Attorney General of U.S.

Not overruled or negatively treated on appealinfoCoverage
United States Court of Appeals, Third CircuitApr 9, 2010
602 F.3d 260 (3d Cir. 2010)

Cases citing this case

How cited

  • Turkson v. Holder

    …16(c)(2). The regulations direct the trier of fact to consider “all evidence relevant to the possibility of…

  • Johnson v. Att'y Gen. U.S.

    …As we recently clarified, what is likely to happen to the petitioner if removed is a factual question which…

lock 138 Citing caseskeyboard_arrow_right

Summaries written by judges


  • holding that the agency's conclusion about "what is likely to happen to the petitioner if removed" is an issue of fact

    Summary of this case from Murguia-Briseno v. Attorney Gen. U.S.

  • finding $10,000 loss requirement established based on allegations in the criminal information to which the noncitizen pled guilty and in the presentence report

    Summary of this case from De Morais v. United States

  • rejecting an argument that In re Frentescu should apply to a crime committed in 1998 where the statutory definition of a particularly serious crime has since changed

    Summary of this case from Infante v. Attorney Gen. of the United States

No. 08-2571.

Argued March 8, 2010.

Opinion filed: April 9, 2010.

Thomas E. Moseley, Esquire (Argued), Newark, NJ, for Petitioner.

Thomas W. Hussey, Esquire, Paul F. Stone, Esquire, United States Department of Justice, Office of Immigration Litigation, Gregory G. Katsas, Assistant Attorney General, Civil Division, Terri J. Scadron, Assistant Director, Manuel A. Palau, Esquire (Argued), United States Department of Justice, Office of Immigration Litigation, Washington, DC, for Respondent.

Before: AMBRO, SMITH, and MICHEL, Circuit Judges.

Honorable Paul R. Michel, Chief Judge, United States Court of Appeals for the Federal Circuit, sitting by designation.


AMBRO, Circuit Judge.

Petitioner Vadim Kaplun, a citizen of the Ukraine, petitions for review of four decisions of the Board of Immigration Appeals resulting in a final order of removal that designated him removable for having committed an aggravated felony, denied him withholding of removal by virtue of having committed a particularly serious crime, and reviewed de novo and reversed the Immigration Judge's finding of a clear probability of future torture if Kaplun were removed. He contends that the Government did not prove he committed an aggravated felony by clear and convincing evidence; as a nonviolent, white collar offense, his offense was not a particularly serious crime; and the BIA applied an improper de novo standard of review to the IJ's finding that Kaplun's torture if removed would be probable. We conclude that the BIA was correct on the first two issues, but applied an incorrect standard of review on the third. Accordingly, we deny the petition for review on the first two claims, and grant it on the third claim.

1. Facts and Procedural History

A. Kaplun's background

Kaplun is a native of the Ukraine who was admitted to the United States in 1977 as a seven-year-old refugee. He later became a legal permanent resident. In 1997 and 1998 he was charged and convicted in two federal criminal proceedings based on his participation in fraudulent stock schemes.

For reasons we describe below, only the 1998 conviction is at issue here. Kaplun there pled guilty to an information. alleging securities fraud with losses of nearly $900,000 under 15 U.S.C. §§ 77q, 77x, and 18 U.S.C. § 2 Per the presentence investigation report (PSR), the total loss for the 1998 offense was described as "at least $700,000 and less than $1,000,000." The $700,000 figure was used twice more in the PSR to calculate the specific offense level and to calculate the maximum fine. Defense counsel made no objection to the PSR. After the District Court adopted the PSR (save for two exceptions not relevant here) and granted a downward departure, the undisputed Guideline range was 51-63 months' imprisonment. Kaplun was sentenced to 56 months' imprisonment for the 1998 conviction, but a fine was waived because of his inability to pay.

Kaplun waived indictment and consented to proceeding by information.

From 1987 to 2001, the United States Sentencing Commission Guidelines Manual had specific provisions for fraud offenses in § 2F1.1. Subsection 2F1.1(b)(1)(K) assigned a 10-level increase for fraud offenses with a loss of more than $500,000. This was used in Kaplun's PSR. U.S.S.G. § 2F1.1 has since been repealed.

B. Removal proceedings before the Immigration Judge

The Government began removal proceedings against Kaplun in 2001 based on the 1997 and 1998 convictions. He denied removability and later submitted an application for asylum. The Government produced the judgment of conviction, the PSR, and the information to establish the 1998 conviction and its surrounding facts. No plea colloquy was produced, though Kaplun does not deny that he pled guilty to the single-count information.

In his application for asylum, Kaplun claimed that, as a Jewish refugee, he would be subjected to persecution and torture if he were removed to the Ukraine. In support of his claims, he procured an expert witness to give testimony on antiSemitism in the Ukraine. This expert gave detailed testimony on the situation and voiced disagreement with various Government reports on the extent of antiSemitism in that country. He also testified that Kaplun would be unable to gain citizenship, get a job, rent an apartment, or even buy a train ticket. It was his expert opinion that Kaplun would be living on the street, destitute, and would be targeted for extortion and torture.

In an April 2004 ruling, the IJ found Kaplun removable based on his prior convictions (though it was unclear which of the two convictions qualified), but granted withholding of removal and protection under the United Nations Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment ("CAT"), crediting the testimony of Kaplun's expert. Despite this, Kaplun appealed the part of the ruling that found him removable. The Government cross-appealed the part of the ruling granting withholding of removal.

Technically a CAT claim is a claim under the Foreign Affairs Reform and Restructuring Act, but we use the convention adopted in our Circuit and refer to it as a CAT claim. See Pierre v. Att'y Gen., 528 F.3d 180, 186 n. 5 (3d Cir. 2008) ( en banc).

C. The BIA's first ruling

The first ruling by the Board of Immigration Appeals was issued in November 2004. It held that the IJ erred in his removal findings by inadvertently relying on the wrong record of conviction, and it vacated the IJ decision and remanded for a determination of whether Kaplun was indeed removable as charged under either the 1997 or the 1998 conviction.

D. The Immigration Judge responds

The IJ issued a second ruling in February 2006, in which he went through the removability charges in great detail. He concluded that the sole sustainable removal charge was the 1998 fraud conviction. The basis for this decision was Kaplun's conviction for an aggravated felony as defined in 8 U.S.C. § 1101(a)(43)(M)(i) ("an offense that . . . involves fraud or deceit in which the loss to the victim or victims exceeds $10,000"). The IJ then certified the case and sent the matter back to the BIA for resolution of the remaining issues on appeal — namely, removability and the prior grant of relief from removal under the CAT.

E. The BIA's second ruling

The BIA issued a second decision in August 2006. It addressed the following: (1) whether the removability charge was sustainable based on the 1998 conviction; (2) whether that 1998 conviction was a particularly serious crime rendering Kaplun statutorily ineligible for withholding of removal, or in the alternative whether Kaplun met his burden of proof of establishing eligibility for withholding of removal; and (3) whether the IJ erred in finding that Kaplun had met his burden of proving probability of torture under the CAT. The BIA concluded that Kaplun was removable on the basis of the 1998 conviction, that it was a "particularly serious crime" (thereby declining to address the alternative argument), and that the IJ erred in granting relief under the CAT. It thus ordered Kaplun removed to the Ukraine.

He filed a petition for review of the final order of removal in August 2006. We did not rule on the petition at that time.

F. The BIA issues a third ruling denying a motion to reopen

In November 2006, Kaplun filed a motion with the BIA to reopen his case based on our decision in Alaka v. Attorney General, 456 F.3d 88, 108 (3d Cir. 2006) (in determining a loss amount for 8 U.S.C. § 1101(a)(43)(M)(i), "we look only to the charges to which the petitioner pled guilty, and not to conduct that was neither admitted nor proven beyond a reasonable doubt"). In February 2007, the BIA denied the motion to reopen because it concluded that Alaka was factually distinguishable.

Kaplun then filed a petition for review with our Court in February 2007. On motion of the Government, we remanded Kaplun's 2007 petition, along with his 2006 petition, to the BIA in September 2007. We asked for a determination of whether the BIA had authority

to reverse the immigration judge's determination that "there [was] a preponderance of evidence in the record leading to a justification for a clear probability finding that this particular respondent . . . is likely to be targeted [for mistreatment,] at least in part, by both governmental and nongovernmental entities within the Ukraine should he be removed to that country . . . [and that such mistreatment will rise to the level of torture.]"
G. The BIA's fourth ruling

In May 2008, the BIA issued a precedential decision answering the question on remand. Matter of V-K-, 24 I N Dec. 500 (BIA 2008). It held that it had the authority to review the IJ's determination de novo, and it reasoned as follows:

We now clarify that while we reviewed the Immigration Judge's factual rulings for clear error, we do not consider a prediction of the probability of future torture to be a ruling of "fact." Although predictions of future events may in part be derived from "facts," they are not the sort of "[f]acts determined by the Immigration Judge" that can only be reviewed for clear error.

. . . .

Accordingly, we conclude that an Immigration Judge's prediction or finding regarding the likelihood that an alien will be tortured may be reviewed de novo because, like a conclusion relating to whether a statutorily prescribed chance of persecution or level of hardship exists, it relates to whether the ultimate statutory requirement for establishing eligibility for relief was met and is therefore a mixed question of fact and law, or a question of "judgment."

24 I N Dec. at 501-02 (alteration in original) (citations omitted).

Kaplun then filed the petition for review that is before us today, as to which we have jurisdiction pursuant to 8 U.S.C. § 1252(a)(1).

II. Standard of Review

When the BIA issues its own decision on the merits and not a summary affirmance, we review its decision, not that of the IJ. Sheriff v. Att'y Gen., 587 F.3d 584, 588 (3d Cir. 2009). As Kaplun is subject to a final order of removal by reason of having committed an aggravated felony, we review his petition only to the extent it raises questions of law or constitutional claims, and purely factual or discretionary determinations are outside our scope of review. See Pierre v. Att'y Gen., 528 F.3d 180, 184 (3d Cir. 2008) ( en banc) (citing 8 U.S.C. § 1252(a)(2)(C)-(D)). We review the BIA's legal determinations de novo, subject to the principles of deference articulated in Chevron v. Natural Resources Defense Council, Inc., 467 U.S. 837, 844, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). Pierre, 528 F.3d at 184; see also Filja v. Gonzales, 447 F.3d 241, 251 (3d Cir. 2006). The Board's interpretation of its own regulation is "controlling unless `plainly erroneous or inconsistent with the regulation.'" Auer v. Robbins, 519 U.S. 452, 461, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997) (citations omitted).

III. Discussion

A. Kaplun's 1998 conviction was shown to be an aggravated felony by clear and convincing evidence.

Kaplun's first claim stems from his 1998 securities fraud conviction. He argues that the sole charge of removability should not have been sustained because the Government did not prove the $10,000 loss required to qualify as an "aggravated felony" with clear and convincing evidence. We disagree.

Federal immigration law provides that any "alien who is convicted of an aggravated felony at any time after admission is deportable." 8 U.S.C. § 1227(a)(2)(A)(iii). The list of offenses that qualify as an "aggravated felony" is given by 8 U.S.C. § 1101(a)(43). As noted, § 1101(a)(43)(M)(i) makes "an offense that . . . involves fraud or deceit in which the loss to the victim or victims exceeds $10,000" an aggravated felony.

The Supreme Court has recently clarified the approach we are to take when determining what evidence can be consulted to determine whether a crime is an "aggravated felony" for purposes of 8 U.S.C. § 1101(a)(43)(M)(i). Nijhawan v. Holder, 557 U.S. ___, 129 S.Ct. 2294, 174 L.Ed.2d 22 (2009). The Court first determined that this provision "calls for a `circumstance-specific,' not a `categorical,' interpretation." Id. at 2300. With regard to the $10,000 figure, "the monetary threshold applies to the specific circumstances surrounding an offender's commission of a fraud and deceit crime on a specific occasion." Id. at 2302.

The Court also rejected a "modified categorical approach," which would limit the evidence to "a jury verdict, or a judge-approved equivalent . . . [such as] charging documents, jury instructions, and any special jury finding . . . [, or] written plea documents or the plea colloquy." Id. Noting that the Government was held to a "clear and convincing" burden of proof (and not "beyond a reasonable doubt"), it found "nothing unfair about [an] rel[ying] upon earlier sentencing-related material" (there, a stipulation for sentencing purposes and a restitution order). Id. It also noted that "the sole purpose of the aggravated felony inquiry is to ascertain the nature of a prior conviction; it is not an invitation to relitigate the conviction itself." Id. (internal quotation marks omitted).

In our case, the record shows that the IJ was presented with clear and convincing evidence that Kaplun was convicted of a fraud crime with a loss exceeding $10,000. Two powerful and uncontroverted items of evidence establish the amount clearly and convincingly. The first is the criminal information to which Kaplun pled guilty, in which he is alleged to have converted nearly $900,000 of investors' funds to his personal use and that of his accomplices. Although no plea colloquy was produced by the Government or Kaplun, he does not deny pleading guilty to the information alleging the criminal conduct involving fraud or deceit.

The PSR is the second piece of evidence establishing a loss greater than $10,000. At three different points in the PSR, the loss is alleged to be at least $700,000. First, in the "Offense Conduct" section, the loss resulting from Kaplun's conduct in the 1998 conviction is described as follows:

As an overview, between late April 1995 and October 1995, VADIM KAPLUN was involved in a securities fraud scheme which resulted in a total loss to over two hundred investors of at least $700,000 and less than $1,000,000.

Second, in the "Offense Level Computation" section, the loss is again determined to be greater than $10,000:

Specific Offense Characteristic: This fraud resulted in a loss of at least $700,000. Pursuant to Guideline 2F1.1(b)(1)(K), ten levels are added.

U.S.S.G. § 2F1.1(b)(1) provided a schedule of offense level enhancements that vary depending on the amount of loss resulting from the offense, giving the defendant every incentive to ensure that it is calculated correctly so as to avoid excess punishment. Third, in the "Fines" section, the loss is once again determined to be greater than $10,000:

For example, subsection (K) provides for ten levels to be added for a loss greater than $500,000. Subsection (J) provides for nine levels to be added if the loss is greater than $350,000 while subsection (L) provides for eleven levels to be added if the loss is greater than $800,000.

Statutory Provisions: . . . For Count One of 98-CR-1387 (JGK), pursuant to 18 USC 3571(d), the maximum fine is $1,400,000, namely twice the loss of at least $700,000 caused by Kaplun's participation in that fraud scheme.

Kaplun made no objection to the PSR, and save for two exceptions not related to the loss amount, the District Court adopted the PSR in its entirety, including its findings of fact.

Like the Supreme Court in Nijhawan, "[w]e can find nothing unfair about the immigration judge's having here relied upon earlier sentencing-related material." 129 S.Ct. at 2303. While the Government could have produced a plea colloquy transcript demonstrating Kaplun's acceptance of the loss amount (and Kaplun suggests that we should demand such evidence), in the face of his guilty plea to the information alleging over $10,000 in losses, the three uses of a loss amount over $10,000 in the PSR that was adopted by the District Court, the absence of any objections to the PSR by Kaplun, and "the absence of any conflicting evidence (and [Kaplun] mentions none), this evidence is clear and convincing." Id.

B. The BIA did not commit legal error when it determined that Kaplun's aggravated felony was a "particularly serious crime"

The BIA found Kaplun ineligible for withholding of removal because the 1998 conviction was for a "particularly serious crime." Removal is withheld if "the alien's life or freedom would be threatened in that country because of the alien's race, religion, nationality, membership in a particular social group, or political opinion." 8 U.S.C. § 1231(b)(3)(A). However, this does not apply if "the alien, having been convicted by a final judgment of a particularly serious crime[,] is a danger to the community of the United States." Id. § 1231(b)(3)(B)(ii).

"Particularly serious crime" includes an "aggravated felony (or felonies) for which the alien has been sentenced to an aggregate term of imprisonment of at least 5 years." Id. § 1231(b)(3)(B). It also provides for a discretionary designation of "particularly serious crime" when the Attorney General determines that, no matter how short the sentence, the crime qualifies as particularly serious. Id. Kaplun was not sentenced to an aggregate term of at least five years, so we look only to the discretionary portion of the statute. We do not have jurisdiction over discretionary decisions, and without any assertion of legal error, Kaplun is not entitled to relief on this claim.

Kaplun makes three legal arguments why he has not been convicted of a "particularly serious crime": (1) the Government has waived the issue, having not brought it before the IJ; (2) the "entire analysis . . . rests upon the legally erroneous conclusion that [his] guilty plea admitted all the allegations in the information"; and (3) the BIA decision "ran contrary to actual Board precedent." We reject all three arguments.

First, Kaplun cites no relevant authority for the proposition that the Government waives the right to challenge withholding of removal if it does not bring it before the IJ. The sole authority cited, Tokatly v. Ashcroft, 371 F.3d 613, 618 (9th Cir. 2004), does not support that proposition. There, the Government "waived waiver" by failing to argue to the BIA that a petitioner had waived an argument, raising the issue for the first time before the Court of Appeals. This situation is distinguishable for two reasons: (1) Kaplun himself did not raise this issue before the BIA; and (2) the Government raised the issue prior to briefing in the federal courts, having brought it to the BIA's attention. We thus decline to hold that the Government waived the argument here.

Second, we reject Kaplun's assertion that the entire analysis rested on a legally erroneous conclusion for the reasons discussed in Part III.A. The Government has established the commission of an aggravated felony by clear and convincing evidence. Nor is there a problem with "tethering," for the 1998 conviction and loss figure was a result of a single-count information, and there is no dispute that the entire loss amount alleged was tied to Kaplun's guilty plea and conviction. See Alaka, 456 F.3d at 108 (holding that "the loss amount relevant to [an alien's] aggravated felony determination is . . . the loss suffered by the victim of the count to which [the alien] pled guilty"). This assertion also fails.

Finally, we reject Kaplun's contention that the BIA decision runs contrary to BIA precedent. Citing to a now-superseded BIA decision, Matter of Frentescu, 18 I N Dec. 244 (BIA 1982), he argues that the BIA "has never held that . . . a non-violent white collar criminal offense could constitute a particularly serious crime." Kaplun does not point out that Frentescu's definition of "particularly serious crime" has been replaced by the definition in 8 U.S.C. § 1231(b)(3)(B). It includes aggravated felonies with a term of imprisonment of at least five years, and there are a number of "non-violent white collar criminal offense[s]" that are aggravated felonies punishable by at least five years' imprisonment. Moreover, nothing in our precedent suggests that a financial crime cannot, as a matter of law, be a particularly serious crime.

The list of aggravated felonies in 8 U.S.C. § 1101(a)(43) is comprehensive and goes beyond "crimes of violence" (which are covered by 8 U.S.C. § 1101(a)(43)(F)). See, e.g., 8 U.S.C. § 1101(a)(43)(D) (money laundering), (P) (document fraud), (Q) (failure to appear for service of sentence), (R) (bribery, counterfeiting, forgery, or trafficking in vehicles with altered Vehicle Identification Numbers), (S) (obstruction of justice, perjury, and bribery of a witness), (T) (failure to appear pursuant to a court order to answer to or dispose of a felony).

C. The BIA applied an incorrect standard of review to the IJ's determination that there was a probability of future torture.

Kaplun's final claim alleges BIA error in reviewing the probability of future torture de novo and not under a "clearly erroneous" standard. Though the question is not as straightforward as it first appears, we agree with Kaplun.

Prior to 2002, the BIA reviewed all aspects of an IJ's decision de novo. In August 2002, procedural reforms were adopted after a period of notice and comment, and codified in the Code of Federal Regulations. The new rule governing BIA review now reads:

(d) Powers of the Board —

. . . .

(3) Scope of review, (i) The Board will not engage in de novo review of findings of fact determined by an immigration judge. Facts determined by the immigration judge, including findings as to the credibility of testimony, shall be reviewed only to determine whether the findings of the immigration judge are clearly erroneous.

(ii) The Board may review questions of law, discretion, and judgment and all other issues in appeals from decisions of immigration judges de novo.

8 C.F.R. § 1003.1(d)(3)(i)-(ii). If the probability of torture is a "finding of fact," the BIA may only review it for clear error, but if it is a "question of law, discretion, [or] judgment," the BIA reviews the IJ's decision de novo.

Eligibility for withholding of removal under the CAT is governed by 8 C.F.R. § 208.16(c). To obtain withholding on this ground, "[t]he burden of proof is on the applicant . . . to establish that it is more likely than not that he or she would be tortured if removed to the proposed country of removal." 8 C.F.R. § 208.16(c)(2). Here, the BIA concluded that a finding of a "probability of future torture" was "a mixed question of fact and law, or a question of `judgment,'" subject to de novo review. 24 I N Dec. at 502. It reasoned that "[a]lthough predictions of future events may in part be derived from `facts,' they are not the sort of `[f]acts determined by the Immigration Judge' that can only be reviewed for clear error." Id. at 501 (alteration in original) (citation omitted). As the Government notes, the BIA spoke even more forcefully in Matter of A-S-B-, 24 I N Dec. 493, 498 (BIA 2008), making the blanket assertion that "it is impossible to declare as `fact' things that have not yet occurred."

We reject the BIA's assertion that a "prediction of future events" is alternatively a "question of judgment" subject to de novo review. In this context, "judgment" means either the application of legal standards or the exercise of discretion, and not a situation where the decision-maker must determine the likelihood of a future event. Only the legal aspects of judgment or exercises of discretion, not the findings of fact, receive de novo review by the BIA. In either case, the BIA must take the facts as given (unless they are clearly erroneous), and only then exercise its judgment or discretion whether those facts meet the particular legal standard under review ( e.g., whether the facts are sufficient to be deemed torture).
This is borne out in the Attorney General's comments on the final rule. The Attorney General interpreted the rule to allow for de novo review of "judgments as to whether facts established by a particular alien amount to [a legal standard]." Board of Immigration Appeals: Procedural Reforms To Improve Case Management, 67 Fed.Reg. 54,878, 54,890 (Aug. 26, 2002). On the topic of discretion to grant relief, the rule was interpreted to mean that "the `discretion,' or judgment, exercised based on . . . findings of fact, and the weight accorded to individual factors, may be reviewed by the Board de novo." Id.

In an opinion published after we heard argument in this case, the BIA continues to take the stance adopted in Matter of V-K- and Matter of A-S-B- that likelihoods cannot be "facts" because they involve future events. See Matter of H-I-H Z-Y-Z-, 25 I N Dec. 209, 212-13 n. 4 (BIA 2010). To the extent that A-S-B- and H-h-H Z-Y-Z- address the standard of review applied to an IJ's determination of a "well-founded fear of future persecution" in the asylum context (and not a likelihood of torture), we do not purport to resolve that issue at this time.

For the reasons that follow, insofar as the BIA interpreted 8 C.F.R. § 1003.1(d)(3) to hold that an IJ's assessment of the probability of future torture is not a finding of fact because the events have not yet occurred, we conclude its interpretation plainly errs. See Auer, 519 U.S. at 461, 117 S.Ct. 905. Accordingly, it is not controlling, and to the extent that the BIA relied on this incorrect statement to justify its interpretation of its own regulation governing the standard of review, we reject its conclusion.

Facts include past events, but they are not restricted to historical events. See Black's Law Dictionary 669 (9th ed. 2009) (defining "fact" to mean "something that actually exists; an aspect of reality," or "[a]n actual or alleged event or circumstance, as distinguished from its legal effect, consequence, or interpretation"; defining "[f]acts" to "include not just tangible things, actual occurrences, and relationships, but also states of mind such as intentions and opinions"); Webster's Third New Int'l Dictionary 813 (1971) (defining "fact" as "the quality or character of being actual or of being made up of facts"; giving as an example "a question of fact hinges on the actual evidence"). A present probability of a future event is something "distin[ct] from its legal effect" that is "made up of facts" and "actually exists" but is "not [a] tangible thing[], [or] actual occurrence[]." This likelihood, while an assessment of a future event, is what a decision-maker in an adjudicatory system decides now as part of a factual framework for determining legal effect.

For example, that "it took 2 hours to drive the 100 miles to grandmother's house last week" may be established and found as a fact through documentary, testimonial, or other evidence. A finding of fact may also stem from an assessment of what is expected to occur in the future. Take, for instance, "It is likely that it will take less than 3 hours to drive the 100 miles to grandmother's house next week." The likelihood (an inferential fact) may be established through, for example, testimony of past experience, testimony on typical traffic conditions, or even expert opinion testimony on traffic patterns. Of course, to call a likelihood "fact" is not to say that the likely outcome will necessarily occur, but the likelihood itself remains a factual finding that can be made ex ante the actual outcome.

Taking another example from the medical malpractice context, experts frequently testify to their opinions (or "predictions") of future disability and future pain and suffering. When a jury chooses to believe the expert's predictions, it makes a factual finding that the plaintiff will be unable to perform certain tasks in the future and will experience some degree of pain and suffering. If a patient plaintiff improves, or dies unexpectedly, we do not say that a jury made a factual error, nor do we conclude — indeed, it would be too late to do so — that the expert should not have been found worthy of belief. The patient plaintiffs future condition is a factual finding made by the jury.

Elsewhere outside of the immigration context, we have held that factual inferences drawn from historical facts (including the likelihood of future events) are reviewed for clear error. See United States v. Stewart, 452 F.3d 266, 273 (3d Cir. 2006) (holding that whether the release of an individual, found not guilty solely by reason of insanity, creates a substantial risk of future danger to society is a finding of fact reviewed for clear error); Martin v. Cooper Elec. Supply Co., 940 F.2d 896, 900 (3d Cir. 1991) ("[W]here the district court was required to draw factual inferences from historical facts . . . [,] we review those inferences for clear error too."); see also Onisheav. Hopper, 171 F.3d 1289, 1300-01 (11th Cir. 1999) ( en banc) (holding that a district court's finding as to the risk of future prison violence based on conflicting evidence was a factual determination reviewed for clear error).

Applied to immigration generally (and specifically this case), a finding of fact by an IJ includes expressions of likelihood based on testimony (both lay and expert) and/or documentary evidence. For example, if an IJ concludes, based on expert opinion, that "Kaplun is likely to be imprisoned if removed to the Ukraine," that finding would be factual and reviewable only for clear error, even though it is possible that Kaplun may be removed and yet never imprisoned. If Kaplun were removed but turned out not to be imprisoned (the likely outcome did not materialize), it would not change the inquiry into the likelihood of imprisonment from a question of fact to a mixed question of law and fact or a question of judgment.

The confusion arises when the question is posed as whether it is more likely than not that Kaplun would be tortured. Torture is a term with legal significance. The regulations outline the evidence to be considered in assessing the possibility of future torture:

In assessing whether it is more likely than not that an applicant would be tortured in the proposed country of removal, all evidence relevant to the possibility of future torture shall be considered, including, but not limited to:

(i) Evidence of past torture inflicted upon the applicant;

(ii) Evidence that the applicant could relocate to a part of the country of removal where he or she is not likely to be tortured;

(iii) Evidence of gross, flagrant or mass violations of human rights within the country of removal, where applicable; and

(iv) Other relevant information regarding conditions in the country of removal.

8 C.F.R. § 208.16(c)(3).

Here, the BIA noted that there were three likelihoods involved: "In his decision, the Immigration Judge credited the testimony of [Kaplun]'s expert witness that [Kaplun] `[1] is likely to need to come into contact with governmental entities and [2] is likely to be a target for extortion and mistreatment that [3] is likely to rise to the level of torture.'" 24 I N Dec. at 501; see also App. 26-27 (IJ Oral Dec). The BIA treated this trio of likelihoods as a single question, namely, whether there was a clear probability that, if removed, Kaplun would be tortured in the future.

In the case of the likelihood of torture, there are two distinct parts to the mixed question: (1) what is likely to happen to the petitioner if removed; and (2) does what is likely to happen amount to the legal definition of torture? The two parts should be examined separately.

The first question is factual. A finding that a petitioner is likely to be imprisoned (based, for instance, on the evidence of gross violations of human rights in the country of removal) is a finding of fact. A finding that a petitioner is likely to be extorted for vast sums of money if removed (based, for example, on the testimony of the petitioner and his expert) is also a finding of fact. So is a finding that a petitioner is likely to be beaten by government officials based on a finding that the petitioner was severely beaten in the past. This is to be distinguished from the legal consequence of those underlying facts.

The second question, however, is a legal question. Torture is a term of art, and whether imprisonment, beating, and extortion are severe enough to rise to the level of torture is a legal question. While the underlying facts vary from petitioner to petitioner, the legal question remains the same: do the facts found by the IJ (and that the BIA determines are not clearly erroneous) meet the legal requirements for relief under the CAT? This is a question of law where the IJ has no comparative advantage over the BIA.

Glueing the two questions together, however, does not entitle the BIA to review the first question, the factual one, de novo. It must break down the inquiry into its parts and apply the correct standard of review to the respective components. This is consistent with the Attorney General's guidance promulgated with the 2002 procedural reforms. See Board of Immigration Appeals: Procedural Reforms To Improve Case Management, 67 Fed.Reg. 54,878, 54,890 (Aug. 26, 2002). There, the Attorney General gave two examples that were consistent with this dichotomy between fact and law. The second example, eligibility for cancellation of removal, is relevant to our inquiry:

Similarly, in cancellation of removal, those facts that [the alien] claims make up "exceptional and extremely unusual hardship" to [the alien]'s putative qualifying relative . . ., and whether the putative qualifying relative is actually a qualifying relative, will be reviewed by the Board only to determine if the immigration judge's determination was clearly erroneous. Whether those facts, as determined by the immigration judge and found not to be clearly erroneous, amount to "exceptional and extremely unusual hardship" . . . may be reviewed by the Board de novo.

Id. (citations omitted). While looking at the hardship necessarily involves ascertaining the future factual consequences that would result from removal of the alien, it is the degree of hardship that constitutes a legal question, namely, whether it is "exceptional and extremely unusual." Id. (citing Matter of MonrealAguinaga, 23 I N Dec. 56 (BIA 2001)).

Returning to the trio of likelihoods in Kaplun's case, the BIA should have addressed the three questions as follows: (1) the likelihood of Kaplun interacting with governmental authorities on return to the Ukraine is a factual question; (2) the likelihood of his being targeted for extortion and mistreatment is also a factual question; and (3) whether the likely mistreatment will amount to torture is a legal question. Here — even though the BIA purported "not [to] find any facts [itself,] but only assessed the facts as found by the Immigration Judge and established by the evidence entered into the record, determining that they were insufficient to meet [Kaplun]'s burden of proof for protection under the Convention Against Torture," 24 I N Dec. at 502 — it appears that the BIA reexamined the record and conducted de novo fact-finding instead of applying the clearly erroneous standard.

On this record, we can assume only that the BIA applied a de novo standard to all of the IJ's relevant factual findings. The BIA justified its reversal of the IJ by declaring it "possess[ed] the authority to review de novo findings deemed by an Immigration Judge to satisfy an ultimate statutory standard," 24 I N Dec. at 502, and at no point did it purport to apply the clearly erroneous standard to the underlying facts. Perhaps a more precise (and correct) statement would be that the BIA possesses the authority to review de novo whether an Immigration Judge's factual findings (that are not clearly erroneous) satisfy an ultimate statutory standard (in this case, torture).

The IJ took testimony from an expert witness (his qualifications and competency to testify are not challenged) whose expert opinion was that Kaplun was likely to be detained and imprisoned, and that the conditions of imprisonment would constitute torture. The IJ credited the expert's testimony and concluded that Kaplun was likely to interact with governmental entities. The IJ then found that there was a "extremely high" likelihood this would make Kaplun a target for detention and mistreatment with the acquiescence or consent of a public official or other person acting in an official capacity for the current government of the Ukraine. Based on these factual findings, the IJ made the legal conclusion that Kaplun's likely detention would constitute torture, distinguishing it from situations where "it is somewhat problematic as to . . . whether the level of [mistreatment] is likely to rise to the level of the regulatory language . . . as to what constitutes torture."

The BIA applied de novo review to the question of whether there was a probability of torture, including the factual findings of the IJ. Specifically, the BIA disagreed with the IJ crediting "the [expert] witness's statements regarding what would happen in [Kaplun's] specific situation" if he were removed, dismissing it as "speculative." This appears to have been reversal of a factual finding under a de novo standard, an impermissible BIA action. The BIA also disagreed with the IJ's factual finding that "public officials would consent or acquiesce" to mistreatment of Kaplun, again appearing to make a de novo reversal. It should have reviewed these factual findings under a "clearly erroneous" standard, and the apparent application of the incorrect standard of review was error. In contrast, the BIA correctly applied a de novo standard of review in determining whether the claimed discrimination or mistreatment would constitute torture under the legal framework.

Although terming an expert's testimony "speculative" may indicate that the BIA considered the factual findings to be clearly erroneous, it did not say so. If it wishes to reverse factual findings it believes are not "established by evidence entered into the record," 24 I N Dec. at 502, it needs to apply the clearly erroneous standard in such a way that reviewing courts understand that standard to be in play.

We caution that we are not precluding BIA review of facts found by the IJ. It is free to revisit the underlying facts (including the likelihood of future events), but it must do so under the clearly erroneous standard that applies to facts. This review is important to the process of assuring that IJ-found facts get proper deference.

IV. Conclusion

We hold that when the Government provides unrebutted evidence of monetary loss in excess of $10,000 through an information, a guilty plea to that information, a presentence report, and the documented lack of objection to that report, it has proven the amount of loss by clear and convincing evidence for the purpose of establishing an aggravated felony under 8 U.S.C. § 1103(a)(43)(M)(i). We further hold that the BIA committed no legal error in designating Kaplun's aggravated felony to be a "particularly serious crime" under 8 U.S.C. § 1231(b)(3)(B)(ii), rendering him ineligible for withholding of removal. Finally, we hold that the BIA erred in reviewing the finding of a probability of torture de novo; it was required to review the factual aspects of that inquiry for clear error, and it was entitled to review only the legal aspects of the inquiry de novo. Accordingly, we grant the petition for review and remand for further proceedings consistent with this opinion.