'Booker' After a Year: New Highs for Sentences, Guidelines Followed

New York Law Journal

March 6, 2006
by William Sloane and

In 2004, Jamie Olis, a former mid-level executive at Dynegy Corp., was sentenced to over 24 years in prison after a federal jury convicted him of accounting fraud. His lengthy sentence was mainly based on the court's finding that he helped cause $105 million in losses to shareholders and that the U.S. Sentencing Guidelines at the time mandated the sentencing range because of those losses.

Mr. Olis's two codefendants pleaded guilty to charges where they faced a maximum of five years in prison, and were sentenced last week to 15 months and one month.

Judge Sim Lake of the U.S. District Court for the Southern District of Texas sentenced Mr. Olis to the minimum within the then-required sentencing range for the U.S. Sentencing Guidelines. At Mr. Olis's sentencing, Judge Lake said he was "required to follow . . . the Federal Sentencing Guidelines," that he took "no pleasure in sentencing [Mr. Olis] to 292 months," but that it was the court's job "to follow the law."[1] Newspapers described the sentence as "eye popping" and "jaw-dropping," and one commentator called Mr. Olis "the poster child for what some perceived as excessive punishment of white-collar crime."[2]

Several months later, the Supreme Court decided the case of United States v. Booker,[3] ruling that the Sentencing Guidelines were unconstitutional. Defense attorneys, and the defendants they represent, celebrated. They predicted that a new, more-lenient era had begun where judges would reduce sentences in the extreme cases like Mr. Olis's, as well as the run-of-the-mill cases. The hope was that defendants would routinely receive sentences significantly below the hefty range designated by the Sentencing Guidelines.

Mr. Olis himself may benefit from the Supreme Court's decision. On Oct. 31, 2005, the U.S. Court of Appeals for the Fifth Circuit ruled that Judge Lake overstated the shareholders' losses attributed to him, and remanded for re-sentencing. Based on the Fifth Circuit decision, the government is now seeking a sentence of just over 15 years in prison, but Judge Lake has greater freedom to sentence Mr. Olis outside of that range. His sentencing was recently postponed and is now expected to occur this summer.

However, if Judge Lake follows the national trend, he will adhere largely to the Sentencing Guidelines. Just over one year after the Supreme Court issued its decision, celebrations among defense attorneys and defendants alike are much more muted since Booker has, instead, led to surprisingly limited changes in sentencing practices, and average and median sentences nationwide are at historic highs, the exact opposite of what was expected.

U.S. Sentencing Guidelines

The Sentencing Guidelines began in 1984 when Congress passed the Sentencing Reform Act (SRA) to promote uniformity and to help sentences better reflect "the seriousness of the offense."[4] The guidelines were formally implemented in 1987 to achieve this goal through a point system based on the defendant's conduct, criminal history, and mitigating factors such as acceptance of responsibility. In order to calculate the defendant's points, judges made findings of fact correlating to each factor. The totaled points corresponded to a narrow sentence range within which the judge was required to comply when issuing a sentence, unless specific and narrowly defined exceptions applied to allow for a reduction in the sentences such as the defendant's substantial assistance to the government.

The guidelines had no shortage of critics. One main criticism was that the guidelines imposed harsh sentences without due consideration to individual factors.

In 2003, Judge John S. Martin wrote that he was returning to private practice rather than continuing as a judge with senior status because of the distress "at being part of a sentencing system that is unnecessarily cruel and rigid."[5] "Every sentence imposed affects a human life and, in most cases, the lives of several innocent family members who suffer as a result of a defendant's incarceration," he wrote. "For a judge to be deprived of the ability to consider all of the factors that go into formulating a just sentence is completely at odds with the sentencing philosophy that has been a hallmark of the American system of justice."

Nonetheless, for 17 years, federal judges followed the guidelines, and with significant and broad-based results. In a report issued in November 2004, the U.S. Sentencing Commission found that the average prison sentence doubled and the percentage of offenders receiving only probation was cut to a third when compared to pre-guidelines statistics.[6] The commission also found that the guidelines had "reduce[d the]unwarranted disparity arising from differences among judges"[7] and that regional disparities also appeared to have been reduced for most crimes, although the data was not conclusive.

'United States v. Booker'

In January 2005, the Supreme Court changed the entire landscape of federal sentencing with its ruling in United States v. Booker. In a pair of majority opinions — one by Justice Paul Stevens and one by Justice Stephen Breyer — the Court found that the U.S. Sentencing Guidelines were unconstitutional because they mandated sentences based on facts neither proved to a jury beyond a reasonable doubt nor admitted by the defendant.

In Justice Breyer's majority opinion, the Court ruled that district courts must still consider the guidelines in delivering the sentences, but that the guidelines were only "advisory." Therefore, courts are now required "to consider Guidelines ranges," but are permitted "to tailor the sentence in light of other statutory concerns as well."[8]

Reaction to 'Booker'

Now judges must only consult the guidelines, but are not required to sentence defendants within the prescribed range. Unleashed for the first time in nearly two decades from the mandatory guidelines which many vehemently opposed, federal judges have so far reacted in a very surprising way — to the general disappointment of the criminal defense bar.

According to statistics released by the U.S. Sentencing Commission, sentences remain at all-time highs and are still largely within the guidelines range, the exact opposite of what most expected.

Since Booker, the average sentence is 56 months, the same as it was in 2004, and up from 52 months in 2003, 51 months in 2002, and 50 months in 2001[9]. The median sentence is 34 months, up from 33 months in 2004 and 30 months in 2003, 2002, and 2001. The type of crime seeing the greatest average increase is white-collar theft/fraud for which, since Booker, the average sentence is 21 months, compared with 19 months for 2004, 16 months for 2003 and 2002, and 15 months for 2001. The median sentence for theft crimes is 12 months since Booker, the same as it was in 2004 and 2003, and up from 10 months in 2002 and 2001.[10]

In addition, since Booker, 61.2 percent of federal defendants nationwide have been sentenced within the guidelines range, a change from 72.2 percent in 2004, 69.4 percent in 2003, and from 65.0 percent in 2002. Of the 38.8 percent of the sentences outside of the guidelines range, the vast majority, 24.4 percent, are based on a government motion, either because of substantial assistance (14.8 percent), or through "fast-track" reductions to speed deportation in immigration cases (9.6 percent). This is a modest change from roughly 17 percent of government-sponsored departures in 2000 through 2002, before the fast-track program was in place.

Judges in only 9.6 percent of the cases nationwide since Booker have sentenced defendants below the range based purely on their new "Booker powers." In other words, federal judges are exercising their newfound "Booker" powers sparingly — less than one in every 10 cases to reduce sentences — and moderately, since overall sentences are still high.[11]

Circuit, District Variations

The statistics that may attract the most interest are those that show variations among the circuits and districts in exercising the "Booker" powers. New York-area lawyers will be interested to see that the Second Circuit leads the nation, with 16.7 percent of its cases since Booker sentenced below the guidelines range based on "Booker" powers, followed by the U.S. Court of Appeals for the First Circuit with 14.1 percent. This compares to national rate of 9.6 percent, and lows of just 6.4 percent in the Fifth and 6.9 percent in the Tenth circuits.

The U.S. Sentencing Commission also recently released figures by districts, which show even larger variations. The U.S. District Court for the Eastern District of New York has the third highest percentage nationally of "Booker" variances below the range, at 23.2 percent, with only the U.S. District Court for the Southern District of Iowa at 25.1 percent, and the U.S. District Court for the District of Massachusetts at 26.7 percent (two districts with substantially lower volume of cases than the Eastern District) ahead of it in the nation. The U.S. District Court for the Southern District of New York reports Booker variances below the guidelines range in 19.8 percent of its cases, also well above the national average.

Above the Guidelines' Range

Unleashed from the guidelines, judges can of course sentence defendants above the guidelines range too, and the First Circuit leads the nation in such sentences at 2.3 percent, nearly twice the national average (1.4 percent), with the Middle (6.1 percent) and Western (5.1 percent) districts of Louisiana, and the Northern District of Texas (4.9 percent) leading the nation by districts.[12]

Just what these regional variations mean is not clear. Despite some press reports and anecdotal accounts to the contrary, these statistics do not yet show that similarly situated defendants in different regions will be treated, since factors such as the volume of cases, types of cases (some regions have high volume of immigration cases, for instance) and local prosecution practices play a significant role in influencing the statistics. The U.S. Sentencing Commission is expected to present a more detailed report sometime this year further analyzing the data.

Thus far, the "Booker variance" cases appear to have allowed judges to sentence defendants outside the guideline range for unique, case-specific reasons, the very kinds of decisions judges should be making. A good example of the measured reaction is, ironically, demonstrated in Attorney General Alberto Gonzales' speech, delivered on June 21, 2005, in which he used four cases as examples of what he said was "an increased disparity in sentences, and a drift toward lower sentences" in the criminal justice system.[13] The National Association of Criminal Defense Lawyers looked at each case cited by Mr. Gonzales and issued a detailed report. Each case revealed that the judge had made a reasoned and well-justified decision under the specific circumstances. The last report indicated that the government was not appealing two of the cases.[14] If this was the worst that the Department of Justice could come up with, then judges must be exercising their new powers in moderation.

Judge Shira Scheindlin's opinion in October of last year involving a defendant convicted of distribution of crack cocaine provides another example of a Booker variance case. Judge Scheindlin discussed the "unsettling" disparity in sentences between crack and powder cocaine sentences, including the attempts to reform it in Congress, and noted that the defendant's sentence under the guidelines would have been 14 years less if convicted of distributing the equivalent amount of powder cocaine. Joining a host of similar decisions across the country, Judge Scheindlin rectified the crack-cocaine disparity for the case, sentencing the defendant to 17 and a half years (the defendant faced a mandatory minimum of 15 years), roughly 10 years below his guidelines range.[15] This is exactly what the criminal justice system should allow: federal judges considering the Sentencing Guidelines and assessing the individual merits of a case, then delivering sentences the judges believe are most fitting for the crime and for society. Decisions like those by Judge Scheindlin also help spur healthy national debate about just punishments for crimes.


One year after the Booker decision, to the general disappointment on the defense side, judges have embraced their new post-Booker powers with a measured, moderate response. The years of delivering high, mandatory sentences appear to have taken their toll, and even now judges continue to deliver high sentences. In time, judges will certainly grow more comfortable sentencing defendants below the guidelines range, but the process has been slower than originally expected. Of greater concern, judges may not have much time to exercise their expanded powers; Congress plans to hold hearings this year to examine whether the post-Booker world needs fixing. Congress will consider new laws, such as a series of mandatory minimum sentences, that again restrict the discretion of judges during sentencing. The data so far certainly do not show a need for reform.

William H. Sloane is a partner, and Kenneth S. Levine, an associate, at Carter Ledyard & Milburn LLP. Both are former state prosecutors.

Reprinted with permission from the March 6, 2006 edition of The New York Law Journal
© 2006 ALM Properties, Inc. All rights reserved. Further duplication without permission is prohibited.


[1] United States v. Olis, No. 04-20322 (5th Cir., Oct. 31, 2005), at 7.

[2] Simon Romero, "Revision of 24-Year Prison Term Ordered in Accounting Fraud," N.Y. Times, November 2, 2005, at C3, quoting Professor Frank O. Bowman.

[3] 125 S.Ct. 738 (2005)

[4] USSG §1A.1.1

[5] Hon. John S. Martin, "Let Judges Do Their Jobs," N.Y. Times, June 24, 2003

[6] U.S. Sentencing Commission, "Fifteen Years of Guidelines Sentencing," November 2004, pages 138-39; available at

[7] Id., page 140

[8] Id. at 757 (citations omitted).

[9] The annual figures are based on fiscal years as defined by the U.S. Sentencing Commission. The 2004 figures reflect data only from cases before the Supreme Court's decision in Blakely v. Washington, 542 U.S. 296 (2004), on June 24, 2004, since that case called into question the constitutionality of the U.S. Sentencing Guidelines.

[10] U.S. Sentencing Commission, Special Post-Booker Coding Project, Dec. 1, 2005, available at

[11] The changes in sentencing patterns, as moderate as they are, are even less so when compared to the sentencing statistics from several years ago. In 2003, Congress passed the PROTECT Act, which granted de novo review for appellate courts on all downward departure decisions, reduced circumstances in which courts could sentence defendants below the guidelines range, and increased reporting to monitor judicial activity. After its implementation, courts across the country sentenced defendants to within the guidelines a higher percentage of the time — a 4.4 percent increase from 2002 to 2003 nationally. Compared to 2001, for instance, overall sentences are now within the guidelines range only 2.1 percent less often, and the Ninth Circuit's patterns are virtually the same as they were in 2000 through 2002 in this regard. In other words, since Booker, courts have largely returned to their pre-PROTECT Act patterns

[12] U.S. Sentencing Commission, "Fifteen Years of Guidelines Sentencing," available at

[13] Prepared Remarks of Attorney General Alberto Gonzales, Sentencing Guidelines Speech, Washington, D.C, June 21, 2005, available at

[14] National Association of Criminal Defense Lawyers, "Truth in Sentencing? The Gonzales Cases," July 7, 2005, available at$FILE/Truth percent20Report percent20Final.pdf.

[15] United States v. Otis Fisher, S3 03 CR 1501 (SAS) (S.D.N.Y., Oct. 11, 2005)

Related practice area:

© Copyright 2018 Carter Ledyard & Milburn LLP