In a key sentencing decision handed down this year, the United States Supreme Court held that the Ex Post Facto Clause is violated when a defendant is sentenced under provisions of the Federal Sentencing Guidelines promulgated after he committed the crime and those new provisions result in an increased risk of greater punishment. In addition to clarifying the proper application of different versions of the Sentencing Guidelines, this is a particularly significant decision because the Supreme Court has now held that even post-Booker, an error in calculating merely advisory guidelines ranges still invalidates the sentence.
Marivn Peugh and his cousin Steven Hollewell were charged in 2008 with nine counts of bank fraud in connection with a check kiting scheme from 1999 to 2000 that allegedly caused the bank to suffer over $2 million in losses. Hollewell pleaded guilty to one count of bank fraud and was sentenced to one year and one day imprisonment. Peugh pleaded not guilty and went to trial where he testified that he had not intended to defraud the banks. Peugh was nonetheless convicted by the jury of five counts of bank fraud, although he was acquitted of the remaining counts.
At the time of Peugh’s offense (in 1999 and 2000), the 1998 Guidelines were in effect. Under the 1998 Guidelines, the base offense level applicable to his offense was six, and thirteen levels were added for a loss amount of over $2.5 million, creating a total offense level of nineteen. The government argued for an additional two level enhancement for obstruction of justice, which brought the total offense level to 21. Since Peugh was a first time offender in criminal history category I, he had an advisory sentencing range of 37-46 months under the 1998 Guidelines.
When Peugh was sentenced in 2010, the district court applied the 2009 Guidelines which were then in effect. Under the 2009 Guidelines, the base offense level applicable to Peugh’s conduct was now seven, and the enhancement for a loss value of over $2.5 million added an additional eighteen levels. After adding the two level enhancement for obstruction of justice, Peugh’s total offense level under the 2009 Guidelines was 27 – six levels higher than under the 1998 Guidelines. With a criminal history category of I, the advisory range for sentencing was 70-87 months – roughly double the range under the earlier version of the Guidelines. The district court sentenced Peugh to 70 months imprisonment, at the low end of the advisory Guidelines and he appealed the decision.
The U.S. Court of Appeals for the Seventh Circuit affirmed the sentence from the district court and quickly dismissed Peugh’s argument that the sentence violated the Ex Post Facto Clause. Relying on its own 2006 decision in United States v. Demaree, the Court held that the advisory nature of the Sentencing Guidelines post-Booker makes moot any argument that the application at sentencing of an increased Guidelines range at sentencing was not in effect at the time of the offense violates the Ex Post Facto Clause. This ruling was no surprise given that the Seventh Circuit has reaffirmed this proposition twice since it issued its 2006 ruling in Demaree.
The Supreme Court granted certiorari to resolve a Circuit split on this issue. On appeal, the focus of the Court’s analysis was on whether the Guidelines – which, post-Booker, are admittedly advisory – are sufficiently material to judges’ decisions about sentencing to warrant application of the Ex Post Facto Clause. In support of his argument, Peugh relied upon empirical evidence showing the judges are indeed influenced in their sentencing decision making by the Guidelines even if those Guidelines are not binding. On the other hand, the government argued that there was no precedential basis for the application of the Ex Post Facto Clause to a provision of law that is merely advisory.
In its holding the Court emphasized that the intent of the Ex Post Facto Clause was that it “ensures that individuals have fair warning of applicable laws and guards against vindictive legislative action.” Even where these concerns are not implicated, the Court held that the Ex Post Facto Clause also “safeguards a fundamental fairness interest.” The Court noted that, while the Guidelines are advisory, judges are still required, under Gall and by statute to begin their sentencing determination by correctly calculating the applicable Sentencing Guidelines range. The Court noted that continued vitality of the Guidelines in encouraging uniformity in sentencing by creating procedural hurdles that make the imposition of a sentence outside the guidelines range less likely. In doing so, the majority rejected the argument in Justice Thomas’ dissent that the advisory nature of the Guidelines means that do not “meaningfully constrain” a judges’ discretion.
The ruling in Peugh provides clear guidance to district judges that the version of the Sentencing Guidelines to be applied is the one in place at the time that the defendant committed his or her conduct constituting an offense. Of course, the Court’s ruling does not resolve how that principle will apply in cases involving charges such as conspiracy that may occur over a substantial period of time during which there may be multiple versions of the Guidelines. That issue and others will undoubtedly be the subject of litigation to come.
A $3 billion fraud scheme, more farcical than dangerous and in any case doomed to fail, led to 20-year sentences in federal prison for all four conspirators. The U.S. Court of Appeals for the 2nd Circuit, however, vacated the sentences on procedural grounds, and U.S. District Judge Stefan R. Underhill of the District of Connecticut, sitting by designation, wrote a concurrence that drew back the procedural curtain to shed light on what he saw as a fundamentally flawed corner of the administration of justice. This was the U.S. Sentencing Guidelines’ loss table, which he said was “divorced from its own objectives and from common sense” in this case.
The case, United States v. Juncal, came to the court on appeal from the District Court for the Eastern District of New York. The appellants – James Campbell, John Juncal, and Rodney Sampson – and their codefendant Emerson Corsey had been convicted of conspiracy to commit mail and wire fraud. The four men, posing as officers of a (fictional) Wyoming-based multinational bank and its client in Buryatia, an obscure region of Siberia, attempted to extract a $3 billion loan from a hedge fund to finance an (imaginary) Siberian oil pipeline. In exchange for the loan, they offered to assign to the hedge fund $5 billion in U.S. Treasury notes, which they claimed would generate a $14 billion return in just five short years. When a broker asked for physical evidence of the T-notes, the defendants explained that they had hidden the notes in Austria for safe keeping. The defendants did, however, send the broker copies of T-notes from their AOL account.
The absurd nature of these facts notwithstanding, the defendants’ offense levels were calculated based on an intended loss amount of $3 billion, and each received the statutory maximum sentence for fraud: 20 years in prison. At sentencing and on appeal, counsel for the defendants highlighted the significant flaws in the loss calculation, arguing that the “30-point mega-enhancement vastly overstated both the seriousness of the offense, and the danger of appellants to their community.
At sentencing, their arguments fell on deaf ears. On appeal, they did not. The Second Circuit questioned the lower court’s failure to apply (or even address the merits of) a reduced sentence and remanded the case for resentencing. Because the case was “clouded by the possibility of error,” the appeals court “felt it appropriate to give the District Court an opportunity to clarify its thinking.” The case was remanded on procedural grounds, and the appeals court declined the appellants’ request to consider whether the sentences were substantively unreasonable.
Judge Underhill began his nine-page concurrence by first agreeing that the sentences should be vacated and remanded for procedural error. However, he also noted that “the real problem is that the sentences are shockingly high.” For that reason, he “would reach the question of substantive reasonableness and would reverse on the merits.” In his view, “the loss guideline is fundamentally flawed, and those flaws are magnified where, as here, the entire loss amount consists of intended loss. Even if it were perfect, the loss guideline would prove valueless in this case, because the conduct underlying these convictions is more farcical than dangerous.”
Underhill went on to explain that the current guidelines are the result of three increases in the recommended ranges for fraud crimes, each of which “was directed by Congress, without the benefit of empirical study of actual fraud sentences by the Sentencing Commission.” He also noted the common perception that the loss guidelines are broken, and highlighted their widely inconsistent implementation among the district judges. However, since this case could be decided on procedural errors, the circuit court was able to remand the case without expressing a view on the substantive issues that Underhill highlighted.
In so doing, however, the appeals court may have overlooked an opportunity to fashion a common- law reasonableness standard to protect the administration of justice in future cases.
There are many arguments to support the avoidance of knotty substantive issues when their examination will not affect the final outcome of the case. As Underhill himself pointed out, courts ordinarily examine the procedural issues first before applying an abuse-of-discretion standard to examine the substantive reasonableness of a sentence. However, that practice creates a slippery slope: district court judges are forced to proceed without meaningful guidelines, and abuses of discretion go unnoticed.
A recent D.C. Circuit Court of Appeals decision narrows the ability of the government to revisit uncharged crimes against a person whose plea has been vacated due to a change in the law.
In 2007, Russell Caso had pleaded guilty to conspiracy to commit honest-services wire fraud, in violation of 18 U.S.C. §§ 371, 1343 and 1346, based on certain conduct during his employment as U.S. Rep. Curt Weldon’s chief of staff. Caso was sentenced to three years’ probation, including a 170-day term of home confinement. In entering its plea agreement with Caso, the government had forgone the right to charge Caso also with a violation of the false statements statute for failing to include certain payments on his annual disclosure statement required by virtue of his status as a federal employee.
Shortly after Caso was sentenced, the U.S. Supreme Court handed down its decision in Skilling v. United States, 130 S. Ct. 2896 (2010) – a decision that substantially limited the permissible reach of Section 1346, the honest-services fraud statute – with the result that Caso was indisputably innocent of the crime for which he was charged and convicted. The government did not dispute this point but nevertheless opposed Caso’s motion to vacate his conviction.
The government argued that Caso had procedurally defaulted his Skilling challenge because he had not directly appealed his conviction on the ground that the conduct to which he pleaded did not constitute an offense, and therefore was barred from raising this issue on a habeas petition. The government also argued that Caso had failed to satisfy the narrow conditions for excusing such a default that the Supreme Court set out in Bousley v. United States, 523 U.S. 614 (1998): (1) “cause” for the default and “actual prejudice” resulting therefrom; or (2) that the defendant is “actually innocent.”
In denying Caso’s petition (which argued only the second of these exceptions), the District Court agreed with the government, and focused on the Bousley Court’s rule that, “[i]n cases where the Government has forgone more serious charges in the course of plea bargaining, petitioner’s showing of actual innocence must also extend to those charges.” (emphasis added) Based on that rule, the District Court held that Caso had to demonstrate his “actual innocence” not only of the crime for which he was charged and convicted (honest-services wire fraud) but also of the separate uncharged offense of making a false statement, a crime that the government argued was at least equally serious as the honest-services fraud charge. Because Caso could not show his actual innocence of the false statement charge in light of the admissions he made as part of his plea agreement, the District Court denied his motion to vacate his conviction and sentence.
The D.C. Circuit reversed this decision based its reading of what constitutes “more serious charges” under Bousley. In doing so, the appeals court rejected the government’s argument that seriousness is to be determined based on the statutory maximum sentence for each crime, and found it far more logical to base the question of seriousness on the way in which each crime is treated in the United States Sentencing Guidelines. Quoting the Supreme Court’s Gall decision, the court noted that Guidelines calculations are still “the starting point and initial benchmark” for every sentencing decision and that “district courts must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing process.”
The court also noted that the United States Attorneys’ Manual, in directing prosecutors to charge “the most serious offense that is consistent with the nature of the defendant’s conduct,” explains that “[t]he ‘most serious’ offense is generally that which yields the highest range under the sentencing guidelines.”
The court also noted that statutory maxima provide the parties with little useful information in the context of plea negotiations, in part because courts rarely sentence defendants to the statutory maxima. Because the Guidelines treat a violation of the false statements statute less seriously than honest-services fraud, the Court of Appeals held that the forgone false statement charge was not “more serious,” and that Caso need not show his innocence of that charge to support his claimed right to vacating of his conviction for honest services fraud.
The fact that that the D.C. Circuit relied upon the Guidelines as the justification for its ruling is particularly interesting given that recent attacks on the reasonableness of some of the Guidelines (particular the Section 2B1.1 loss tables) have sapped the Guidelines of some of their authority. It is possible that this ruling could change the way in which prosecutors structure their pleas, but circumstances such as this one, in which a defendant is found innocent of convicted charges because of a change in the law, are rare enough that this is not likely. To the extent that courts face similar cases, they will have to address issues left unresolved by the D.C. Circuit, such as whether there must be contemporaneous evidence that prosecutors considered the forgone charge at the time, and whether a crime of “equal seriousness” (and not “more serious”) falls within the Bousley rule.
The U.S. Court of Appeals for the 11th Circuit recently ruled on an issue lying at the intersection of fraud conspiracies and the U.S. Sentencing Guidelines: the government’s separate burden of proof against each co-defendant when multiple plea bargains are entered. Specifically, the 11th Circuit was addressing whether the government presented sufficient evidence to show, in a credit card fraud case, that the defendant’s criminal activity affected at least 250 victims. Finding that the government had come dramatically short of meeting its evidentiary burden, the appeals court opened its opinion with a flare of witty admonition: “Sometimes a number is just a number, but when the number at issue triggers an enhancement under the Sentencing Guidelines, that number matters.”
The facts of this case are as interesting as the court’s tone. The defendant was Gary Washington, who pleaded guilty to four offenses related to his role in a credit card fraud conspiracy that affected more than 6,000 individual cardholders. At first blush, it stands to reason that his sentence was calculated using a level-6 enhancement, which is reserved for crimes affecting 250 or more victims. However, there was a critical issue that the government and the district court failed to appreciate: Washington didn’t enter the conspiracy until four months after its inception, so the full victim count couldn’t be summarily applied to him.
Remarkably, before the sentencing hearing, Washington conceded that “in all probability there were more than 250 victims.” However, his sticking point was that he wanted the government to submit “hard evidence” supporting a level-6 enhancement in place of its “verbal assurances.” The government essentially ignored his requests and proceeded to the hearing without submitting additional evidence. Washington objected again at the sentencing hearing, but the district court overruled his objection and applied the level-6 enhancement, noting that the figure had been applied to the other defendants’ sentences.
On appeal, the 11th Circuit found the government’s representations insufficient and stated that “evidence presented at the trial or sentencing hearing of another may not – without more – be used to fashion a defendant’s sentence if the defendant objects.” The appeals court pointed out that it was especially inappropriate to use the other co-defendants’ sentences as a guide, because Washington joined the conspiracy well after it began. Following this reasoning, the appeals court set aside Washington’s sentence and remanded the case to the lower court for resentencing. The 11th Circuit declined the government’s request to present additional evidence on remand, because nothing had prevented it from presenting sufficient evidence at the original sentencing hearing.
This case is another example of federal prosecutors and trial courts losing sight of our system’s fundamental canon: a defendant is innocent until proven guilty. In some instances, the procedural safeguards that protect this system may seem inefficient and unnecessary. However, the alternative would beckon trial courts down the slippery slope of replacing actual evidence with assumptions. Fortunately, the appeals courts are present as a way of reining them in.
Former Enron executive Jeffrey Skilling reportedly has negotiated a deal with federal prosecutors that is likely to result in a significant reduction of the prison sentence he will serve for his role in the collapse of Enron. Under the new agreement, Skilling faces between 14 and 17.5 years in prison — a 27 to 42 percent reduction relative to his previous sentence of 24 years. Apparently, Skilling’s aggressive defense wore prosecutors down in such a way that they are now willing to give up almost half of Skilling’s prison sentence to resolve the case once and for all.
In May 2006, Skilling was convicted on one count of conspiracy, 12 counts of securities fraud, five counts of making false statements to auditors, and one count of insider trading. As a result, he was sentenced to roughly 24 years in prison and ordered to pay $45 million in restitution.
Skilling appealed the convictions and sentence with some success. First, the U.S. Court of Appeals for the 5th Circuit vacated his sentence on the grounds that the U.S. Sentencing Guidelines had been misapplied. Then, the U.S. Supreme Court held that the trial record did not support his conviction for conspiracy to commit “honest services” wire fraud. On remand, the 5th Circuit found the “honest services” error to be harmless and upheld the conviction so all that remained was for Skilling to be resentenced.
Skilling’s attorneys were preparing to request a second trial based on newly discovered evidence, but the prosecutors evidently decided that the fight was not worth it. According to prosecutors, the government has invested extraordinary resources in bringing Skilling to justice, and a second round would impose even greater costs, delay resolution, and delay restitution payments to Skilling’s victims.
The parties’ agreement will facilitate closure by stipulating that a sentence in the range of 14 to 17.5 years is reasonable. Both parties have agreed not to contest a sentence within that range and have reserved their right to contest a sentence outside that range.
U.S. District Judge Sim Lake, the sentencing judge, is likely to agree with the parties, as a sentence outside the agreed-upon range would burden the parties with costs they would rather avoid.
Skilling is scheduled to be resentenced in the Southern District of Texas on June 21.
The U.S. Court of Appeals for the 3rd Circuit is currently considering a sentencing issue of great significance in cases in which a number of individuals work together to bring about a financial fraud. The question posed is the extent to which a defendant can and/or should be punished based on the profits made through the fraud when the defendant did not receive as much money from the fraud as his co-conspirators.
In Kluger v. United States, the appeals court must determine whether former attorney Matthew Kluger’s sentence was unduly harsh. Kluger was one of three men who pleaded guilty to insider trading last year in federal district court in Newark, New Jersey. In his plea, Kluger, who is 51, admitted that he stole data on about 30 transactions during 17 years at law firms that included Skadden, Arps, Slate, Meagher & Flom and Wilson Sonsini Goodrich & Rosati. The companies involved include Sun Microsystems, 3Com Corp., and Acxiom Corp. Kluger gave that information to his co-defendant, Kenneth Robinson, who in turn gave them to trader Garrett Bauer, who traded on the information and then sold at a great profit when the deals went public. Following the scheme, Bauer then distributed the money to his partners. Over the last four years of this arrangement, according to prosecutors, Bauer made about $32 million in illicit profits, while Robinson made more than $875,000. Kluger claims to have made more than $500,000.
The sentences that were meted out to Kluger and Bauer did not track this huge disparity in the benefit that each received from their illegal activities. Bauer was sentenced to nine years imprisonment. Kluger received a sentence of 12 years – the longest prison sentence ever given for insider trading, eclipsing the 11-year sentence received by Galleon Group co-founder Raj Rajaratnam. In sentencing Kluger, Judge Katherine Hayden said that she wanted to send a strong message about the “radiating effect of the loss of confidence in the market” caused by insider trading. Judge Hayden also emphasized Kluger’s abuse of trust given his position as a lawyer. Robinson, who cooperated with authorities and secretly recorded the other men for the FBI, received a sentence of only 27 months.
The notion that a defendant may be sentenced based on the aggregated gains of his co-conspirators is nothing new. Section 1B.1.3(1)(a)(1)(B) of the U.S. Sentencing Guidelines expressly provides that, “in the case of a jointly undertaken criminal activity,” relevant conduct (which sets the amount to be used to calculate upward adjustments in the loss table of Section 2B1.1) includes “all reasonably foreseeable acts and omissions of others in furtherance of the jointly undertaken criminal activity . . .” But the acceptance of this approach may be strained by cases of insider trading and other white-collar crimes that increasingly involve astronomical amounts of money, and therefore expose all participants to draconian criminal sentences.
In appealing Kluger’s sentence, his attorneys stressed that the district court appeared not to have considered the disparity in the amount of money that Kluger actually received as a result of the insider trading compared with at least one of his co-conspirators. This argument echoes some of the reasoning of Judge Jed Rakoff in his sentencing of Rajat Gupta, who likewise received far less benefit from insider trading than his co-conspirator, Rajaratnam. The issue raises an interesting question: Should a defendant’s sentence be commensurate only with his or her own personal gain? Or is the measure of the proper severity of a sentence the total gain obtained by all of the participants – an approach that appears to be more in step with the concept of “relevant conduct” that plays an important role in calculating advisory ranges under the Sentencing Guidelines?
The Third Circuit’s determination on this issue may signal the direction that the courts take on this issue, or may be just the first ruling in what becomes a split among the circuits. The resolution of this issue will be particularly important in cases in which Section 2B1.1 (the loss value table) plays a critical role in determining Guidelines sentences.
Justice may or may not be blind; but she can buckle under pressure. It may take years, millions of dollars and armies of attorneys, but if you have the resources to test her mettle, you too may tip the balance in your favor.
Almost seven years after his conviction on fraud and other charges, former Enron executive Jeffrey Skilling may finally be succeeding in his effort to cut down his prison sentence that was originally set at more than 24 years. His investment in his battle is nothing short of impressive. He apparently spent some $70 million on his defense in the underlying trial that ended in 2006 … and that doesn’t include the subsequent seven years of activity, which involves more than 1300 docket entries as of March 2013.
Skilling’s persistence may be paying off. The Department of Justice recently issued a notice on a proposed sentencing agreement with Skilling. (The notice provided that victims have until April 17, 2013, to express their views on the prospective agreement. No further timetables have been officially set.)
It may seem surprising that the Justice Department would consider entering a sentencing agreement with someone who has already been convicted and sentenced and is serving time. But this is a product of Skilling’s aggressive efforts since his conviction, which have resulted in several appearances before the U.S. Court of Appeals for the Fifth Circuit and in one successful trip to the U.S. Supreme Court.
In 2009, the Fifth Circuit vacated Skilling’s sentence – which is where the recently announced sentencing agreement comes into play. In 2010, the Supreme Court ruled that one of the legal theories behind Skilling’s conviction (the honest-services fraud theory) was unconstitutionally vague and remanded the case to the Fifth Circuit to decide whether any of the charges should be invalidated.
After more yo-yoing between courts (the Fifth Circuit upheld the conviction in 2011, the Supreme Court declined to hear a second subsequent appeal in 2012, and Skilling renewed his request for a new trial based on new evidence after the failed Supreme Court appeal), the Justice Department may be raising a white flag of sorts and opting to settle upon a sentence that is mutually acceptable to Skilling and prosecutors. The DOJ may be unwilling to spend more public resources on a man who won’t go away until he gets his way.
It is hard to say what the sentencing agreement will provide. We previously opined that in resentencing, the judge could sentence Skilling to somewhere between 15 and 30 years under the sentencing guidelines. Obviously a more stringent sentence than the previous 24-year sentence is not going to be the result of the prospective agreement between Skilling and the DOJ. Regardless of the terms, the agreement will need to be approved by the sentencing judge. And he will invariably have to balance, along with the scales of justice, the public outcry if the sentence is too light and the costs of continuing to do battle with Skilling.
There can be no dispute that the death of Aaron Swartz – the Internet activist who took his own life on Friday, January 11 – is tragic. There can also be no dispute that the grief and anger his family feel is very real. The question is what the appropriate focus for that anger should be in order to give meaning to Swartz’s life – and death.
Swartz, who had blogged about his own battles with depression, was a leading activist involved with the movement to make information freely available on the internet, and is credited with helping to lead the protests that ultimately defeated the Stop Online Piracy Act (SOPA) – a statute that would have significantly broadened law enforcement powers in policing internet content that may violate U.S. copyright laws. Swartz’s suicide came as he faced federal charges of wire fraud and computer fraud arising from his alleged efforts to make freely available an enormous archive of research articles and similar documents offered by JSTOR, an online academic database, through computers at the Massachusetts Institute of Technology. The allegations in the indictment he faced were a tribute to Swartz’s computer acumen, describing the technological means that Swartz had used to access and download approximately 2 million documents from the JSTOR subscription archive by unauthorized access to the computers at MIT.
Swartz’s family has released a statement in which they blame his death on the decision by federal prosecutors in the District of Massachusetts to pursue “an exceptionally harsh array of charges, carrying potentially over 30 years in prison, to punish an alleged crime that had no victims.” Contrary to the family’s assertion that the prosecution caused Swartz to take his own life, we suggest that the appropriate focus here is not on prosecutorial overreaching, but rather on Congress’s decision to criminalize certain conduct and to set sentencing guidelines that would likely have led to imprisonment if Swartz were convicted.
It is true that the maximum statutory sentence of imprisonment for the wire fraud charge in the indictment against Swartz is 30 years. But there is no question that the likely sentence that Swartz would have faced if convicted of wire fraud and/or the other charges in the indictment would have been far less than that. The advisory range under the U.S. Sentencing Guidelines would have depended on the loss (or intended loss) suffered, among other things, but Swartz likely faced (based on back of the envelope calculations) a sentence of no more than two to four years in prison – a fact that he almost certainly knew from the lawyer who represented him. While four years in federal prison is significant, it is much less than the 30-year sentence mentioned by the family.
It is also not entirely clear that the prosecutors’ decision to pursue charges against Swartz was unreasonable. This is not just a case alleging the distribution of materials protected by copyright law – an issue on which there is fair debate as to whether conduct should be criminalized. Rather, in this case, Swartz was accused of having accessed the MIT computer systems and the JSTOR subscription (for which MIT paid approximately $50,000) through illicit means. There were also allegations that Swartz’s computer intrusions crashed some computers and caused some legitimate subscribers to the JSTOR service to lose access for a period of time. Thus, assuming the truth of the allegations in the indictment, the alleged crime here was not entirely victimless. Moreover, everyone agrees that illegally accessing a computer system is not conduct that should be condoned. For these reasons, Swartz’s family’s attacks on the prosecutors as overreaching – while understandable given their grief and anger – may actually be misplaced.
On the other hand, there is a fair question whether the conduct with which Swartz was charged is really the kind of conduct for which we need to send a person with no other criminal record to prison for a period of years. That, however, is not an issue of decision-making by the prosecutor’s office. Rather, that is a question for Congress, both in terms of establishing criminal liability and in terms of setting astronomical maximum statutory sentences (which increased the base offense level for this crime). And it is a question for the U.S. Sentencing Commission, which has raised Guidelines levels over the years. It is also a question for Congress in terms of setting Guidelines scoring that increasingly fails to reflect any expertise of the Sentencing Commission, but rather reflects only a congressional mandate to support increasingly harsh advisory sentences under the Guidelines for white-collar offenses.
Prosecutors may have been justified in seeking charges against Swartz for his conduct. But if his family, friends and supporters wish Swartz’s death to have as much meaning as his life, they should focus instead on the decisions that created the harsh potential penalties that Swartz faced.
Andrew Strempler, a Canadian citizen who helped to pioneer the cross-border online pharmacy industry, was sentenced on January 9, 2013, to four years in prison in connection with allegations that his former company sold fake and misbranded drugs to U.S. citizens.
The sentence follows Strempler’s guilty plea in October in federal court in Miami to a charge of conspiracy to commit mail fraud. Strempler also agreed to forfeit $300,000 and pay a $25,000 fine. A hearing will be held to determine if Strempler will also need to pay restitution.
Strempler operated companies that sold foreign pharmaceuticals to consumers in the United States, where drug costs are significantly higher than in other countries. The drugs were obtained in markets with lower prices on drugs, but the U.S. government has long taken the stance that selling these drugs is illegal because the sources of the drugs could not be assured.
Under the plea agreement, the guidelines range for Strempler’s sentence would be 46 to 57 months, on a charge that carries no mandatory minimum sentence. The government recommended a sentence of 57 months. Prosecutors had originally sought up to 20 years in prison and the forfeiture of $95 million.
Counsel for Strempler asked the court for a downward variance and a sentence of 24 months. Strempler’s attorneys argued that since he is a Canadian citizen, any sentence imposed on him would be more difficult and onerous than an identical sentence imposed on an American citizen. They contended that he would likely not be assigned to a minimum security prison, even though he would likely otherwise qualify based on the nature of the offense and his lack of criminal history. Additionally, as a Canadian citizen Strempler would not be allowed to participate in an early release to a community corrections facility. After he serves his sentence he will be sent to immigration custody, where he will likely be held until his removal from the country.
Strempler’s attorneys also noted that the pre-sentence investigation report states that “there is no evidence that any victim sustained an actual loss or physical injury as a result of this offense.” Additionally, the forfeiture judgment of $300,000 to the government that Strempler agreed to pay prior to sentencing was nearly doubled the agreed-to loss amount.
According to court papers, Strempler believed that the drugs his company was selling were “safe and effective,” and his attorneys noted that he purchased the same drugs for his family and had sample drugs tested by a lab in Canada. His attorneys argued that he did not act with malice and had no actual belief that the drugs were fake and ineffective. He believed that the drugs were safe because they were purchased in accordance with the regulations of foreign countries.
The court essentially rejected the arguments by Strempler for a more lenient sentence and went along with the government’s request for a lengthy sentence. It appears to us that Strempler received a long sentence for a first-time nonviolent offender who did not act with malice. It seems that this is more of a regulatory violation parading in the clothing of a criminal case.
By asking for such a significant sentence, the government may have been trying to serve notice that this type of case will not be taken lightly. Given the stance taken by the prosecution in this case, it will be interesting to see if this leads to further prosecutions for related offenses.
On November 9, 2012, in a unanimous opinion in United States v. Fair, the U.S. Court of Appeals for the D.C. Circuit found that the district court had abused its discretion in ordering restitution in the amount of $743,000 in a criminal copyright infringement case. The appeals court vacated the lower court’s restitution order, finding that the order was based on “a clear legal and factual error.”
The appeals court emphasized that a restitution order may not be based solely on the ill-gotten gain of the defendant but must be directly related to the victim’s actual loss, which is not always the same thing.
In this case, Gregory Fair had offered customers an appealing but illegal way to acquire up-to-date Adobe software at less than half the retail price. Fair’s company sold outdated Adobe software (Photoshop and PageMaker) on eBay and included numerical codes that the buyers could use to purchase an update of the same software directly from Adobe. This scheme lasted from February 2001 until September 2007, when Fair was shut down by the United States Postal Inspection Service.
In 2009 Fair pleaded guilty to charges in exchange for a reduced sentence. Although Fair admitted to receiving roughly $1.4 million in revenue from the sale of pirated software on eBay, his plea agreement was based on an infringement category of greater than $400,000 but less than $1 million. Based on the Sentencing Guidelines for that category, Fair was sentenced to 41 months in prison followed by three years of supervised release. At sentencing, the prosecution also insisted that the court order the maximum restitution. The district judge ordered restitution of $743,000 to Adobe, based on prosecutors’ calculations of Fair’s ill-gotten gains.
At first blush that makes sense, right? Not so quick; there is a fatal flaw. Under federal law, restitution is not based on what the defendant gained; it’s based on what the victim actually lost. In many cases those are the same, but here it certainly was not.
At sentencing, Fair’s attorney raised this point in several different ways, emphasizing that the prosecution had completely failed to prove any actual harm to Adobe. This was a critical issue in this case, because Fair was actually directing each of his buyers to make a legitimate $200 purchase from Adobe. Fair’s attorney argued, correctly, that the burden was on the prosecution to show actual, provable loss (i.e., that purchasers of Fair’s outdated material, which Adobe no longer offered for sale, would have actually purchased the full-price, up-to-date merchandise from Adobe AND that the aggregate sales that Fair directed to Adobe were less than the sales that he supposedly thwarted).
The trial judge ignored Fair’s arguments, referred to the prosecution’s unsubstantiated calculation as “hard proof,” and fallaciously based the restitution order on his belief that it was “undisputed that Fair’s revenue from the sale of pirated products was at least $767,000.”
In so doing, the trial judge overstated the weight of the prosecution’s evidence and misinterpreted the law regarding restitution. As the appeals court explained, the purpose of the Mandatory Victim Restitution Act (MVRA) is “to compensate victims for the loss caused by the defendant’s criminal conduct.” Thus the trial court’s restitution order must be “limited to the actual, provable loss suffered by the victim” at the hands of the defendant. The appeals ruling made it clear that the prosecution must “articulate specific factual findings underlying its restitution order,” and it “may not substitute the defendant’s ill-gotten gains for the victim’s actual, provable loss.”
Prosecutors and judges must not lose sight of the fact that victims are free to seek full restitution in separate civil lawsuits. In fact, civil restitution suits actually allow for the disgorgement of all of the defendant’s profits, including those in excess of the victim’s loss. The Fair case is a perfect example of federal prosecutors and criminal trial courts losing sight of their role in the justice system. Fortunately, the appeals court stepped up to rein them in. In the words of D.C. Circuit Judge Judith Rogers, “the abuse-of-discretion standard may be generous, but it is not one that will countenance the clear legal and factual error present here.”