Sentencing Leakers: David Petraeus vs. Jeffrey Sterling

On April 23, 2015, David Petraeus — a retired four-star general, former commander of U.S. and NATO forces in Afghanistan, and former director of the CIA — was sentenced to two years probation and a $100,000.00 fine for mishandling classified information. Petraeus pleaded guilty to a single misdemeanor charge based on his disclosure of classified information to his mistress, Paula Broadwell, who was writing his biography.

On May 11, 2015, former CIA officer Jeffrey Sterling was sentenced to 42 months in prison for his unlawful disclosure of classified information to James Risen, a reporter for the New York Times. Sterling was convicted at trial of nine felony counts of violating the Espionage Act and related offenses for leaking information to Risen about a classified CIA operation to undermine Iran’s nuclear weapons program.

The two sentencings, occurring so close to each other, prompted inevitable comparisons.  For example, Mark Berman, a reporter with the Washington Post, tweeted that Sterling was “sentenced to same prison time as Petraeus, plus 3.5 years.”  The theme of much of the commentary, such as articles here and here, was that treating Sterling more harshly than Petraeus was unfair because the two men had committed similar crimes.  Many suggested Petraeus had received more lenient treatment simply because he, unlike Sterling, was powerful and connected.

Sterling’s lawyers also pointed to the Petraeus case at Sterling’s sentencing, arguing that the two cases were comparable and that “Mr. Sterling should not receive a different form of justice than General Petraeus.”

So what was the basis for the different sentences, and are the comparisons and criticisms justified?


Facts of United States v. David Petraeus

While he was commander of military forces in Afghanistan, Petraeus maintained small black notebooks that contained his daily schedules and personal notes from briefings and meetings that he attended. During his time in Afghanistan he filled up eight such notebooks. They contained classified information including identities of covert officers, war strategy, intelligence capabilities and mechanisms, and diplomatic discussions, along with details from National Security Council meetings and Petraeus’ conversations with the President.

After Petraeus returned from Afghanistan and retired from the military, the Department of Defense began collecting classified materials that he had accumulated during his tenure. Petraeus never informed the DOD officials gathering his materials about his black notebooks, and kept them in his possession at his house.

In 2011 author Paula Broadwell was working on a biography of Petraeus. She and Petraeus, who was married, were also having an affair. In August of 2011, Petraeus gave the notebooks to Broadwell for a few days so she could review them. (Broadwell apparently has a security clearance, but it is unlikely her clearance would have authorized her to view all of the material in the notebooks.) None of the classified information from the notebooks ended up in the biography, which was published in 2012.

In October of 2012, during an interview with the FBI, Petraeus lied and said he had never provided any classified information to Broadwell. In the wake of the investigation and the revelation of his affair with Broadwell, Petraeus ultimately resigned from the CIA in November, 2012.  On April 5, 2013, pursuant to a search warrant, the FBI seized the black notebooks from an unlocked desk drawer in Petraeus’ house.

 CIA logo

Facts of United States v. Jeffrey Sterling

Jeffrey Sterling was a ten-year CIA employee with access to extremely sensitive information. In particular, he was deeply involved in a project known as “Operation Merlin,” a covert operation to undermine Iran’s nuclear weapons program. Merlin involved using a former Russian scientist, now living in the United States and cooperating with the CIA, to feed faulty nuclear weapons plans to Iran. The idea was to send Iran down some blind alleys in order to delay and disrupt its development of a nuclear weapon.

In the early 2000’s, Sterling began filing a series of personnel grievances and lawsuits against the CIA, alleging racial discrimination in connection with his employment and with CIA decisions about whether he could publish certain material in his memoirs. Over several years his lawyer made escalating demands to the CIA for financial settlements of the disputes, which the CIA refused.

During this same time period, Sterling began having meetings and telephone conversations with James Risen. Sterling disclosed classified information about Operation Merlin to Risen and apparently gave him copies of classified documents. According to the evidence at trial, Sterling provided Risen with a distorted and inaccurate view of the operation, in order to make it appear the CIA had bungled the program and may have actually aided Iran rather than hindering it.

In the spring of 2003, Risen informed the CIA of his intention to publish a story about Operation Merlin. National Security Advisor Condoleezza Rice and other government officials met with Risen’s editors to express grave concerns about the national security damage that could result if the story was published. In light of the government’s concerns, Risen’s editors agreed to kill the story. Risen was reportedly furious. Two years later he wrote a book about the CIA, State of War, in which he devoted a chapter to Operation Merlin and disclosed the classified information that his editors at the Times had agreed to protect in the interest of national security.

Comparison of the Two Cases: Apples to Apples?

The Petraeus and Sterling cases reached their denouements just a few weeks apart, with dramatically different results for the two defendants. But although it’s tempting to equate the two cases, as many have done, it is really a comparison of apples and oranges. Both involve the improper disclosure of classified material, but that’s about where the similarities end.

Intent: A defendant’s intent is always an important factor when determining the appropriate charges and punishment for a given offense. The government’s evidence at trial was that Sterling intended to harm the government and was motivated, as the government put it, by “pure vindictiveness” and “spite.” Angry with the CIA over what he perceived as wrongful personnel and other actions, he disclosed highly sensitive material and painted a distorted picture of a covert operation in an effort to embarrass and harm the agency. In so doing he appears to have had little concern for the possible damage to national security that might result.

Sterling also plainly intended that the classified information he revealed would be publicly and widely disclosed. By giving the information to a prominent national security journalist for one of the nation’s leading newspapers, he maximized the likelihood that the sensitive, classified material would be made public – which of course also maximized the chance that national security would be harmed.

With Petraeus, there was no evidence of any malicious intent or desire to harm national security. It’s hard to say what exactly motivated him; probably a combination of ego, wanting to make sure that the historical record of his accomplishments was as full and accurate as possible, and a desire to please the woman with whom he was having a relationship. But there was never any suggestion that Petraeus intended to cause any harm.

There also was no evidence that Petraeus intended for any classified information to be made public; it appears he was trying to give Broadwell background information and context for the book. Even if Broadwell had included classified material in a draft, it’s likely that review of the proposed book by government agencies or by Petraeus himself would have caught it and stopped it from being published.  Petraeus (unlike Sterling) had no reason to think that the recipient of the classified material he disclosed had any desire or motive to reveal it to the world.

This difference in their level of intent also helps to explain the different charges of which the two men were convicted. The Espionage Act, the felony Sterling was found guilty of violating, includes a requirement that the defendant knew the disclosed information could be used to injure the United States and to benefit a foreign nation. Mishandling classified information, the misdemeanor to which Petraeus pleaded guilty, contains no such element.

Actual Harm Caused: The government maintains that the damage caused by Sterling makes his one of the most serious leak cases in recent memory. Sterling’s disclosures, according to the government’s evidence at trial, shut down an active covert operation and one of the few mechanisms the government had to restrain Iran’s nuclear ambitions. Unlike other recent leakers, Sterling disclosed an ongoing, classified operation involving human foreign intelligence assets. The same type of operation had been used not only against Iran but against other nations as well. Former National Security Advisor Rice testified at Sterling’s trial about how important Operation Merlin had been and how damaging the disclosure of the program was.

There was also testimony at trial, including from the Russian scientist himself, that as a result of Sterling’s disclosures the scientist and his family now live in fear that a foreign government may harm them in retaliation for his cooperation with the United States.

At Sterling’s sentencing, the judge noted that his leaks effectively disclosed the identity of a covert asset who was working with the CIA and to whom Sterling had been assigned. The judge observed that this kind of disclosure is one of the most serious breaches of a CIA official’s duty. It betrays the trust that an asset placed in the CIA and may make it more difficult for the government to recruit other foreign human assets in the future.

By contrast, by allowing Broadwell to read his notebooks, Petraeus appears to have caused no actual harm at all. As discussed above, none of the material was ever publicly disclosed, and there appears to have been little risk that it would be. Although allowing Broadwell to review the materials and keeping them in an unlocked desk in his home was a serious breach of the rules for handling classified information, it was more of a procedural violation than an actual betrayal.

Plea vs. Trial: Petraeus accepted responsibility for his conduct and pleaded guilty. Sterling never accepted responsibility for his conduct, took the case to trial, and was ultimately convicted by a jury.

Defendants who plead guilty almost always get some credit from the government, as well as from the sentencing judge. This age-old concept is also built in to the federal Sentencing Guidelines, which provide significant benefits to a defendant who pleads guilty before trial. A guilty plea is a sign of contrition and that the defendant has accepted responsibility for his conduct, behavior worthy of favorable consideration by the sentencing judge. In addition, by pleading guilty a defendant saves the government from the considerable time, effort and expense required to take a case to trial. In a case like Sterling’s, this is particularly important because the trial itself may require the government to reveal at least some classified information that it would prefer to remain secret.

Sterling’s lawyers argued that it would be unfair to penalize him because he exercised his constitutional right to put the government to its proof. But this simply gets it backwards: it’s not that a defendant who goes to trial is penalized; it’s that a defendant who pleads guilty gets a benefit. That has always been true, and is another factor that distinguishes the Sterling case not only from Petraeus but from other recent leak cases pointed to by Sterling’s lawyers involving John Kiriakou and Stephen Kim, who also pleaded guilty.

We don’t know what kind of plea discussions took place between the government and Sterling, but there’s little doubt that if Sterling had been willing to step up and admit responsibility early in the case he could have received a much more favorable deal and a more lenient sentence. Similarly, had Petraeus stonewalled the government and refused to plead, he very likely would have faced far more serious charges.

Personal background: A judge always takes a defendant’s personal history into account when fashioning an appropriate sentence. That’s one of the reasons that Presentence Reports are prepared, to give the sentencing judge detailed information about the background and accomplishments of the defendant. It’s also the reason people write letters to the sentencing judge in support of a defendant and urging leniency, such as the more than 400 letters that were sent to the judge who was recently preparing to sentence former Virginia Governor Bob McDonnell, attesting to McDonnell’s good character and record of public service.

Petraeus came before the court after 37 years of serving the country in the military and, most recently, in the CIA. He was considered one of the most able military leaders in the nation and to have done an admirable job commanding forces in Afghanistan. That type of career of public service entitles a defendant to considerable credit. It makes his offense look all the more like an isolated lapse of judgment in an otherwise sterling career where he sacrificed a great deal for his country.

For his part, Sterling has some impressive personal accomplishments as well, including graduating from law school, his career at the CIA, and a subsequent successful career as a fraud investigator for an insurance company. But in terms of overall record and service to the nation, he is no David Petraeus. That is particularly true considering his efforts, near the end of his CIA career, to undermine and damage the agency and country he had sworn to serve.

*  *  *

 In sum, it’s much too simplistic to look at the two cases, see that they both involved leaks of classified information, and conclude that the different sentences are unjustified or arise from some improper reason. Every case is unique, and here there are substantial differences that explain the different sentences.

Actually, in both cases the sentences seem about right to me. In Sterling’s case, it’s worth noting that the sentencing guidelines actually called for a sentence of about 19 to 24 years in prison. The judge rightly recognized that in a case involving leaks to the media, as opposed to actual espionage against the United States, such a sentence would have been crazy. (Even the government seemed to acknowledge this and did not argue for a sentence within the guidelines, as it normally would, but simply asked that the sentence be “severe.”) In light of the harm caused by Sterling’s conduct, his malicious intent, and the sentences received by other recent leakers of classified information to the press, Sterling’s sentence of 42 months appears eminently fair, if not a bit lenient. Even his own lawyer, after the sentencing, said that the judge “got it right.”

As for Petraeus, I’m not entirely unsympathetic to the view that he got a pretty sweet deal. This is particularly true considering he initially lied to the FBI about revealing classified material to Broadwell. Lying to the FBI can be prosecuted as false statements, a five-year felony. There were reports that the FBI agents working on the case were upset and thought that Petraeus should have been treated more harshly, given his lack of candor. That view has some force.

On the other hand, few defendants who end up entangled in the justice system have Petraeus’ record of a lifetime of distinguished and valuable service. The notebooks were his personal notes, not original classified materials, and one can see how he might have viewed them as somewhat less sensitive. And his lie — a simple denial of guilt when interviewed by the FBI — is a type of case that, as a matter of policy, the Department of Justice generally does not prosecute.

This was a serious screw-up by someone who knew better. But given his overall history, lack of intent to harm the country, lack of any actual damage resulting from his conduct, and willingness to accept responsibility and plead guilty, I can’t say that Petraeus should be going to jail or that his plea deal seems unreasonable.

The Key Issues in Bob McDonnell’s Appeal

Update:  On August 11, the Fourth Circuit denied McDonnell’s motion for rehearing and rehearing en banc.  McDonnell will now seek review by the U.S. Supreme Court.

Update:  On July 10, the three-judge panel of the Fourth Circuit unanimously affirmed McDonnell’s convictions.  He is now seeking en banc review by the entire court.  The panel opinion is here

Former Virginia Governor Bob McDonnell’s appeal will be argued tomorrow, May 12, before the U.S. Court of Appeals for the Fourth Circuit in Richmond, VA. McDonnell and his wife Maureen were convicted of multiple counts of corruption for accepting a series of extravagant gifts and sweetheart loans from businessman Jonnie Williams in exchange for using the power of the Governor’s office to promote Williams’ product, Anatabloc.

McDonnell 2

The key legal issues in the appeal can be broken down into four categories:

1) Were the things McDonnell did in exchange for the gifts from Williams “official acts” for purposes of federal bribery law?

2) Did the trial judge err by refusing the McDonnells’ request that they be tried separately?

3) Did the trial judge err during jury selection by not probing more thoroughly whether pre-trial publicity had prejudiced any of the potential jurors against the McDonnells?

4) Did the trial judge err in several evidentiary rulings against the defendants?

McDonnell has an impressive array of supporters before the Fourth Circuit. Nearly a dozen amicus briefs were filed on his behalf by groups including the National Association of Criminal Defense Lawyers; six former Virginia Attorneys General; forty-four non-Virginia state Attorneys General; several Virginia law professors; and members of the Virginia General Assembly. To one degree or another, most of these supporters claim that political life in Virginia – if not American democracy itself – will be irreparably harmed if McDonnell’s conviction is not overturned.

I believe these concerns are misplaced. As an example of quid pro quo corruption, the McDonnell case is really not that remarkable. In fact, the most remarkable thing about it is that the central legal argument throughout the case has been about the language of a statute that is not even part of the indictment. How the Fourth Circuit deals with that issue in particular may end up being the key to McDonnell’s fate.

The Definition of “Official Acts” – A Focus On the Wrong Question

By far the most significant issue on appeal, and the one likely to consume most of the court’s attention, is the first: the question of “official acts.” Since before the case was even indicted, the backbone of McDonnell’s defense has been that whatever he may have done for Williams in exchange for the gifts, his actions were merely political courtesies and did not violate federal bribery law because they were not “official acts.” Arguments about this point consume the bulk of the briefs on both sides, as well as most of the amicus briefs.

The term “official acts” comes from the federal bribery statute,18 U.S.C. § 201, which defines it as “any decision or action on any question, matter, cause, suit, proceeding, or controversy” brought before a public official in their official capacity. 18 U.S.C. § 201(a)(3).  Section 201, however, applies only to federal public officials and was not part of the McDonnell indictment. The corruption charges they were convicted of fall under two other federal statutes: honest services fraud and Hobbs Act extortion under color of official right. Both of these statutes operate as bribery by another name and provide a vehicle for federal prosecution of state, local, and even private sector bribery.

As I have argued in an earlier post, however, bribery for purposes of honest services fraud and the Hobbs Act is not limited to the definition contained in 18 U.S.C. § 201. Those statutes apply more broadly to general common law bribery, not to a specific statutory definition. Indeed, not a single case cited in any of the appellate briefs stands for the proposition that bribery under honest services fraud or the Hobbs Act is defined by the language of 18 U.S.C. § 201.

Nevertheless, throughout this case, both sides have taken the position that the government is required to prove McDonnell performed “official acts” as defined in Section 201. It’s not completely clear how the case arrived at this posture, but it is clear that the emphasis on this particular language favors the defense. Had the parties and the court focused on a more generic definition of bribery, the case would have looked quite different.

For example, Virginia state law and the Model Penal Code both define bribery as a public official accepting a pecuniary benefit in exchange for being influenced in a “decision, opinion, recommendation, vote or other exercise of discretion as a public servant.” VA Code §18.2-447(2). There is no requirement of an “official act” as defined in Section 201, and the broad application to “recommendations” and “exercises of discretion” certainly seems to cover much of McDonnell’s conduct.

There are scattered references throughout the appellate briefs and amicus briefs to the idea that the definition of bribery in Section 201 may not in fact be the correct standard, with the defense continuing to insist that it is.  In the end, though, for the most part everyone proceeds on the assumption that the language of Section 201 applies, as the parties did during the trial. As a result, much of the legal argument is devoted to whether McDonnell’s conduct amounted to a “decision” or an “action” “on” a “matter, cause, suit, proceeding, or controversy” within the meaning of Section 201’s definition. Rarely has so much appellate ink been spilled over the language of a statute that no defendant in the case was charged with violating.

The Effect of the Focus on “Official Acts”

This obsession with the nature of McDonnell’s actions has caused the real issue to be obscured. In a bribery case the focus is less on the nature of the act itself and more on the corrupt deal: the agreement of the public official to be influenced in the exercise of his or her official powers in exchange for the bribe. In fact, although you’d never know it from the pleadings in this case, even federal bribery under Section 201 does not require an “official act” – that is only one of three alternative ways to violate the statute. The definition of “official act” is actually much more important to the lesser crime of gratuities (also in Section 201) than it is to bribery, and indeed many of the main cases relied upon by McDonnell are gratuities cases, not bribery cases. The implication of many of the arguments that federal bribery law always requires “official acts” is simply incorrect.  What bribery does always require is the element of influence.

Bribery is considered corruption because it alters the behavior of a public official, who acts not for the good of all as he or she is sworn to do but in response to an improper benefit received from a particular individual. This corrupt exchange in a bribery case is often referred to as the quid pro quo. In McDonnell’s case, the excessive focus only on the “quo” – McDonnell’s actions — is what leads to the fears expressed by the defense and many of the amicus briefs about the implications of McDonnell’s conviction. They argue that if the types of things McDonnell did for Williams are criminalized, it will “wreak havoc” on the political life of Virginia, “criminalize wide swaths of political life,” and make “virtually every elected official in the Fourth Circuit a criminal.”

What these arguments fail to grasp is that it is not the nature of McDonnell’s actions that’s the problem: it’s the corrupt deal that led to them. Routine political acts and favors done for supporters are generally not bribery not because they are not “official acts,” but because they are not done as part of a corrupt quid pro quo.  Law abiding politicians who are simply serving their constituents and not making corrupt deals have nothing to fear from McDonnell’s conviction.

The defense and various amici further obscure this point by noting that it’s possible for a campaign contribution to be a bribe. As a result, they claim, if the McDonnell conviction is upheld then any politician who takes any kind of routine action for someone who has donated to their campaign would risk being branded a felon.

Again, the flaw in this argument is that it focuses on only one side of the equation – in this case, the quid – instead of focusing on the corrupt agreement itself. Legal campaign contributions generally do not support criminal charges simply because they are not considered part of a corrupt bargain. For better or worse we have a system of privately financed campaigns in this country, where it is legal and appropriate for supporters to contribute to politicians in the hope that they will act in certain ways and for politicians to respond to the concerns of their lawful supporters. Properly reported campaign contributions within legal limits, without more, are not corrupt.

Gifts like those in the McDonnell case are another matter. Secret sweetheart loans, payments for a daughter’s wedding, expensive golf outings, Rolex watches, and steps taken to hide all of the above from the public – these are all things that, unlike legal campaign contributions, have the whiff of corruption about them. Contrary to the defense’s implication, not all quids are created equal.

A campaign contribution can be a bribe – in rare cases – but the other indicia of corruption must be much greater, including an explicit agreement by the politician to take a particular action in direct exchange for the contribution.  The point is that in a bribery case it is the overall corrupt deal and the question of influence that has to be examined, not just the quid, and not just the quo.

The focus on “official acts” has been largely a diversionary tactic that allows the defense to deflect attention away from the corrupt deal that the jury, by its verdict, necessarily found existed between Williams and the McDonnells. But the logical implication of the defense argument is that a businessman could secretly give the Governor tens of millions of dollars in exchange for an agreement to host an event at the Governor’s mansion and that deal would not be illegal because it did not involve an “official act.” That is definitely not the law – not even in Virginia. The powers and resources of the Governor’s office are not up for sale to the highest bidder.

The real issue is not whether the actions McDonnell took to help Williams met some precise statutory definition. The crux of the case is that he exercised his discretion and the powers of his office to benefit Williams in exchange for a two-year pattern of secret gifts and loans. That is classic quid pro quo corruption. The interesting question on appeal will be whether the Fourth Circuit focuses on that fact or gets sidetracked into debates about the meaning of a statute that is not even part of the case.

The Other Three Issues: Not Likely to Be a Major Factor

Voir Dire:  The defense also is challenging the process by which the jury was selected, known as voir dire. Their claim is that the judge failed to probe sufficiently whether negative pretrial publicity about the case might have prejudiced potential jurors against the McDonnells.

The process of jury selection is firmly committed to the discretion of the trial judge and it’s very rare for cases to be overturned on this basis. Appellate courts recognize that, particularly when it comes to pre-trial publicity, the trial judge is in the best position to evaluate what the nature of that publicity has been and how best to deal with it, because the judge also lives in the same community. The judge is also in the best position to judge the demeanor and credibility of the potential jurors standing in the courtroom and responding to questions about their ability to be fair and impartial.

Another important factor is that the jury acquitted the McDonnells on several counts of the indictment. This suggests a jury that did its job and considered each charge individually on its merits, not one that was biased and predisposed to convict regardless of the evidence.

Although there was a lot of pretrail publicity, this is not a gruesome murder, racially-charged case, or other case where there might be heightened concern about the passions of the community being inflamed. It’s possible this could be a sleeper issue, with some of the appellate judges concerned that the voir dire was a bit too perfunctory. In the end, though, I doubt this argument will gain much traction in the Court of Appeals.

Severance: McDonnell also argues that it was error for the trial judge to refuse to try the Governor and his wife separately. He claims that if had been tried alone his wife would have testified at his trial and provided helpful evidence.

There is a presumption that co-defendants will be tried together, particularly in a conspiracy case, and severance is rarely granted. It may be true that Mrs. McDonnell’s testimony would have been useful to the Governor, but that’s not the standard. Separate trials are only required if the failure to sever the defendants would deprive the defendant of a fair trial and result in a miscarriage of justice.

As the government has pointed out, much of the proffered testimony from Mrs. McDonnell ended up being presented to the jury through other witnesses. Although she may have denied that the Governor knew about some of the gifts, she could not deny that he knew about many of the others, including the expensive golf outings and the loans from Williams that he personally requested. Much of her testimony would have been vigorously challenged and impeached by the government. And of course her effectiveness as a witness would have been limited by her obvious bias and motivation to exonerate both her husband and herself. The appellate court is extremely unlikely to second-guess the trial judge’s refusal to sever the trials.

Evidentiary Rulings:  Finally, the defense argues that the judge erred in a number of the evidentiary rulings throughout the case. These include the judge’s refusal to allow defense expert testimony about Virginia’s financial disclosure forms and to admit evidence about McDonnell’s completion of those forms; the decision to admit evidence suggesting that the Governor was actively looking for places where he could play golf for free; the decision to allow evidence about another $23,000 golf vacation paid for by another supporter of McDonnell’s that the Governor failed to report; the refusal to allow additional evidence and defense expert testimony about the immunity agreement granted to Jonnie Williams; and the failure to allow the defense access to the contents of William’s iPhone.

A judge is called upon to make dozens, if not hundreds, of such rulings during a lengthy trial like this one, and those rulings will not be disturbed unless the judge abuses his or her discretion. The government argues that each of these rulings was correct. Even if any of them was an error, however, it’s extremely unlikely that the error would be significant enough to justify a new trial.

The Bottom Line

The “official acts” question has been the central issue throughout this case and will likely determine the outcome of the appeal.  A good result would be for the Fourth Circuit to write an opinion recognizing that, even though the government was not required to prove the exact elements of Section 201, the evidence at trial was more than sufficient to establish common law, quid pro quo bribery.  McDonnell’s best hope is that the Court of Appeals gets bogged down in all the debate about “official acts” and loses sight of what the case is really about: a corrupt arrangement where he and his wife put the powers of the Governor’s office up for sale.

Click here to sign up for the Sidebars mailing list and receive notifications of future posts.

The New Sentencing Guideline for Fraud Cases

Update: the proposed amendments to the fraud sentencing guideline discussed in this article took effect on November 1, 2015.

On April 30 the United States Sentencing Commission formally submitted to Congress its annual proposed amendments to the Federal Sentencing Guidelines. Unless Congress intervenes, the amendments will take effect on November 1st.

Several amendments affect the guideline for economic crimes and fraud cases, § 2B1.1, probably the most commonly-used guideline for white collar crimes. The fraud sentencing guideline has been criticized for years for being too driven by loss amounts that may be a poor measure of culpability or difficult to calculate, and for calling for unreasonably high sentences. The proposed amendments seek to address some of those concerns, but stop short of the wholesale overhaul that some had hoped for.

The overall effect of the amendments will be to reduce the recommended sentence in many fraud cases, particularly those involving large dollar amounts. At a public hearing on March 12, the U.S. Department of Justice opposed most of the proposals, arguing that any move to reduce the sentences in fraud cases would be bad policy and would ignore the “overwhelming societal consensus” in favor of harsh punishment for these crimes.

Given the current realities of federal sentencing, I believe DOJ’s opposition was misplaced.   There are legitimate concerns about the effect of the fraud guideline in some cases. If those concerns are not addressed within the guidelines structure, the guidelines will become increasingly irrelevant to the real world situations faced by sentencing judges and their force and influence will continue to wane. In the long run, that’s a far more serious concern.

The Proposed Amendments

There are five key proposed amendments affecting § 2B1.1:

Inflation Adjustment: The guidelines operate by calculating an offense level, ranging from 1 to 43, for every criminal conviction. That offense level, along with the defendant’s criminal history, is used in the Sentencing Table to find the recommended sentencing range. The greater the offense level, the greater the sentence.

In any fraud case the single biggest driver of the offense level is the loss caused by the defendant’s conduct. That amount is plugged into the “fraud table” in § 2B1.1(b)(1) and the corresponding number of offense level points is added; if the loss exceeds $5,000.00 two points are added, if it exceeds $10,000.00 four points, and so on. At the top of the fraud table, for losses exceeding $400 million, thirty points are added to the offense level.

Since the fraud table was last revised in 2001, inflation has led to “bracket creep.” A crime that caused $100,000.00 in damage in 2001 would cause more than $130,000.00 in damage today, simply due to the effect of inflation. That would place it higher in the fraud table and result in a greater sentence for the identical crime. The Commission proposes using a multiplier based on the Consumer Price Index to adjust the amounts in the table that define the different levels and bring them back into line with what they represented in 2001.

Because of its potential to affect almost any case under § 2B1.1, the inflation adjustment is probably the most far-reaching of the proposed amendments. According to the Commission’s research, this modification of the fraud table alone will reduce the average sentence under § 2B1.1 by more than 20%.

Victims Table: The “victims table” in § 2B1.1(b)(2) provides a two level enhancement if the offense affected ten or more victims, four levels if it affected fifty or more victims, and six levels if it affected 250 or more. This focus on the number of victims may lead to inappropriate results in some cases.

For example, suppose a defendant in a penny stock scheme defrauds 500,000 people of only $5.00 each. Another defendant in a different kind of scheme defrauds five people of $500,000.00 each, devastating them and wiping out their life savings. The total loss is identical and so each would receive the same increase from the fraud table. But the first would receive an additional six levels based on the number of victims while the second, who caused far greater harm to his individual victims, would not. Six additional levels may lead to a doubling or more of the recommended sentence, but probably few would argue that the first defendant deserves twice the jail time of the second.

The proposed amendment shifts the focus away from the number of victims and toward the seriousness of the harm caused. The two level enhancement remains for offenses that affect 10 or more victims, and also applies if the offense resulted in substantial financial hardship (such as bankruptcy or loss of retirement savings) to one or more victims. If the offense resulted in substantial financial hardship to five or more victims there is a four level increase, and substantial financial hardship to 25 or more victims results in a six level increase. The four and six level increases based solely on the number of victims are eliminated.

Sophisticated Means: The fraud guideline contains a two level enhancement for offenses that involved multiple jurisdictions, overseas bank accounts, shell corporations, or other types of “sophisticated means.” § 2B1.1(b)(10). Some have criticized this provision for failing to distinguish offenses that truly were unusually sophisticated from those that merely employ devices that have now become commonplace. There was also some concern about courts applying this enhancement to a defendant when it was other participants in the scheme, and not that defendant, who actually employed the sophisticated means in question.

The Commission ultimately addressed only the latter concern. It proposes modifying the enhancement to specify that it should apply only if the particular defendant “intentionally engaged in or caused the conduct constituting sophisticated means.” The rest of the enhancement remains intact.

Intended Loss: The loss caused by the defendant’s behavior that is plugged into the fraud table is the greater of either the actual or the intended loss. There has been some dispute in the courts over whether “intended loss” means all losses that are reasonably foreseeable based on the nature of the defendant’s conduct (objective standard) or only those losses that the defendant actually and purposely sought to inflict (subjective standard). The Commission proposes revising the commentary in the fraud guideline to adopt the subjective standard and define intended loss as the loss the defendant “purposely sought to inflict.”

Fraud on the Market: “Fraud on the market” cases are those that involve the fraudulent increase or decrease in the price of a publicly traded security. The amendment removes language in the commentary providing a rebuttable presumption that the proper way to calculate loss is the change in the actual value of the security, which can be extremely difficult to calculate and lead to astronomical sentences. The new commentary states that a court may use any method for calculating loss that is appropriate and practical under the circumstances.

DOJ building: the Department opposed several of the amendments to the fraud sentencing guideline

DOJ’s Opposition: Fighting the Wrong Guidelines Battles

At the March 12 hearing the Department of Justice opposed the inflation adjustment; opposed the amendments concerning sophisticated means, intended loss, and fraud on the market; and supported the new enhancement based on causing victims substantial hardship. In other words, DOJ opposed virtually any amendment that could lead to lower sentences while supporting changes that could lead to higher ones.

DOJ was a lonely voice at the hearing and is definitely swimming against the tide by opposing these amendments. There is a widespread and growing belief that guidelines sentences in major fraud cases have become excessive. More broadly, there is an emerging bipartisan movement in the country favoring criminal justice reform, including measures to reduce skyrocketing sentences (particularly for non-violent offenders) and our enormous prison population. There aren’t many causes today on which conservatives like Senator Ted Cruz and liberals like Senator Dick Durbin can find common ground, but this appears to be one of them.

Law professor Frank Bowman provided some compelling hearing testimony tracing the history of the fraud guideline and demonstrating how various forces, both intentional and unintentional, have combined over the years to escalate the sentences in such cases dramatically. As he pointed out, given the large dollar values involved in some recent Wall Street frauds and the multiple enhancements that have been added over the years to § 2B1.1, it’s relatively easy for a white-collar defendant to rocket to the top of the sentencing table and end up with a recommended sentence of 30 years or even life in prison – on a par with sentences recommended for homicide, treason, or a major armed bank robbery.

It’s hard to see what criminal justice purpose is being served by the ever-escalating sentences in fraud cases. The prospect of prison does have a powerful and important deterrent effect that is unique to criminal law. But for a typical business executive it’s hard to believe there’s much additional marginal deterrent value in a possible twenty or twenty-five year sentence as opposed to, say, a ten or fifteen year one.

By appearing almost reflexively to oppose any amendment that might lead to lower sentences, DOJ risks losing credibility and being marginalized in the broader conversation about sentencing reform. For example, the inflation adjustment should have been relatively uncontroversial. There’s no logical reason why a crime today should result in a higher sentence than the identical crime fifteen years ago, simply due to the increased cost of living. Yet DOJ opposed the adjustment, largely because it would result in lower sentences – which is, of course, precisely the point. Members of the Commission pronounced themselves “baffled” by DOJ’s position.

What’s more, legal developments have rendered DOJ’s position in favor of higher guidelines sentences increasingly beside the point. It’s been ten years since the Supreme Court ruled in United States v. Booker that the mandatory sentencing guidelines were unconstitutional and the guidelines must be advisory only. Later in Kimbrough v. United States the Court made it clear that a judge is free to depart from the recommended sentence if the judge disagrees with a policy decision underlying the guidelines.

In this legal environment, DOJ’s continued push for higher guidelines looks like a struggle to keep the barn door closed when the horse left for greener pastures long ago. In the post-Booker/Kimbrough world, if judges believe a sentence called for by the guidelines is out of whack they will simply reduce it. For example, in the recent public corruption case involving former Virginia Governor Robert McDonnell, the judge called the recommended guidelines sentence of six to eight years in prison “ridiculous” and proceeded to sentence McDonnell to only two years.

There’s substantial evidence that the same thing is already happening in fraud cases. According to the Sentencing Commission’s data, judges sentence below the recommended guidelines range in about 28% of fraud cases (not counting those cases where the government itself requests a reduced sentence). But in the Southern District of New York, home to Wall Street and many of the big-dollar fraud cases, judges depart below the guidelines in more than 48% of such cases. It does little good for DOJ to push for extremely high guidelines numbers only to have judges ignore the guidelines and impose the lower sentences that they feel are just and reasonable.

But DOJ’s advocacy for higher guidelines sentences in fraud cases is worse than futile: it’s counter-productive. The more that judges come to regard the guidelines as calling for inappropriate sentences, the more comfortable they may become not following them. This could lead to even more widespread departures from the guidelines not merely in fraud cases but in cases across the board, accelerating a decline in the force and influence of the guidelines that so far has been held relatively in check since Booker.

Guidelines amendments that may result in lower sentences do not tie a prosecutor’s hands. In truly extraordinary cases the government is always free to ask the judge to exceed the recommended guidelines sentence and depart upward. This would achieve DOJ’s goal of seeking enhanced punishment in cases where it may truly be called for while still promoting greater overall respect for the guidelines in the average case. DOJ would be better served by adopting such an approach while still supporting modest sentencing reforms for the majority of cases.

The Future of the Fraud Guideline

The proposed amendments reflect some tinkering around the edges of § 2B1.1, but won’t have a dramatic impact on many of the very large fraud cases that have prompted the most concern. None of the Commissioners appeared completely satisfied with the amendments, and it appears the Commission considers them to be merely a first step. The April 9 meeting where the amendments were formally adopted revealed some disagreement, though, over whether the best course in the future would be additional tweaks to the fraud guideline or a more fundamental reform of the entire sentencing system. The latter could include working with Congress to implement a completely new system of mandatory guidelines that would satisfy Booker.

Following the post-Enron hysteria of more than a decade ago, the pendulum swung too far in the direction of severe sentences for fraud cases. The proposed amendments represent a welcome step back towards sentencing sanity, but more remains to be done if the guidelines are not to become virtually irrelevant in major fraud prosecutions.

Click here to sign up for the Sidebars mailing list and be notified of future posts.