Sunday, October 30, 2011

2B, Or Not 2B?

United States v. Bahel, No. 08-3327 (2d Cir. October 26, 2011) (Pooler, Raggi, CJJ, Korman, DJ)

Sanjaya Bahel, a chief procurement officer at the United Nations, was convicted of honest services fraud and bribery offenses in connection with a kickback scheme in which he improperly steered lucrative U.N. procurement contracts to a friend, in exchange for money. This long opinion covers a lot of very fact-specific issues. This post focuses only on the sentencing claim.

Both U.S.S.G. § 2B1.1 and U.S.S.G. § 2C1.1 can apply to fraud convictions. The difference is that § 2C1.1 applies to specifically to “public officials” and carries a higher base offense level. The district court sentenced Bahel under § 2C1.1, over objection; on appeal, and the circuit rejected his claim that the court should have used § 2B1.1.

Under the relevant definition, which is to be “construed broadly,” Bahel was a “public official.” That he was employed at an international organization does not take him out of § 2C1.1. Indeed, the commentary to § 2B4.1 specifically indicates that that “officials” of, inter alia, “public international organizations” are covered by Chapter 2, Part C. And other guideline provisions support this analysis. For example, the commentary to § 2C1.1 defines “public official” to include anyone who is “in a position of public trust with official responsibility for carrying out a government program or policy.”

As Chief of Commodity Procurement, Bahel was clearly a “high ranking U.N. official” and not the “baggage porter” with a “ministerial job” that he compared himself to, in reliance on a 1921 Supreme Curt decision. Rather, Bahel’s position was “closer to that of a foreign diplomat, political party official, or a tribal leader, all of whom are expressly covered by” § 2C1.1, not § 2B1.1.


Family Matters

United States v. Banki, N o. 10-3381-cr (2d Cir. October 24, 2011) (Cabranes, Pooler, Chin, CJJ)

Defendant Banki is an Iranian-born United States citizen. Starting in 2006, his family transferred about $3.4 million from Iran to the United States, all of which was effectuated through the “hawala” system. Banki’s hawala broker used a “matching” system to facilitate these transfers. When he knew that Banki’s family wanted to send money to the United States, he would find someone in the U.S. who wanted to send approximately the same amount to Iran. The U.S.-based contact would transfer into Banki’s account a sum comparable to the amount Banki’s family wished to send. Banki’s hawala broker would then pay an equivalent sum to the U.S.-based contact’s intended recipient, or broker, in Iran. Ultimately, Banki received some 56 hawala-related deposits.

Banki would typically email a family member to confirm receipt of each payment. Although most of his emails did not acknowledge a corresponding payout in Iran for each deposit, at least one such email - relating to a $6,000 payment in August of 2006 - displayed his knowledge that that particular sum of money was moving to Iran.

Banki’s financial activity came to OFAC’s attention in 2008 and, after receiving administrative subpoenas, Banki gave some spurious answers. He was ultimately charged, inter alia, with two counts of violating the Iranian Transaction Restrictions (the “ITR”). After a two-week jury trial, Banki was convicted of those counts, along with one count of operating an unlicensed money transmitting business and two counts of making false statements.

The circuit reversed the ITR convictions based on a flawed jury instruction. While the ITR have a service-export ban, at trial Banki argued that non-commercial remittances to Iran, specifically family remittances, were exempt. He sought a jury instruction to this effect, but the district court refused to give one. This, according to the circuit, was error.

The regulation at issue authorizes U.S. “depository institutions” to process transfers of funds to or from Iran, if the transfer “arises from an underlying transaction that is not prohibited by this part, such as a ... family remittance not related to a family-owned enterprise.” The government argued that this language permits such family remittances only if they are processed through a U.S. “depository institution.” But, to the circuit, the reg was “at a minimum” ambiguous.

Clearly, “family remittances” are “not prohibited” by the ITR. And, while this does not necessarily lead to the conclusion that they are permitted by the complete regulatory scheme, the language at issue “suggests that such actions do not contravene other applicable laws or regulations.”

And, more importantly, the government’s view - that only U.S. depository institutions are authorized to process the permitted transfers - is inconsistent with the language of the reg. A “fair reading” of the reg is that it tells U.S. depository institutions that they are permitted to process such remittances, but does not provide that they are the only entities that may do so. “Indeed, nothing in [the reg] specifically prohibits anyone from making a family remittance.”

Accordingly, after a lot of back-and-forth over the two sides’ competing views of the meaning - and purpose - of the reg, the court concluded that it was ambiguous. Interpreting it in Banki’s favor under the rule of lenity required that the two ITR counts be reversed.

The court also vacated Banki’s convictions relating to operating an unlicensed money transmitting business, again on a flawed jury charge. Banki wanted the district court to define a “money transmitting business” in a way that would make clear that it had to be an “enterprise” - not a “single transaction” - that was “conducted for a fee or profit.” Not only did the district court refuse, it actually told the jury that “a hawala is a money transmitting business.”

The charge as given was error. Banki’s requested instruction was “legally correct,” and there was a “foundation in the trial evidence” that the government proved Banki’s knowledge of money going to Iran in only a single transaction - the $6,000 transfer in August of 2006 - which the jury could have concluded was a one-time favor for a family friend. Moreover, the “hawala is a money transmitting business” charge compounded the error by “arguably reliev[ing] the government of its burden of proving that Banki’s knowledge that money was moving to Iran extended beyond the $6,000 transaction” and by suggesting that if it found that Banki participated in a hawala, “then he necessarily operated a money transmitting business.”


Between the Cracks

United States v. Rivera, No. 10-1199 (2d Cir. October 21, 2011) (Katzmann, Chin, CJJ, Gleeson, DJ)

This interesting decision answers an unanswered question in the circuit’s jurisprudence on § 3582(c)(2) motions. The outcome is favorable for Mr. Rivera, but will likely not last. An amended version of U.S.S.G. § 1B1.10 goes into effect on November 1, 2011, that is, at least arguably, intended to render defendants in his situation ineligible for a sentence reduction.


Convicted by a jury, Rivera faced a base offense level of 38 for trafficking in more than 1.5 kilograms of crack cocaine. There were no adjustments, so 38 was also his total offense level. He was in criminal history category IV, so his range would have been 324 to 415 months. But, he was a career offender. The highest offense level in the career offender table is 37, so the district court correctly “borrowed” the actual offense level of 38, and matched it to the career offender criminal history category of VI. This produced a range of 360 to life. The court then departed down from level 38 by 3 levels due to Rivera’s mental health; at level 35 and category VI, the range was 292 to 365. The court sentenced him to 292 months’ imprisonment.

Rivera moved pro se for a sentence reduction based on the 2007 retroactive reduction in the guidelines for crack cocaine offenses, but the district court appointed counsel for him. It then twice held that, as a career offender, Rivera was ineligible for a reduction. Here, the circuit reversed.

The Court’s View

Two circuit opinions serve as the background to this. In Martinez, see “PC World,” posted July 28, 2009, the court held that a defendant who received a career offender sentence was ineligible for a sentence reduction under § 3582(c)(2) and § 1B1.10, because his sentencing range “remains unaltered by the crack cocaine amendments.” In that situation, the sentence was not “based on” a range that the Sentencing Commission has subsequently lowered. On the other hand, in McGee, see “Crack a Smile,” posted January 25, 2009, the court held that, where the district court departs from the career offender range to the range provided by the offense guidelines, the defendant is eligible because the sentence is “explicitly based on the range produced by the offense guideline.”

This case is unlike either of those, between the cracks, as it were. Rivera received neither a career offender sentence nor a sentence within the original, non-career offender range. Rather, in his case, the court departed by three levels from the career offender range due to Rivera’s mental health, and the resulting sentence was below, not within, the original offense guideline range.

The court's view of the case turned on a fine analysis of the relevant provisions - § 3582's requirement that the original sentence be “based on” a range that has subsequently been lowered, and § 1B1.10's requirement that a defendant is only eligible if his “applicable guideline range” has been lowered. The court held that the two phrases should be understood to mean the same thing - “the range the initial sentence was ‘based on’ within the meaning of the statute is also generally the range that was ‘applicable’ within the meaning of the guideline.”

And, with that, the court had to decide which range Rivera’s sentence was “based on” - “that is, what was his ‘applicable’ sentencing range?” Was it the career offender range of 360 to life - or was it the departure range of 292-265?

The career offender guideline range would not change under the retroactive amendment - the offense level would drop from 38 (that was the crack quantity, which was higher than the career offender level of 37) to 37 (because applying the amendment to Rivera’s 3.3 kilograms of crack would reduce it to 36, so the career offender level would apply), but both 37 and 38 produce a range of 360 to life at CHC VI.

But, if his sentence was “based on” the range to which the court departed, he would be eligible for a reduction of up to 30 months. Under this scenario, the the starting point would have been level 37, and a departure to 34 would produce a range of 262 to 327. And this is the scenario that the majority went with: “We hold Rivera’s sentence was ‘based on’ the range produced by subtracting three offense levels from the career offender computation. The resulting range was the one the sentencing judge found to be ‘applicable’ to Rivera, and he chose a sentence at the low end of that range. That range is lowered when the retroactive amendment at issue is plugged into its calculation, even if everything else remains the same. Rivera is therefore eligible for a reduction.”

In the end, the circuit followed the reasoning of McGee, rejecting the government’s view that the sentence should be considered “based on” the pre-departure career offender range. Here, as in McGee, the court concluded that it would be excessively formalistic to conclude that the “applicable” range was the one rejected by the sentencing court. The court also found support for this methodology in the Supreme Court’s recent decision in Freeman, in which the plurality held that a court should “isolate whatever marginal effect the since-rejected Guideline had on the defendant’s sentence” and “revisit a prior sentence to whatever extent the sentencing range in question was a relevant part of the analytic framework the judge used to determine the sentence.” Here, that “marginal effect” is “easily isolated” as the 30-month difference between 262 and 292.

Finally, the court reinforced its view that its role in construing retroactive guideline amendments is so as to “allow inequalities to be fixed.” Accordingly, the court “let[s] lenity play a role in the construction of the Guidelines where there is doubt about their scope.” "Where the sentencing judge departs from a range computed under the career offender guideline to a lower range, the sentence imposed was ‘based on’ the latter range” under both § 3582(c)(2) and § 1B1.10. “If a subsequently-lowered guideline range was a relevant part of the analytic framework the judge used to determine the sentence ... a § 3582(c)(2) proceeding [should] be available to allow the sentencing court the opportunity to remedy an injustice.”

The opinion with a long rejection of the approach of other circuits, which have construed the phrase “applicable guideline range” in career offender cases to be limited to the career offender range, regardless of the range within which the defendant is sentenced. To the Second Circuit, that phrase should be construed differently in the context of § 1B1.10 cases, where the question becomes not what the original range was, but what range the sentencing court actually applied, “even if, as in this case, that range different from the one that was the starting point of the initial sentencing proceeding.” Thus, “[n]ow that Rivera seeks a modification of his sentence, the 292-365 month range to which his sentencing judge departed is his applicable range.”

The court closes with an acknowledgment that the November 1, 2011, version of § 1B1.10 will “dramatically alter” this “landscape,” because it will codify a construction of “applicable guideline range” that the court “refuse[s] to give the existing guideline,” and will limit it it to the “pre-departure range from the initial sentencing.” Recognizing that this change would “render Rivera himself ineligible for a sentence reduction if it were applied to his case,” the court held that it could not “fairly be applied retroactively to Rivera” on remand.

Judge Katzmann wrote a concurring opinion, in which he agreed with both the majority’s outcome and its reasoning. But Judge Katzmann would rely more heavily on the rule of lenity. To him, the relationship between § 3582(c)(2)’s “based on a sentencing range that has subsequently been lowered” and § 1B1.10's “applicable guideline range” is a “close and difficult question.” In McGee, the court resolved that question in the defendant’s favor under the rule of lenity, and to Judge Katzmann, McGee requires the same outcome here. The “ambiguity described in McGee does not disappear merely because Rivera’s departure falls under Chapter Five of the Guidelines Manual and was based on his diminished mental condition,” as opposed to the Chapter 4 departure at issue in McGee.

Labels: ,

Saturday, October 22, 2011

Thorn, Again

United States v. Thorn, No. 11-37-cr (2d Cir. October 20, 2011) (Jacobs, Sack, Raggi, CJJ)

This is Joseph Thorn’s third time in the circuit. Thorn ran an upstate asbestos removal company; he performed dangerous, substandard work, and used the money he earned to grow the business. In 2000, a Northern District jury convicted him of money laundering and environmental crimes, and the district court sentenced him to 65 months’ imprisonment. On the government’s appeal, the circuit vacated the sentence - the guidelines were mandatory then - and on remand the district court downwardly departed to 168 months from what it thought was a 235-month guideline minimum. The government appealed and won again. By this time the guidelines were advisory, however, so while the guideline minimum was now up to 292 months, the district court sentenced Thorn to 144 months.

Three years later, Thorn filed a 2255 motion, seeking to vacate his money laundering conviction as legally insufficient under United States v. Santos, 553 U.S. 507 (2008), which held that the term “proceeds” in the money laundering statute means “net profits,” and not “gross receipts.” The district court granted the motion, vacated the money laundering count, and resentenced Thorn to 132 months on the remaining counts.

On this, the government’s third appeal, the circuit reversed, agreeing that the Santos claim was procedurally barred because Thorn did not raise it on direct appeal, and no exception to procedural default applied.

While Thorn’s direct appeal argued that the money laundering conviction was legally insufficient, he challenged only the “intent to promote” element, not the “proceeds” element. And he could not establish cause and prejudice to excuse the default. Thorn argued that the “proceeds” theory was “so novel that its legal basis [was] not reasonably available to counsel” at the time. But the circuit disagreed. The futility test is strict - it asks not whether it would have been difficult to raise an issue but whether the claim was “available at all.” It was. By the time of Thorn’s direct appeal, several attorneys had argued for a narrow construction of the term “proceeds,” and the question was open in the Second Circuit. Even if it is true that the circuit would likely have rejected the claim had Thorn pursued it, he still could not establish cause - a defendant does not establish cause by showing “simply that a claim was unacceptable to that particular court at that particular time.” Accordingly, since Thorn did not establish cause, the circuit skipped the question of prejudice.

Thorn argued alternatively that he was actually innocent, which can also excuse a procedural default. But actual innocence means “factual innocence,” not mere legal insufficiency. Thorn had to demonstrate that “in light of all of the evidence,” it is “more likely than not that no reasonable juror would have convicted him.” Here, even assuming that Santos would have required proof that Thorn laundered only the profits, and not merely the receipts, of his fraudulent asbestos abatement scheme, he could not meet this standard. The trial evidence clearly established that Thorn’s company used money realized from existing abatement jobs to finance new projects, and the realized monies included profits.

With this, the court vacated the amended judgment and 132-month sentence, and ordered the court to reinstate the money laundering conviction and the 144-month sentence it had previously imposed.

Labels: , ,

Monday, October 10, 2011

PC World

There have been two interesting per curiams in the past couple of weeks.

Sometimes immigration law and criminal law intersect. They did in Prus v. Holder, No. 10-599-ag (2d Cir. September 28, 2011) (Calabresi, Wesley, Lynch, CJJ). Here, the court held that the New York offense of promoting prostitution in the third degree under Penal Law §§ 20.00 and 230.35 is not an aggravated felony. The term "prostitution" is not defined in the aggravated felony statute. But, under the immigration statute rendering aliens who enter the United States to engage in prostitution inadmissible, prostitution is defined as "promiscuous sexual intercourse for hire." Since identical words in different parts of the same act are construed to have the same meaning, the same definition should be used in the ag-fel section. But the New York offense of promoting prostitution encompasses a definition of "prostitution" - it includes "sexual conduct" - that is clearly broader. Since the New York offense includes acts other than the "specific act of sexual intercourse," it is not an aggravated felony.

United States v. Leslie, No. 10-2994-cr (2d Cir. October 3, 2011) (Calabresi, Wesley, Lynch, CJJ) (per curiam) rejected the argument that a defendant's incarceration, standing alone, should be treated as a withdrawal from the conspiracy for sentencing purposes. Leslie was the architect of a bank fraud scheme that capitalized on a flaw in the People's Bank ATM system. He not only devised the scheme, he also taught others how to do it. Leslie started serving a four-year state bank fraud sentence in 2005, but during his incarceration others continued the scheme. In 2009, Leslie was transferred to federal custody and pled guilty to the same scheme. The total loss was more than $300,000, but Leslie argued that he was only responsible for the losses incurred until his July 1, 2005, incarceration, although he admitted that he had taken no affirmative steps to withdraw from the conspiracy, such as cooperating with law enforcement or telling his co-conspirators that he had abandoned it. While the circuit agreed that incarceration could be some evidence of withdrawal, the defendant had the burden of showing something more, even for sentencing purposes. "The defendant's imprisonment is but one factor to consider in deciding whether withdrawal occurred."

A Bridge To FARC

United States v. al Kassar, No. 09-1051-cr (2d Cir. September 21, 2011) (Jacobs, Hall, CJJ, Scheindlin, DJ)

Defendants were convicted of various terrorism offenses in connection with a sting operation in which a CI, who was working for the DEA, introduced al Kassar to two undercover DEA agents posing as members of FARC, the left-wing Colombian guerrilla group. The defendants agreed to supply FARC with weapons, including surface to air missiles (“SAM”s) to use against United States military personnel and equipment in Colombia.

All of the criminal conduct occurred outside of the United States - mostly in Lebanon, Spain, Bulgaria and Romania. The district court denied the defendants’ motions to dismiss the indictment on jurisdictional grounds, and in this opinion, the circuit affirmed.

There is a presumption that acts of Congress do not apply extraterritorially, but even if the statute is not explicit, an intent can be inferred from the nature of the offense. Here, four of the five statutes of conviction contain explicit provisions applying them extraterritorially. The fifth, conspiracy to kill U.S. officers of employees under 18 U.S.C. §§ 1114, 1117 does not, but the nature of the offense - “protecting U.S. personnel from harm when acting in their official capacity” - implies an intent that it apply outside of the United States.

Even so, due process limits the extraterritorial reach of federal statutes. “There must be a sufficient nexus between the defendant and the United States, so that such application would not be arbitrary or fundamentally unfair.” For non-citizens acting entirely abroad, a jurisdictional nexus exists when the “aim of that activity is to cause harm inside the United States or to U.S. citizens or interests.”

Here, the nexus existed because the defendants’ stated aim was to “sell arms to FARC with the understanding that they would be used to kill Americans and destroy U.S. property.” Nor did it matter here that this case involved a sting operation taking place entirely outside of the United States and involving only foreign citizens. Since the goal of activity was to harm U.S. citizens or interests, or threaten the security or government functions of the United States, there is a sufficient jurisdictional nexus. And the goal alone is enough. It does not matter that the defendants never came close to harming any U.S. interest. Jurisdiction is determined by the “aims of the conspiracy, not its effects.”

The circuit also rejected the defendants’ “fair warning” argument. “Fair warning does not require that the defendants understand that they would be subject to criminal prosecution in the United States, so long as they would reasonably understand that their conduct was criminal and would subject them to prosecution somewhere.”

Finally, the circuit rejected the defendants’ claim of “manufactured jurisdiction.” Manufactured jurisdiction is a collection of three distinct defense theories: outrageous government conduct, entrapment, and a failure by the prosecution to prove an essential element of the crime.

None of those theories applies here. As to outrageous government conduct - also raised as an independent appellate claim - the DEA’s pervasive involvement in the weapons deal did not violate due process. There was no coercion, intimidation or physical force by the DEA agents. And the absence of a conspiracy prior to the government’s involvement merely shows that the government created the “opportunity for illegal conduct.” The tactics used were neither coercive nor outrageous - they were simply the “commonplace and often necessary tactics for infiltrating criminal enterprises.” Thus, while this was an “elaborate and prolonged” sting operation, “nothing done was outrageous or a shock to the conscience.”

Nor were the defendant’s entrapped “as a matter of law,” since they clearly had a predisposition. They already knew how to procure and smuggle arms, and reacted positively to the idea that the arms would be used to kill Americans.

Finally, the “unproved-element” theory is satisfied if the government “supplies an essential element of a crime”; “in effect,” the government has failed to prove it. But even if the government “initiates” an element of the crime, jurisdiction is not manufactured where the defendant “takes voluntary actions that implicate” it.” Here, every element of the crimes of conviction was established by voluntary action by the defendants. And the DEA agents did not create the jurisdictional nexus by “injecting the notion that the weapons were going to FARC for use against Americans.” That the DEA agents “lied to the defendants” does not make the nexus artificial or invalid.


Saturday, October 08, 2011

Aliens vs. Predator

United States v. Archer, No. 10-4684-cr (2d Cir. September 20, 2011) (Newman, Calabresi, Hall, CJJ)

Thomas Archer, a solo-practitioner immigration lawyer in Queens, ran a visa fraud mill. His specialty was the I-687, an amnesty program that permitted certain aliens who were here illegally in the 1980's to adjust their status and receive a visa. In 2004 and 2005, Archer filed nearly 240 I-687 applications; the DHS denied them all.

Convicted of visa fraud and conspiracy to commit visa fraud, his appeal concerned both trial issues - centered around his claim that he did know know that his assistants were filing forms with false information - and sentencing issues. The circuit affirmed Archer’s conviction, but remanded for resentencing and recalculation of the restitution.

The Trial Issues

At trial, an immigration agent who had reviewed 175 I-687 applications that Archer’s office filed, testified that almost all of them had certain suspicious factual allegations in common. That said, however, only three clients actually testified about the preparation of their fraudulent I-687's, and the government entered only four applications into evidence. The aliens' stories had much in common: their I-687 applications contained information that they knew was false; Archer or his staff gave them supporting affidavits for others to sign that were already filled in, and; Archer’s office told them to abandon the application process once they had received temporary work permits, but before their interview.

Archer’s principal trial defense was that he was unaware of the fraudulent actions of his staff. To this end, he requested two jury instructions. First he sought a “Philips” instruction that “the fact that a defendant is a solo practitioner, without more, is an insufficient basis from which to infer his guilt because, even though he is the only lawyer in the office, he may not be aware of everything his staff is doing.” He also sought a “Maniego” instruction that “attorneys are not held to a higher duty to investigate than non-lawyers and have no special obligation to verify independently information give to them by clients.” The district court rejected these requests and instead gave a fairly generic “knowingly” charge that simply told the jurors to consider whether Archer knew that the visa applications contained false statements but nevertheless presented them.

The circuit affirmed. Although it agreed that both the Philips and the Maniego instructions contain legally sound principles, here the instruction that the court gave was accurate and “left no room for the jury to convict Archer if it believed that he merely ran an office from which fraudulent documents were filed.”

The Sentencing Issues

The circuit found fault with the district court’s findings on two sentencing enhancements and its restitution order.

First, based solely on the agent’s statistical review of the 175 I-687 applications, the court enhanced Archer’s guideline range by nine-levels for creating 100 or more fraudulent documents, even though only the government admitted only four applications at trial. The circuit found that this statistical review failed the Shonubi “specific evidence” test. The principal problem was that the government “presented no evidence that the four applications proven false at trial were ... a representative slice of the 175 applications” that the agent reviewed because the government did not “randomly select[]” them; they were the applications associated with particular witnesses that the government chose to call and, most likely, “the most egregious cases.”

In addition, even though there was a suspicious statistical similarity among the applications reviewed, there was no “baseline” - evidence of what the national pool of I-687's, most of them likely filed by honest lawyers, actually looked like. Finally, the government did not explain why the similarities among Archer’s I-687 applications were “in themselves incriminating.” The only facts at issue were dates of entry and of travel and those facts were not “so peculiar” or “obvious” “that no further explanation is needed.”

The second sentencing error concerned the obstruction of justice enhancement. The district court imposed it because Archer texted a former employee, Singh, asking him whether he was going to be a government witness. When Archer concluded that the answer was “yes,” he texted Singh again, this time calling him a “Pussy.” According to the PSR, Singh felt “very threatened” by those messages.

Here, there was no direct and obvious threat; Archer’s statements to Singh were ambiguous. Where this occurs, the circuit usually defers to the district court’s findings on the speaker’s meaning and intent, but here the district court made none. And a “reasonable reading” of the messages would not support a finding of intent sufficient to support the enhancement. To the circuit, the “most obvious” reading of the texts was that Archer wished to know whether Singh would testify against him, and was displeased to learn that he would. Thus, even though Archer called Singh an “unpleasant name” and Singh was, subjectively, “afraid,” it was error to impose the obstruction enhancement.

The district court also erred in imposing restitution to 234 of Archer’s former clients - the total was more than $300,000 - because there was insufficient evidence that all of the clients were “victims” under the MVRA. The clients were only “victims” if Archer’s conspiracy to commit visa fraud caused their losses. This turns not so much on whether the aliens had “clean hands” but on whether their losses arose from the visa fraud or from an uncharged consumer fraud - Archer’s effort to cheat them of their money. After all, a person can commit visa fraud without accepting any money from the applicants.

If Archer’s clients though they were buying his honest legal services, then they may well have been victims of the visa fraud conspiracy. But, if they knew they were buying the “cover that his law practice gave to their false visa applications” then the visa fraud was not the proximate cause of their loss. There are some cases where it will be clear that no reasonable person would have given the defendant money if he had known of his plan. In those cases, a generalized description of the fraud is enough to support restitution. But, where it is “plausible that some individuals would have paid the defendant even if they had been informed of his fraudulent plan, then the government must proffer some individualized evidence to meet its burden of showing that each alleged ‘victim’ was actually a victim.”

This case is in that latter category. At least some of the aliens clearly knew that their visa applications contained falsehoods, but went along with the process anyway. In addition, filing a “false but plausible I-687 application was anything but a sure loser.” While the application was pending, the alien obtained a temporary work permit, and there was always the possibility, however, small the applicant would receive a visa. Given the resulting lack of certainty as to which clients were victims, the court remanded for recalculation of the restitution order.

Procedures on Remand

This decision has a particularly interesting discussion of the procedures that the district court is to undertake on remand. For sentencing issues, the “consensus” among the other circuits is “where the government knew of its obligation to present evidence and failed to do so, it may not enter new evidence on remand” unless the “government’s burden was unclear,” the “trial court prohibited discussion of the issue,” or the “evidence was, for a good reason, unavailable.”

The circuit “join[ed] that consensus” - sort of - but it still punted. It did not resolve the issue other than to remand for resentencing with instructions that the district court “consider in the first instance whether the justifications ... for allowing the government to present new evidence on remand exist in this case.” If the court allows no new evidence, it should recalculate Archer’s sentencing range without the 100-or-more-documents and obstruction enhancements. It if chooses to consider new evidence it should recalculate the range based on its findings with respect to that evidence and impose a sentence based on this and, of course, all of the other § 3553(a) factors.

For the restitution problem, there is yet another twist to the remand. To satisfy its burden, the government will have to show that each fee-paying client “did not know of the fraud and would not have paid a fee had” he known. Since the government has to “prove two negatives,” “some refinement in the proper allocation” of the evidentiary burdens is necessary.

Ordinarily, according go the circuit, “where the prosecution’s burden of proof would require it to prove a negative and the facts at issue are more readily ascertainable by the defendant,” the defendant assumes a burden of producing “at least a triable issue as to the fact at issue,” after which the prosecution assumes the burden of persuasion. That is the “appropriate” allocation of burdens here. Archer is “more likely than the government to ascertain whether a client knew of the fraud or would have paid a fee even if the client had known of the fraud,” since this knowledge would have come from Archer himself or someone in the firms.

Since the district court did not follow this program in the first instance, on remand, the court will have to reconsider which of Archer’s clients is entitled to restitution. The court will also have to consider whether to allow new evidence, as discussed above. But on the restitution question, the lack of clarity on the parties’ respective burdens “would seem to favor allowing additional evidence on the issue.” Even if there is no new evidence, however, the parties are “free to make new arguments based on the evidence already in the record.”


Mr. Archer now faces something of a dilemma. On remand, each client that he alleges knew what was going on - and hence is not a “victim” for restitution purposes - will add one at least more admittedly fraudulent document to the potential offense level enhancement. Will he stop at ninety-nine to keep his offense level lower, and make restitution to the others? Or will he go all the way, and claim that none of his clients were "victims?" In other words, which will he choose - money or freedom?

Labels: , , , ,

Wednesday, October 05, 2011

No Need to Remand Me

United States v. Elbert, No. 10-72-cr (2d Cir. September 19, 2011)(Jacobs, Cabranes, CJJ, Kravitz, DJ)

A recent anomaly in circuit practice has been its treatment of cases where the district court did not provide a written statement of reasons for the sentence that complies with 18 U.S.C. § 3553(c)(2). In cases where appellate counsel files a merits brief, counsel can waive a remand for a statement of reasons. But, where counsel files an Anders brief, under United States v. Hall, 499 F.3d 152 (2d Cir. 2007), failure to provide a statement of reasons always necessitates a remand.

Until now. This decision abrogates Hall to the “limited extent that it uniformly require remand in these circumstances.” Hall was based on the court’s understanding that the statement of reasons “assists” the BOP and the Sentencing Commission “in the collection of data.” While that is “no doubt for the good,” its effect is to require appellate counsel to seek a remand when “a remand would be of no benefit for the client” or “the court.” All Hall does is “set[] the lawyer to work for the” BOP and the Commission, and “doing a futile job of it in any event.” Thus, the bright line rule of Hall “may undermine, rather than serve, the goals of vigorous representation” that underlie the Anders brief procedures.

In a case where the absence of a written statement could benefit the client, counsel can seek a remand. But a bright line rule is not necessary to effectuate this, particularly since there might be situations where a remand for a written statement of reasons would be “detrimental to a defendant,” such as where the statement would memorialize the defendant’s cooperation or other information he would not want disseminated. “Requiring a defendant’s lawyer to elicit such information goes against the grain of advocacy.”

Even in the context of an Anders brief, then, counsel should “make an independent judgment as to whether deficiencies in a written statement of reasons present a non-frivolous appellate issue.”

Labels: ,

Tuesday, October 04, 2011

Re: Joinder

United States v. Page, No. 10-3150-cr (2d Cir. September 16, 2011) (Walker, Hall, Chin, CJJ)

Defendant was tried on five drug counts and a felon-in-possession count. In the district court, he moved to sever the gun count so that the jury considering the drug charges would not learn that he had a felony conviction. The court denied the motion and the circuit, finding no prejudice, affirmed.


In 2007 and 2008, Page was selling drugs - first crack, then heroin - in Norwich, Connecticut. During this time, he became involved in an altercation outside a bar, and brandished a gun; to avoid trouble, he stashed the gun at his girlfriend’s apartment. Agents raided the apartment the next day and found the gun and some drugs.

Page ultimately faced a six-count indictment; the first five counts alleged drug offenses - although the government ultimately dropped one of these - and count six charged the felon-in-possession. The district court refused to sever count six, noting that Page’s stipulation to the felony conviction did not describe its underlying facts, and that there would be a limiting instruction. These together assured a lack of undue prejudice. Page was convicted, and received a 210-month sentence.

The Circuit’s Decision

On appeal, Page argued that the gun count should have been severed, or at least bifurcated, from the others, citing United States v. Jones, 16 F.3d 487 (2d Cir. 1994). But the circuit found no abuse of discretion, particularly given the heavy burden of establishing prejudice by an allegedly improper joinder.

First, there was a “sufficient logical connection” between the drug counts and the gun count. The gun was recovered along with some of the drugs and Page admitted that both were his. For this reason, separate trials would have required “much of the same evidence.” Evidence of the presence of the drugs in his girlfriend’s apartment would have been probative of his knowing possession of the gun, while conversely, at a separate drug trial, evidence of the gun would have been admissible as a “tool of the trade.”

In addition, the district court took adequate measures to avoid prejudice. The stipulation was bare-bones and the limiting instruction was adequate, even though it did not specifically charge that the prior conviction could not be considered in relation to the narcotics counts.

Finally, there was overwhelming evidence against Page - his own confession and the testimony of the girlfriend.

The court also distinguished Jones on its facts. There, the felon-in-possession count appeared only in a superseding indictment after the jury deadlocked 10 to 2 for acquittal in a bank robbery case. Although the court reversed the bank robbery conviction because the felon-in-possession should have been severed, it did so because it looked to the court like the government had added the count only to “buttress its case” on the robbery. There was also a “retroactive misjoinder” problem with respect to a second felon-in-possession charge. But, since neither of those “unique circumstances” was present here, the district court did not abuse its discretion in refusing to follow Jones.

Jones does not stand for the proposition that a felon-in-possession count must always be severed or bifurcated from other charges. Where “there is a logical connection between” them, a “similarity in the evidence necessary to prove the different charges,” the trial court takes steps to limit the prejudice and gives a proper limiting instruction, and there is no unfair prejudice, it is not an abuse of discretion to refuse to sever or bifurcate. If, on the other had, the district court concludes that a bifurcation would “better protect the defendant from prejudice than a limiting instruction would” it is free to do so.

Labels: , ,