Fraud and Deceit

White-collar crimes “involving fraud and deceit” generally fall under the Sentencing Guidelines Manual section 2B1.1. Those offenses begin with a base offense level of 6 or 7, which then can be increased by a host of what are called “specific offense characteristics.” See U.S. Sentencing Guidelines Manual § 2B1.1 (2014). Each characteristic is assigned a numeric value that, when added together, provides the “offense level.” Matching the offense level with the defendant’s “criminal history category” gives the guideline sentencing range. See id. § 1B1.1; ch. 5, pt. A.

The most significant offense characteristic for white-collar defendants may be the amount of loss or, in cases like insider trading where loss cannot be determined, the defendant’s gain. See id. §§ 2B1.1(b); 2B1.4(b)(1). Beginning with an additional two levels for a loss exceeding $5,000, the calculation ratchets up quickly, adding, for example, 10 levels for a loss of over $120,000, 16 levels for a loss of over $1 million, 22 levels for a loss of over $20 million, and more for larger amounts, topping out at an additional 30 levels. See id. § 2B1.1(b)(1). A fraudster with no prior criminal history and no other offense characteristic whose victims lost $2 million would face a guideline calculation of 46–57 months, even if the fraudster thought (perhaps unreasonably) that his or her victims would recover their money. That same defendant would be looking at 0–6 months if his or her scheme was uncovered before the victims lost anything. No other offense characteristic comes close to having the same impact on the guideline calculation.

The broad definition of “loss” puts this 800-pound gorilla on steroids. Loss is defined as the greater of actual or intended loss, where actual loss is the “reasonably foreseeable pecuniary harm that resulted from the offense.” Id. § 2B1.1 cmt. n. 3(A), (C). “Gain” is specially defined for insider trading offenses and is arguably not limited to “reasonably foreseeable” gain. Compare United States v. Kluger, 722 F.3d 549, at *7 (3d Cir. 2013) (foreseeability not required), with United States v. Royer, 549 F.3d 886, 904 (2d Cir. 2008) (foreseeability required). Trading by downstream tippees—especially when those tippees are trading for large institutions—can significantly increase the amount of gain attributed to a tipper, even if the tipper made little or nothing on the deal. See, e.g., United States v. Santarlas, No. 1:09-cr-01170 (S.D.N.Y. Dec. 10, 2009) (non-trading defendant received $32,500 for his tip, but adding the tippee’s gain increased his sentencing range from 0–6 to 30–37 months); United States v. Martoma, No. 1:12-cr-00973 (S.D.N.Y. Sept. 8, 2014) (gain attributed to SAC portfolio manager increased guideline range from 0–6 to 188–235 months).

Sentencing enhancements driven by the amount of loss or gain have increased dramatically in recent years for white-collar defendants. The percentage of such defendants for whom the amount of loss or gain added 10 levels—which raises the guideline calculation by roughly 300 percent—or more has risen almost every year since 2006. In the most recent year for which data are available (2013), 45.6 percent of fraud defendants—and 88.6 percent of insider trading defendants—fell into this category. See U.S. Sentencing Comm’n, Use of Guidelines and Specific Offense Characteristics (2006–2013). A recent study by Reuters found that insider trading sentences have increased by 31.8 percent over the past five years and that this “rise is at least partly driven by the bigger profits being earned through the illegal schemes[.]” Nate Raymond, Insider Traders in U.S. Face Longer Prison Terms, Reuters Analysis Shows, Reuters (Sept. 2, 2014), www.reuters.com/article/2014/09/02/us-insidertrading-prison-insight-idUSKBN0GX0A820140902.

Some argue that high guideline calculations, significantly driven by the enhancement for loss, unfairly coerce defendants with meritorious defenses from proceeding to trial. According to the former special counsel to the U.S. Sentencing Commission and chair of the commission’s Practitioners Advisory Group, the “draconian guideline range driven by an aggressive loss calculation” allows the government to “more readily extract guilty pleas in exchange for negotiated charges and facts.” David Debold and Matthew Benjamin, “Losing Ground”—In Search of a Remedy for the Overemphasis on Loss and Other Culpability Factors in the Sentencing Guidelines for Fraud and Theft, 160 Penn L. Rev. 141, 153 (2011). In high-gain insider trading cases such as Galleon, “the government may bargain away millions of dollars in loss by stipulating to specific calculations—again, in exchange for guilty pleas.”

Prior to 2005, courts were largely bound to sentence defendants within the narrow range provided by the guidelines. See 18 U.S.C. § 3553(b)(1). But in United States v. Booker, 543 U.S. 220 (2005), the Supreme Court held that this mandatory sentencing regime violated the Sixth Amendment. Now judges need only “consider” the guideline calculation, along with the sentencing factors prescribed in 18 U.S.C. § 3553(a).

At first glance, fraud defendants don’t appear to have benefited from Booker. Their average length of imprisonment has actually increased since 2005, while the average term of imprisonment for all defendants fell from 49.9 months in 2005 to 45 months in 2013. For fraud defendants, that figure increased during that same period from 16.6 months to 26 months. See U.S. Sentencing Comm’n, Sourcebooks of Federal Sentencing Statistics (2005, 2013). But because sentences are driven largely by the facts, a better way to gauge Booker’s impact may be to examine the extent to which courts have imposed non-guideline sentences. Under that analysis, white-collar defendants under the Booker regime have fared relatively well, receiving, on average, more frequent and larger discounts from the guidelines than other defendants.

In the two years preceding Booker, the average fraud defendant was about as likely to receive a below-guideline sentence as any other non-cooperating defendant. While the figure jumped for both populations after Booker—and has continued to rise steadily since—fraud defendants post-Booker appear to receive non-government-sponsored, below-guideline sentences significantly more often than their non-fraud counterparts. Almost 26 percent of fraud defendants sentenced in 2013 received such sentences, up from 5.5 percent in 2004. For all defendants, those figures were 18.7 percent in 2013 and 5 percent in 2004. See U.S. Sentencing Comm’n, Sourcebook of Federal Sentencing Statistics (2004–2013). Insider trading defendants have fared even better, averaging 40.3 percent a year since 2006. See U.S. Sentencing Comm’n, Quarterly Sentencing Updates (2006–2014). And not only are below-guideline sentences for fraud defendants more frequent, they are also more substantial. Between 2005 and 2013, fraud defendants received a sentence that was, on average, 62.9 percent lower than the low end of the guidelines, while for non-fraud defendants that figure was only 35.4 percent.

The Galleon Group was one of the largest hedge fund management firms in the world, at one time managing over $7 billion in assets. According to prosecutors, Galleon and its founder, Raj Rajaratnam, sat at the hub of a broad informational network that included hedge fund managers, traders, lawyers, bankers, and public company insiders of all ranks. Some of the participants tipped information, some traded on it, and some did both. Over two dozen individuals were criminally charged with participating in some aspect of this network.

Seven defendants went to trial. Seven defendants pleaded guilty without the benefit of a cooperation agreement. Fourteen others agreed to cooperate with the government as part of their plea deal. To date, 25 defendants have been sentenced, with punishment ranging from probation to custodial sentences of anywhere between 6 months and 11 years. For these defendants, what made a difference in their sentencing?

A certain conventional wisdom says that defendants who unsuccessfully take the government to trial will, all other things begin equal, receive a substantially harsher sentence than had they pleaded guilty—a belief that can operate as a powerful disincentive to fight. Was this the case in Galleon? Were the defendants who lost at trial inordinately penalized by their decision to put the government to its proof?

Unsuccessful Defendants

The unsuccessful Galleon trial defendants, who on the whole were not minor players, got sentences ranging from 24 to 132 months, with reductions of between 1 percent and 70 percent (with a median of 35 percent) from their guideline calculation. Relative to the guidelines, the seven defendants who pleaded guilty without a cooperation agreement fared worse. Whereas four of the six defendants convicted at trial received a sentence substantially lower (30–70 percent) than the guideline calculation, the seven pleading, non-cooperating defendants received an average discount of 9 percent, with none getting more than 25 percent and most receiving a discount of 10 percent or less.

The seven non-cooperating defendants’ guideline calculations generally reflect a three-level reduction for acceptance of responsibility, so arguably their “discount” was already built into the calculation. See U.S. Sentencing Guidelines Manual § 3E1.1 (2014). But even adjusting the comparison to account for these three levels, the reduction for these defendants is still within the range of—and sometimes less than—that for the four trial defendants who received a substantial discount.

More interesting, the sentences for five of the defendants convicted at trial were within or below the guideline range that would have applied had they pleaded guilty and received a three-level reduction for acceptance of responsibility. And the Galleon defendants are not alone. Defendants in other recent high-profile insider trading cases who also lost at trial still received a sentence substantially less than the low end of their guideline range. See, e.g., United States v. Jiau, No. 1:11-cr-00161 (S.D.N.Y. Dec. 14, 2011) (97–121 months under the guidelines; 48-month sentence); United States v. Fleishman, No. 1:11-cr-00032 (S.D.N.Y. Dec. 21, 2011) (87–108 months under the guidelines; 30-month sentence); United States v. Chiasson, No. 1:12-cr-00121 (S.D.N.Y. Feb. 7, 2012) (97–121 months under the guidelines; 78-month sentence).

One possible explanation for the relatively paltry discounts received by the non-cooperating defendants may be that, in plea negotiations, they got an additional benefit not apparent from their guideline calculation. For several of those defendants, the gain figure used to calculate their guidelines was substantially less than what they were charged with. Danielle Chiesi, for example, was charged with a $4 million gain, but at sentencing the government pegged her gain at only $1.6 million, resulting in a two-level savings. The plea agreement for Robert Moffat, the alleged source of several tips that resulted in Chiesi’s $4 million gain, provided that the information he gave did not result in any profits, and so he escaped even the $1.6 million gain on which Chiesi was sentenced. Particularly where a pleading defendant elects not to cooperate, limiting the loss or gain as part of plea negotiations can make a substantial difference at sentencing.

Still, it is worth considering whether, in forgoing trial, these non-cooperating defendants did any better than if they had tried to beat the charges at trial, where at least they would have given themselves a chance to walk away entirely as Rengan Rajaratnam, Raj’s brother, did. The analysis is complicated by such a small population as well as all the factors a judge must consider at sentencing. See 18 U.S.C. § 3553(a). Even so, these cases suggest that a defendant who goes to trial in a high-profile, high-loss/gain case is not foreclosed from a sentence substantially below the guidelines. Indeed, Raj Rajaratnam lost at trial and received a 44 percent discount from his minimum guideline sentence—a discount greater than that of any of the non-cooperating defendants who pleaded guilty.

How did gain affect the sentences doled out in the Galleon cases? It was certainly significant in calculating the guideline range. For all but one of the defendants for whom complete sentencing information is available, the amount of gain was responsible for between 58 percent and 79 percent of their total offense level, with most falling into the 70–79 percent range. But while gain was helpful in predicting guideline calculations, it was much less so in predicting sentences, a relationship that is especially apparent when examining those defendants with less than outsized ($10 million) gains.

Galleon Defendants: Gain (<$10mm) Relative to Minimum Guidelines Calculation

Galleon Defendants: Gain (<$10mm) Relative to Sentence Imposed

So gain seemed to have substantially less effect where it mattered: on the sentence imposed. As Judge Rakoff saw it in sentencing one Galleon tipper, while the guidelines are driven by “unpredictable monetary gains,” “[t]he heart of Mr. Gupta’s offenses . . . is his egregious breach of trust. . . . Yet the Guidelines assess his punishment almost exclusively on the basis of how much money [his tippee] gained by trading on the information.” United States v. Gupta, 904 F. Supp. 2d 349, 350–52 (S.D.N.Y. 2012). The defendant, who did not trade and received no direct compensation for his tips, received a sentence of 24 months, a significant departure from his 78–97 months guideline calculation.

Even a direct personal gain may not carry as much weight in the actual sentence. Doug Whitman, a hedge fund manager who, unlike Gupta, did directly benefit by trading, also received a significant discount. Whitman’s gain did not include “unpredictable” amounts made by others, and he did not breach any fiduciary duty. In other words, Whitman possessed the element of personal gain that Judge Rakoff found lacking in Gupta, but not the breach of a personal fiduciary duty that the court found so culpable. Nevertheless, Judge Rakoff departed significantly from Whitman’s 51–63 months guideline calculation and sentenced him after trial to 24 months, the same as Gupta.

Regardless, loss/gain remains an important factor. No Galleon defendant with a gain of over $5 million avoided prison, and the three defendants who received the longest sentences had the first-, third-, and fourth-largest gains. And as long as the guidelines place so much weight on loss, that element will continue to have at least an in terrorem impact on defendants. That said, six Galleon defendants with a gain in excess of $1 million received no prison sentence, while six defendants with less than $1 million in gain received significant prison time. Two other factors seemed to play a more determinative role for the Galleon defendants: cooperation and a law degree.

Judges may depart downward from the guidelines when the defendant provides “substantial assistance.” U.S. Sentencing Guidelines Manual § 5K1.1 (2014). Based on the sentences imposed for the cooperating Galleon defendants, perhaps the clearest lesson learned is that cooperation pays. The defendants who pleaded guilty and did not cooperate were each sentenced to between 18 and 66 months in prison. On the other hand, each of the 12 cooperators sentenced to date, some of whom played significant roles or received substantial benefits from the scheme or both, received little or no prison time, with 10 avoiding prison altogether. According to available information, those 10 had an average guideline calculation of 37–46 months, even after deducting three levels for acceptance of responsibility.

But those benefits aren’t free. Cooperation can be long and taxing, and the difference between good cooperation and bad cooperation may mean the difference between probation and a custodial sentence. The guidelines list a number of nonexclusive factors for the court to consider in evaluating a defendant’s cooperation. Id. § 5K1.1(a) (2014). Of these, the timeliness and quality of cooperation seemed to be the most significant.

As in many cases, early cooperation is particularly valuable to the government. In the beginning, prosecutors are often in the dark about basic facts. Hence, early cooperation saves the government time and resources, and it allows the government to pressure other parties to cooperate sooner. Providing information early on also lowers the odds another cooperator will get there first with the inside scoop. Getting there early also gives the cooperator more time to provide more assistance (and more value) before sentencing. Recording conversations with co-conspirators was particularly helpful in the Galleon cases, not only for the evidence itself but also because the government used those conversations to obtain wiretaps, which led to further incriminating evidence. Finally, early cooperation is sometimes perceived as a token of genuine remorse. As the government said at the sentencing of one Galleon cooperator,

[t]he speedy cooperation of Kumar shortly after his arrest was highly significant for the Government . . . . Kumar did not hesitate . . . . Kumar did not weigh his options . . . . Instead, Kumar did what almost all individuals involved in Rajaratnam’s insider trading schemes notably did not do—Kumar decided to cooperate immediately with the Government.

Sentencing Memorandum by U.S. Attorney as to Anil Kumar at 11, United States v. Kumar, No. 1:10-cr-00013-DC (S.D.N.Y. July 16, 2012) (emphasis in original).

If you are going to be a domino, it pays to be the first to fall.

Quality of Cooperation

Not surprisingly, the novelty, scope, and usefulness of the information are important in determining the quality of cooperation. Facts not previously known to the government usually add significant value. One Galleon defendant, for example, provided information that led to the discovery of a separate insider trading network. Another potentially subjected himself to more trouble by disclosing conduct without knowing whether the government already knew about it. And overcoming impediments to cooperation—whether it be personal safety or, as in the case of one Galleon defendant, close family and professional ties to a government target—also scores points.

Cooperation that goes beyond a government interview can be especially valuable. Recording one conversation can be of considerable benefit, especially in insider trading cases. Doing so multiple times is even better, and some of the Galleon cooperators recorded hundreds of conversations. Trial testimony, especially when it’s effective, may be the gold standard in cooperation. According to the government, one Galleon cooperator provided extensive and “devastating” testimony against two defendants, and another cooperator’s testimony about the use of phony instant messages “neutralized” the key defense at the trial of several other defendants.

The only two cooperating defendants who did not receive probation were Roomy Khan and Brien Santarlas. Both seem to have provided substantial cooperation consistent with that provided by other defendants who received probation. Both cooperated early, recorded conversations, and testified at trial. But while the court and the government acknowledged that they had provided substantial cooperation, and while they received significant sentencing benefits, they still went to prison. How come?

For Khan, the answer is easy. Although her cooperation was substantial, it was also horribly flawed. Between the time she agreed to cooperate and the date she was sentenced, she lied to federal investigators, destroyed evidence, tipped off co-conspirators, and committed perjury in an unrelated case. As Judge Rakoff said in sentencing her, “[t]he message from this court is that you cannot have it both ways . . . to cooperate and then obstruct justice.” Patricia Hurtado & Bob Van Voris, Khan Gets One Year in Prison in Insider Trading Case, Bloomberg (Feb. 1, 2013) www.bloomberg.com/news/articles/2013-01-31/roomy-khan-gets-one-year-in-prison-in-insider-trading-case-1-.

At the same time, Khan may also be the best example of the benefits of cooperation. She had a prior wire fraud conviction for passing inside information. She was a key player in the Galleon conspiracy, receiving and providing tips and personally benefiting by at least $1.5 million, with her tippees realizing another $25 million. And she had serious, multiple problems with her cooperation. Yet the government still filed a 5K1.1 motion and she received a 12-month sentence, far below her 97–121 months guideline range.

Santarlas’s case was different. As a lawyer, he faced the consequences of having breached not only fiduciary duties but also the professional responsibilities unique to lawyers. While there were fiduciary breaches aplenty in the Galleon cases, not all of the fiduciary defendants fared poorly by comparison with their non-fiduciary counterparts. Several cooperating senior executive tippers were sentenced to probation despite guideline calculations that called for three years’ imprisonment or more. One non-cooperating tipper fiduciary received a prison term of 18 months, a 25 percent reduction from his guideline calculation. And even Gupta, the only fiduciary to go to trial, received a 24-month sentence, 70 percent off his guideline calculation and substantially more discounted than that of any other trial defendant.

Lawyer Defendants

But this pattern did not hold true for the lawyer defendants. Of the seven defendants who pleaded guilty but did not cooperate, the two lawyer defendants fared worst. Arthur Cutillo received only $32,500 for tipping his client’s confidential information but was sentenced to 30 months in prison, which represented no discount from his guideline calculation. Judge Sullivan scolded him for failing “to respect and protect the confidences of [his] clients” and emphasized the need to deter fiduciaries from violating their clients’ trust. Brian Baxter, Ex-Ropes & Gray Lawyer Gets 30 Months in Galleon Insider Trading Case, Am Law Daily (July 1, 2011) http://amlawdaily.typepad.com/amlawdaily/2011/07/cutillo-sentence.html. Jason Goldfarb fared little better, his 36-month sentence amounting to just a 3 percent discount from the guidelines.

Santarlas received only $32,500 for his tips, cooperated immediately, wore a wire, and testified at two trials. Yet, he still got six months in prison. While Judge Sullivan, who also sentenced fellow lawyers Cutillo and Goldfarb, appears to be one of the tougher sentencing judges, he still sentenced four other cooperating defendants—three with the same or higher guideline calculations than Santarlas’s—to probation. It is not difficult to appreciate that lawyer defendants may draw harsher scrutiny by some judges at sentencing. In the Galleon cases, every one of them went to prison. Perhaps it is no accident that a lawyer received the longest insider trading sentence in U.S. history (12 years). See Kluger, 722 F.3d at 568 (“[W]e cannot overlook the circumstance that Kluger served as the source of the information that permitted the scheme to function. Furthermore, Kluger was an attorney who took an oath to uphold the law.”).

Still, non-lawyers should not rest too easily. This same bias may weigh against other gatekeepers and professionals, such as accountants, or individuals who breach a duty owed to a public or quasi-public entity. While very few insider trading cases have resulted in sentences above the minimum guideline calculation, several that did so involved the abuse of positions of public trust. In both United States v. Liang and United States v. Johnson, the courts emphasized the defendant’s abuse of that trust. See Liang, No. 8:11-cr-00530 (D. Md. Mar. 5, 2012) (FDA chemist sentenced to 60 months on low-end calculation of 57 months); Johnson, No. 1:11-cr-00254 (E.D. Va. Aug. 12, 2011) (NASDAQ managing director sentenced to 42 months on low-end calculation of 37 months).

Judges have considerable discretion in sentencing, and their decisions are subject to limited appellate review. All that discretion means that the luck of the judicial draw can have a major impact at sentencing. While some judges look more critically at the role of loss in sentencing, others adhere closely to the guidelines. As one court observed, “sentences in high-loss cases will remain wildly divergent as some district judges apply the loss guideline unquestioningly while others essentially ignore it.” United States v. Corsey, 723 F.3d 366, 378 (2d Cir. 2013) (Underhill, J., concurring).

Sentencing Disparities

The extent to which courts impose below-guideline sentences without government sponsorship varies from district to district. Nationally, about 18.7 percent of all defendants and 25.9 percent of fraud defendants received such sentences in 2013. The district courts in Brooklyn, Chicago, Manhattan, and San Francisco, all of which handle a substantial number of white-collar cases, handed down these sentences even more frequently. See U.S. Sentencing Comm’n, Federal Sentencing Statistics by District, Circuit & State for Fiscal Year 2013. Even within judicial districts, views of an appropriate sentence can vary from judge to judge. One study concluded that “the typical sentence handed down by a federal district court judge can be very different than the typical sentences handed down for similar cases by other judges in that same district.” Surprising Judge-to-Judge Variations Documented in Federal Sentencing, TRACReports (Mar. 5, 2012), http://trac.syr.edu/tracreports/judge/274/. Intra-district sentencing disparities for white-collar defendants were even greater than for other types of defendants. Judges develop reputations as strict or lenient sentencers for a reason.

One could point to the Galleon cases as evidence of this disparity. The five judges who sentenced the non-cooperating defendants differed significantly in the extent to which they departed from the guideline calculation, with average deviations of 0 percent, 7 percent, 10 percent, 39 percent, and 63 percent. In fairness, this sample size is too small to reach a conclusion about any particular judge, especially given the variety of individualized factors a court is required to consider. See 18 U.S.C. § 3553(a). But these figures illustrate the broad discretion judges have. At the same time, in light of the very substantial discounts given to the defendants who cooperated, it is probably safe to say that the judges shared a similar view of the value of cooperation.

Clearly, it is important to know the judge’s history and philosophy on sentencing. Does the judge generally adhere to the guideline range? Is the judge moved by any factors in particular, e.g., a history of community involvement or charitable giving? Does the judge tend to be tougher on white-collar defendants? And remember that discretion is a two-way street. While it is rare for a defendant to be sentenced above the guideline range, and none of the Galleon defendants was, it is certainly within a judge’s power to do so: In 2013, 2.2 percent of both all defendants and all fraud defendants nationwide received above-guideline sentences. See U.S. Sentencing Comm’n, Sourcebook of Federal Sentencing Statistics (2013).

Lessons to Learn

Can we learn anything from the Galleon saga, or is each case so dependent on its own facts and the idiosyncrasies of the players involved that a review of these cases may be interesting but not particularly helpful? The defendants’ varied circumstances caution against sweeping generalizations, but at least three points are worth making.

First, the guidelines—and particularly loss or gain—should be the starting point in handicapping a potential sentence. But be careful not to overstate their impact. Only two of the Galleon defendants were sentenced within the guideline range, and those who were sentenced below that range—often substantially so—included the most significant defendants, including some who went to trial. Appreciating whether a likely sentence could be below, even well below, the guideline range could make some more willing to roll the dice and go to trial.

Second, cooperation may be the most significant sentencing factor of all. It appears to have had the most consistent and dramatic impact of any factor on the Galleon defendants’ sentences. Just how much of an impact will depend on when and how cooperation occurs, which makes an early assessment of its pros and cons that much more important. And there is much you can do to help your client be a star cooperator:

• Ensure that, from the beginning, your client is prepared in detail for each of his or her government debriefings. Providing more information sooner is more valuable, and a consistent story makes for a more useful witness.
• You and your client may know better than the government about ways your client can cooperate. Look for opportunities to help prosecutors understand the business, the transactions, the documents, and the people involved. Does your client have information about other persons or matters the government would be interested in, even if the government is unaware of them and even if the information is unrelated to the current matter?
• Help your client appreciate that cooperation is a job that needs to be done well and consistently from the start. Slipping up at any point, even just a little, can reduce a cooperator’s value significantly and irretrievably. Conduct having nothing to do with the case—a DUI, domestic violence, or, heaven forbid, perjury in another matter—can seriously diminish a cooperator’s value.
• Stay involved with your client throughout his or her cooperation. Because cooperators are usually sentenced after all of the other defendants’ cases have been resolved, it can be years before your client is sentenced—plenty of time to mess up.
• Keep track of your client’s cooperation as well as your own efforts that may assist the government. Count the hours, meetings, telephone calls, and pages of documents reviewed. Because cooperation may unfold over years, today’s prosecutor may be gone by the time sentencing rolls around. You’ll want to tell the judge what a great cooperator your client was, and the numbers help back it up.

Finally, the time to start thinking about sentencing is not at the end of the case but at the beginning. Predicting a client’s sentence is difficult enough and will be especially so before you know all the facts, including the identity of your judge. But that does not make early strategizing any less important. All cases may be different, but the Galleon prosecutions offer useful insight into what will matter when sentencing rolls around for a white-collar defendant.

Read a judge's comments about the difficulty of devising appropriate sentences.