approach . . . may permit the sentencing court to go beyond the mere fact of conviction in a narrow range of cases where a jury was actually required to find all the elements” of the currently relevant offense, such as generic burglary at issue in Taylor. Taylor v. United States supra, at 602. The Supreme Court has noted that some lower courts refer to this second “step of the Taylor inquiry,” pertaining to which elements of an overly broad statute led to conviction, “as a `modified categorical approach.'” Gonzales v. Duenas-Alvarez, supra, at 819. Yet some courts have also considered the “modified categorical approach” to include a search for criminal record facts that do not pertain to the “elements” of the predicate offense but that can nonetheless be deemed reliably established by the conviction record. However, bot Taylor and Shepard confine a review of the conviction record to such items as charging documents, jury instructions, “the terms of a plea agreement or transcript of colloquy between judge and defendant in which the factual basis for the plea was confirmed by the defendant, or to some comparable judicial record of this information,” and do so as part of a search for the elements that led to a prior conviction, not th facts that were involved in the crime. Shepard v. United States supra, at 26.
Similar to the sentence enhancements at issue in Taylor an Shepard, deportability under section 237(a)(2)(A)(iii) of the Act depends on the existence of a “conviction” for a federally defined category of offenses, i.e., “aggravated felonies.” Hence, the courts of appeals have a long-standing practice of applying the categorical approach to aggravated felony determinations under the immigration law, a practice acknowledged in Gonzales v. Duenas-Alvarez, supra, at 818-19. We have also traditionally applied an analysis that closely resembles the categorical approach to determine whether an alien has a “conviction” that falls within a federally defined category of crimes leading to deportation. See Matter of Gertsenshteyn, 24 IN Dec. 111, 112 (BIA 2007); Matter of Sweetser, 22 IN Dec. 709, 715 (BIA 1999) Matter of Pichardo, 21 IN Dec. 330, 335 (BIA 1996); Matter of Madrigal, 21 IN Dec. 323, 327 (1996); Matter of Ghunaim, 15 IN Dec. 269, 270 (BIA 1975), modified on other grounds, Matter of Franklin, 20 IN Dec. 867 (BIA 1994); Matter of S-, 2 IN Dec. 559 (C.O., BIA 1946; A.G. 1947); Matter of S-, 2 IN Dec. 353, 357 (BIA, A.G. 1945) Matter of N-, 1 IN Dec. 181 (BIA 1941). Indeed, in Matter of Sweetser, supra, at 715, we characterized our approach to “divisible” statutes as being “identical to how the federal courts have applied the categorical approach.” Our unpublished cases undoubtedly include decisions in which we applied a “modified categorical approach” to search for conviction record “facts” that
may not have been “elements” necessary for conviction under the pertinent criminal statute. However, our published law applies either a categorical or a divisibility analysis, where the actual elements leading to conviction are the determining factor for removal charges hinging on a conviction for a crime. See Matter of Pichardo supra, at 335 (emphasizing that it is the elements of the crime an alien is actually convicted of, and not the crime he or she may have committed, that is determinative of deportability).
Yet, as recognized by the United States Court of Appeals for the Third Circuit, in which this case arises, the applicability of the categorical approach in the aggravated felony context necessarily depends on the language of both the statute of conviction and the particular aggravated felony category to which that approach might be applied. Singh v. Ashcroft, 383 F.3d 144, 161 (3d Cir. 2004) (stating that whil Taylor‘s “formal categorical approach presumptively applies in comparing [predicate offenses and aggravated felony categories], under certain conditions, both the enumerating statute and the statute of conviction can require a departure from the formal categorical approach”). Moreover, according to Singh, the reference in section 101(a)(43)(M)(i) of the Act to $10,000 in loss to the victims is the “prototypical example” of such a nonelement qualifier that would “invite inquiry into the facts underlying the conviction at issue,” because “it expresses such a specificity of fact that it almost begs an adjudicator to examine the facts at issue.” Id.
We, too, have recognized that application of the categorical approach depends on the nature of the determinations required by the statute in the aggravated felony context. For example, in Matter of Gertsenshteyn, supra, we considered section 101(a)(43)(K)(ii) of the Act, which provides aggravated felony treatment for an offense “described in section 2421, 2422, or 2423 of title 18, United States Code (relating to transportation for the purpose of prostitution) if committed for commercial advantage.” Because only one of the three enumerated statutes employs the phrase “commercial advantage” as an element (and none had done so at the time of enactment of section 101(a)(43)(K)(ii)), we determined that a strict application of the categorical or modified categorical approach would likely render the provision a nullity. Concluding that such an eventuality was inconsistent with legislative intent, we held that the parties could offer evidence outside the limits of a “record of conviction” in proving the “committed for commercial advantage” component of that aggravated felony charge. Id. at 115-16.
The extent to which the categorical approach applies to the determination of loss arising from a conviction for an offense that “involves fraud or deceit” under section 101(a)(43)(M)(i) of the Act thus depends on whether this immigration provision envisions loss as an element of the criminal offense or as a separate factor for assessment in relation to removability. To answer this question, we look to the statute.
C. Section 101(a)(43)(M)(i) of the Act
We begin with the statutory language. INS v. Cardoza-Fonseca, 480 U.S. 421, 431 (1987); INS v. Phinpathya, 464 U.S. 183, 189 (1984). Section 101(a)(43)(M) of the Act provides that an aggravated felony includes
an offense that —
(i) involves fraud or deceit in which the loss to the victim or victims exceeds $10,000; or
(ii) is described in section 7201 of the Internal Revenue Code of 1986 (relating to tax evasion) in which the revenue loss to the Government exceeds $10,000.
Viewed in isolation, the “loss to the victim” language of section 101(a)(43)(M)(i) could refer to an “element” of the underlying fraud offense. On the other hand, the victim loss aspect of that section can be considered as a “limiting” fact, unrelated to the statutory elements of the underlying offense, that excludes less serious fraud crimes from being aggravated felonies. When viewed in context, however, it is evident that the victim loss aspect of the statute was not intended to refer to an “element” of any underlying fraud offense.
Many of the aggravated felony provisions within section 101(a)(43) of the Act identify some limiting fact that excludes otherwise-qualifying crimes from the scope of the definition. For example, in sections 101(a)(43)(J), (Q), and
(T), which provide that an offense is an aggravated felony only if a certain minimum sentence may be imposed for the crime, the qualifier is the statutory penalty available for the offense of conviction. A number of other provisions limit crimes based on facts relating to the particular conviction at issue, rather than to the generi offense of conviction. Specifically, sections 101(a)(43)(F), (G), (P), (R), and (S) exclude crimes from the aggravated felony definition where the actual sentence imposed was too short (regardless of the sentence that could have been imposed), while sections 101(a)(43)(K)(ii), (N), and (P) provide that convictions are included in, or excluded from, the aggravated felony definition based on the particular offender’s motivation for committing the crime. I n Matter of Gertsenshteyn, supra, we noted that Congress used the length of sentence as one of the “aggravating” factors to distinguish ordinary crimes from aggravated felonies, even though the sentence resulting from a conviction is not an element of the crime itself.
Three other provisions-including the one at issue here-provide that an offense is an aggravated felony only if the financial implications of the crime were sufficiently far-reaching. See sections 101(a)(43)(D), (M)(i), (M)(ii) of the Act (covering, respectively, money laundering offenses or offenses involving the proceeds of specified unlawful activity in which “the amount of the funds exceeded $10,000,” offenses involving fraud or deceit “in which the loss to the victim or victims exceeds $10,000,” and tax evasion offenses “in which the revenue loss to the Government exceeds $10,000”).
Most importantly, both sections 101(a)(43)(M)(i) and (ii) depend in part on the “loss” occasioned by the offense of conviction. Yet section 7201 of the Internal Revenue Code, 26 U.S.C. § 7201, referenced in clause (ii), contains no element pertaining to revenue loss to the Government. Consequently, no violation of section 7201 itself could ever be an aggravated felony if deportability under clause (ii) depended on a showing that the revenue loss to the Government was an element of the crime, which is an absurd result. The sensible reading of section 101(a)(43)(M)(ii), then, is that the amount of revenue loss to the Government is an aggravating factor, similar to the length of sentence or the “commercial advantage” requirement at issue in Matter of Gertsenshteyn, supra, that may be proven by reference to evidence unrelated to the statutory elements of the tax evasion offense. In our view, the reference to victim loss in clause (i) serves the same purpose as the “revenue loss” aspect of clause (ii). Our conclusion is bolstered by the fact that we have been unable
to identify, despite an extensive (if not exhaustive) inquiry, any Federal or State fraud statute that contains an element requiring loss to the victims exceeding $10,000.
Furthermore, our review of the legislative history of section 101(a)(43)(M) has uncovered nothing that would lead us to conclude that Congress understood the loss amount to be a required element of the predicate offense, as opposed to a limiting provision or an aggravating factor. Similarly, we have found nothing in the legislative history accompanying the 1996 change from the original $200,000 figure to the current $10,000 figure that suggests the newer figure was intended to be tied to the statutory elements of fraud or deceit convictions. See generally H.R. Conf. Rep. No. 104-828 (1996), available at 1996 WL 563320.
The Federal courts of appeals, including the Third Circuit, have found other aspects of section 101(a)(43)(M) to be plain and unambiguous See Alaka v. Att’y Gen., 456 F.3d 88, 106 (3d Cir. 2006) (stating that the statute was plain and unambiguous in relation to the need to predicate “`removal on a convicted offense resulting in losses greater than $10,000,'” as opposed to looking to losses arising from additional fraudulent activity that did not result in conviction (quotin Knutsen v. Gonzales, 429 F.3d 733, 736 (7th Cir. 2005))). Moreover, the circuits have consistently applied some form of categorical analysis when seeking to ascertain whether the offense in question had “fraud or deceit” as an element. E.g., Valansi v. Ashcroft, supra, at 214.
But we are not aware of any precedential majority opinion declaring that the victim impact aspect of the statute, i.e., a loss exceeding $10,000, must
correspond to a loss element in the criminal statute. See Li v. Ashcroft, 389 F.3d 892, 899-901 (9th Cir. 2004) (Kozinski, J., concurring) (arguing that the lack of an element in the criminal statute requiring a loss exceeding $10,000 should preclude a finding of an aggravated felony conviction under section 101(a)(43)(M)(i)). Indeed, most published court of appeals rulings seem to accept that a loss exceeding $10,000 need not be a formal element of the criminal offense. And as reflected above, our own construction of the statute is that the $10,000 loss threshold is a limiting or aggravating factor that need not be tied to an element of any criminal statute.
D. Modified Categorical Approach Under Circuit Law
Our conclusion that the $10,000 loss figure in section 101(a)(43)(M)(i) of the Act was not intended to describe an “element” of a “fraud or deceit” crime takes this victim loss aspect of the statute outside the scope of the categorical approach of Taylor an Shepard. Nevertheless, the weight of authority at the circuit court level is that proof of such loss must be derived from the same types of evidence (such as the charging instrument or plea agreements) that are typically consulted under the “modified categorical approach” by a court seeking to identify the elements of an offense defined by an overbroad statute. E.g., Dulal-Whiteway v. U.S. Dep’t of Homeland Sec., No. 05-3098-ag, 2007 WL 2712941 (2d Cir. Sept. 19, 2007) (holding that a loss of more than $10,000 may be shown only by information appearing in a record of conviction that would be permissible under Taylor an Shepard); Kawashima v. Gonzales, Nos. 04-74313, 05-74408, 2007 WL 2702330 (9th Cir. Sept. 18, 2007) (holding that the modified categorical approach limits the scope of review to the record of conviction in relation to a loss exceeding $10,000 where there was no element of loss in the criminal statute); Conteh v. Gonzales, 461 F.3d 45 (1st Cir. 2006) (requiring clear and convincing evidence derived solely from the criminal record of the aggravated felony charge, even if the criminal statute lacked a “loss” element); Alaka v. Att’y Gen., supra, at 107-08 (stating, in relation to the conduct resulting in conviction, that “[b]ecause our consideration is limited to the offense of conviction, we look only to the charges to which the [alien] pled guilty, and not to conduct that was neither
admitted nor proven beyond a reasonable doubt”); see also Tokatly v. Ashcroft, 371 F.3d 613, 622 (9th Cir. 2004) (describing the categorical and modified categorical approaches as being based “on the nature of the alien’s conviction, rather than on the alien’s actual conduct,” but then requiring that conviction-related facts that are not elements of the underlying criminal statutes nonetheless be established by reference to the “record of conviction”).
Despite the rulings in cases such as Dulal-Whiteway, Conteh, an Tokatly, we do not believe there is any sound legal principle that constrains inquiry to the record of conviction if the search involves aspects of the crime that go beyond the elements of the offense. I Tokatly, the Ninth Circuit specifically rejected the Government’s argument that the ground of removal pertaining to crimes of domestic violence, section 237(a)(2)(E)(i) of the Act, required a bifurcated evidentiary approach. The Government acknowledged that proof of a “conviction” for a “crime of violence,” with all its attendant “elements,” was required. On the other hand, the Government maintained that the existence of a “domestic” relationship between the offender and the victim was not understood by Congress as describing an element of the predicate offense but was instead a separate “immigration fact” that should have been subject to independent proof-outside the “record of conviction” if necessary. See Tokatly v. Ashcroft, supra, at 620. The Ninth Circuit rejected this approach, describing it as a “somewhat convoluted and bipolar methodology to the statutory inquiry.”Id. The principal grounds for rejection, however, were the Ninth Circuit’s Taylor-related approach to assessing convictions and our adherence to a comparable approach in cases such as Matter of Pichardo, supra, and Matter of Teixeira, 21 IN Dec. 316 (BIA 1996).
We have a different perspective. We accept that a “categorical” approach is to be applied with respect to any aspect of a removal charge that depends on a conviction under a criminal statute containing elements that must be shown to establish removability. The situation is much different, however, if the removal charge requires proof of some fact that is not an element of the predicate offense. For example, a finding of removability under section 237(a)(2)(A)(i) of the Act requires proof of a conviction for a crime involving moral turpitude that was committed within 5 years after admission. In implementing this statute we do not look to the conviction record to determine the date when an alien was admitted to the United States under the immigration laws because the date of admission is not an element of any criminal offense,
except perhaps in the rarest of cases. Similarly, although the precise date an offense was committed may be relevant in a criminal prosecution to provide a defendant with adequate notice of charges or to rebut a statute of limitations or ex post facto defense, it is not an element of the crime or an aspect of convicted conduct that must necessarily be established in removal proceedings by reference to the conviction record.
As to crimes of moral turpitude committed within 5 years after an alien’s admission, the statute clearly combines the need for a qualifying “conviction” with other requirements bearing both on the crime and on the alien’s immigration history. It may be awkward from the perspective of the “categorical” or “modified categorical” approach for a particular removal statute to include both “offense-element” facts and other “nonelement” facts pertaining either to the conviction (e.g., length of sentence or the date the offense was committed) or the impact of the crime on the victim. But when deportability depends in part on proof of nonelement aspects of a crime, we do not understand how the inquiry into those facts can automatically be constrained by rules that have been developed solely to ascertain the elements of the crime. By definition, the inquiry must go beyond the elements in such cases. The question, then, is whether the conviction record is an adequate source for the information demanded by the removal statute.
E. “Record of Conviction” Evidence To Assess “Loss” to the Victim
A conviction record may serve as the best source of information for nonelement facts about a crime. But this will depend on the nature of the nonelement information. No wooden rule, denominated as a “categorical” or “modified categorical” approach derived fro Taylor and Shepard, or employed by Pichardo, can answer that question in advance.
The length of sentence is a nonelement fact for some aggravated felonies that can be readily ascertained from the record of conviction, and it would not make sense to invite other proof, absent a scrivener’s error. On the other hand, in Matter of Gertsenshteyn, supra, we found that independent proof as to the “committed for commercial advantage” component of section 101(a)(43)(K)(ii) aggravated felonies should be allowed in order to give life
to the statute, as evidence of this aspect of a crime would not likely be found in records of conviction.
The loss threshold of section 101(a)(43)(M)(i) falls between these extremes. Some criminal complaints may allege specific loss amounts, despite the absence of “loss” elements in the criminal statutes. Restitution is frequently included in criminal judgments, and the amount of loss may be discussed during a plea colloquy or included in the sentencing court’s SOR. Indeed, the respondent concedes that a conviction record can establish “loss to the victim,” but only if the alien admitted the amount of loss during the criminal proceedings. In this respect, he argues that restitution orders are based on calculations that do not necessarily match the “loss to the victim” focus of the removal statute, and that such orders, as well as a PSR (even if adopted by the court without change), do not establish the quantum of loss by “clear and convincing evidence” because they need only have been found by a “preponderance of the evidence.”
Accordingly, even if a criminal court judge’s finding as to loss is clear from the record, there may be grounds for questioning whether such a finding, standing alone, is sufficient to prove the amount of loss by the “clear and convincing evidence” standard that applies in removal proceedings. Obasohan v. U.S. Att’y Gen., 479 F.3d 785, 790-91 (11th Cir. 2007). However, restitution orders have served as evidence in some removal cases, even when challenged. See James v. Gonzales, 464 F.3d 505
(5th Cir. 2006) (rejecting specific challenges to use of the restitution order); Ferreira v. Ashcroft, 390 F.3d 1091 (9th Cir. 2004) (finding a California State court order of restitution reliable in part because California law, unlike Federal law, limited restitution to the actual loss suffered by the Government agency victim). The Third Circuit has stated that a restitution order “may be helpful” in determining loss to the victim if the plea agreement or indictment is unclear as to loss Munroe v. Ashcroft, 353 F.3d 225, 227 (3d Cir. 2003).
We conclude that restitution orders can be sufficient evidence of loss to the victim in certain cases, but they must be assessed with an eye to what losses are covered and to the burden of proof employed. We do not agree with the respondent that restitution orders and presentence reports can never suffice if a preponderance of the evidence standard was employed in entering the order or developing the report. Facts found by a preponderance may also meet more
stringent evidentiary tests. For example, a defendant’s failure to contest, during the criminal proceedings, a fact found by a preponderance would bear on whether that fact was reliable for removal purposes as well, especially in the absence of any showing in removal proceedings that there was error in the criminal proceedings respecting that fact.
Further, a defendant’s admission during the criminal proceeding as to the amount of loss will suffice to meet a clear and convincing showing in removal proceedings, if the admission pertained to losses arising from the conduct in the particular charges or criminal counts covered by the conviction. This most likely includes a plea to a criminal count alleging a fraudulent transaction in excess of $10,000, unless the convicting jurisdiction treats the plea as only an admission of the bar elements of the crime. E.g., Bobb v. Att’y Gen., 458 F.3d 213 (3d Cir. 2006) (finding that a guilty plea to passing a forged check in an amount of $13,277 confirmed that the amount of loss was greater than $10,000, despite no “loss” element in the criminal statute). In addition, conviction record evidence as to loss generally may be sufficient if the record in question arose in a jurisdiction that has adapted t Apprendi and its progeny by requiring sentencing factors to be proven to the jury beyond a reasonable doubt. However, even a plea to a fraudulent transaction exceeding $10,000, or a sentencing fact found beyond a reasonable doubt, may be suspect if the admission or sentencing factor covered losses associated with transactions outside the particular count or counts covered by the conviction. See Alaka v. Att’y Gen. supra (holding that losses arising from fraudulent activity outside the actual offense of conviction could not be included for section 101(a)(43)(M)(i) aggravated felony purposes).
F. Independently Assessing “Loss”
While reliance on the criminal “record of conviction” is still possible in many cases, the record of conviction is an uncertain source of reliable information on loss to the victim. The statutory $10,000 threshold is not an element of the crime except, perhaps, in rare cases. The information generated on loss is routinely done for sentencing purposes, not for “conviction” purposes, and may have been assessed against a “preponderance of the evidence” standard. Admissions by defendants should suffice in many cases, but even a plea colloquy need not actually focus on the loss stemming from a particular count in an indictment, if there is no loss element needed to prove
guilt, and may instead focus on a broader loss determination relating exclusively to sentencing or restitution. Because of these limitations on the probative value of “record of conviction” evidence relating to victim loss, we discern no sound reason for prohibiting Immigration Judges from considering other reliable evidence that bears on this question, including but not limited to the testimonial admissions of the respondent made during the removal hearing. Furthermore, because we are looking for a fact that was part of the crime, but not a fact that must have been proved to establish guilt, the independent assessment of that fact during a removal hearing does not encroach on the principal purpose of the criminal proceedings, which was the determination of guilt under the elements of the criminal statute. Hence, we see no legal objection to the independent resolution of this “victim impact” issue during the course of a removal hearing. Accordingly, we hold that an Immigration Judge may consider any evidence, otherwise admissible in removal proceedings, including witness testimony, bearing on the loss to the victim in an aggravated felony case involving section 101(a)(43)(M)(i) of the Act.
III. RESPONDENT’S CASE
In this case, the combination of the counts to which the respondent pled, including a broad conspiracy charge spanning a period from July 2002 to January 2003, the order of restitution, and the information in the PSR and SOR would carry the DHS’s burden to show a loss exceeding $10,000 by clear and convincing evidence, absent countervailing evidence. Both the conspiracy count and the individual count alleged a “scheme” that the defendants “did execute” as part of the conspiracy or, as relates to the individual count, “for the purpose of executing such scheme . . . caused a bill for unnecessary medical treatment to be mailed . . . to Liberty Mutual Insurance Company.” The charges included conduct well beyond a simple agreement to violate the law.
The Immigration Judge refused to consider the PSR, however, because he believed it to be beyond the scope of permissible evidence. We acknowledge that a PSR may have shortcomings. There can be questions as to the accuracy of the report, and the information in it may have been assessed against a “preponderance of the evidence” standard. Thus, the respondent should have an opportunity to object and to offer evidence of inaccuracies that may not have had a bearing on the criminal case, but that do have a bearing on the removal case. Importantly, neither party approached the hearing below with the understanding that testimonial evidence could be offered directly to the Immigration Judge. While the respondent did not offer evidence to refute the DHS’s exhibits, we cannot discount the possibility that he possesses such evidence. In any event, we are not permitted to make findings of fact for the first time on appeal. Accordingly, a remand is required.
Our ruling today represents a departure from the precepts that have been presumed to apply in immigration hearings involving aggravated felony charges arising under section 101(a)(43)(M)(i) of the Act. We leave for another day any questions that may arise with respect to circuit law that may be in tension with this decision, as we ordinarily follow circuit law in cases arising within the particular circuit and the grounds for any departure would need to be developed in the context of specific cases. See Nat’l Cable Telecomms. Ass’n v. Brand X Internet Services, 545 U.S. 967 (2005).
Nevertheless, we conclude that the statute governing removal for an aggravated felony conviction involving fraud or deceit with a loss exceeding $10,000 demands two types of determinations. The first is a categorical inquiry into a conviction under a criminal statute with an element of fraud or deceit. The second is an ordinary evidentiary inquiry into whether the loss associated with the fraudulent conduct encompassed by the conviction exceeds $10,000. The second inquiry cannot be confined to the categorical or modified categorical approach because it does not involve a search for the elements of the crime, even though conviction record information may suffice in making this independent “loss” determination.
In this case, the Immigration Judge erred in declining to consult evidence outside the record of conviction in order to determine the amount of loss resulting from the respondent’s criminal conduct. Accordingly, we find it necessary to remand the case to the Immigration Judge so that both parties may have the opportunity to submit evidence bearing on the question of removability.
ORDER: The appeal of the Department of Homeland Security is sustained.
FURTHER ORDER: The record is remanded to the Immigration Judge for further proceedings consistent with the foregoing opinion and for the entry of a new decision.
(1970) (emphasis added); see also Mullaney v. Wilbur, 421 U.S. 684, 697-99 (1975); Black’s Law Dictionary 520 (6th ed. 1990) (defining “[e]lements of crime” as “[t]hose constituent parts of a crime which must be proved by the prosecution to sustain a conviction”). In sentencing cases, the Supreme Court has concluded that any fact, other than a prior conviction, that is necessary to support a punishment must be admitted by the defendant or proven to a jury beyond a reasonable doubt. Cunningham v. California, 127 S. Ct. 856 (2007); United States v. Booker, 543 U.S. 220 (2005); Blakely v. Washington, 542 U.S. 296 (2004) Apprendi v. New Jersey, 530 U.S. 466 (2000). We are not now treating sentencing facts that are outside the statutory definition of the crime as “elements,” although we acknowledge that there may be reasons for deeming sentencing factors found in accordance with Apprendi to be part of the elements leading to punishment.