362 F.3d 160
UNITED STATES of America, Appellee-Cross-Appellant,
Niels LAUERSEN, Defendant-Appellant-Cross-Appellee.
United States of America, Appellee,
James Rinaldo Jackson, Defendant-Appellant.
United States Court of Appeals, Second Circuit.
Petition for Rehearing argued: January 9, 2004.
Finally submitted: January 30, 2004.
Decided: March 24, 2004.
Nathaniel Z. Marmur, New York, N.Y. (Stillman & Friedman, P.C., New York, N.Y., David I. Schoen, Montgomery, Ala., on the brief), for Defendant-Appellant-Cross-Appellee Lauersen.
Vida M. Alvy, Great Neck, N.Y. (Gail Jacobs, Great Neck, N.Y., on the brief), for Defendant-Appellant Jackson.
James G. Cavoli, Asst. U.S. Atty., New York, N.Y. (James B. Comey, U.S. Atty., Gary Stein, Lisa Horwitz, Adam Siegel, Asst. U.S. Attys., New York, N.Y., on the brief), for Appellee.
Before: NEWMAN, WINTER, and B.D. PARKER, Circuit Judges.
JON O. NEWMAN, Circuit Judge.
The United States has petitioned for rehearing in each of the captioned cases to challenge our rulings that permit the sentencing judges on remand to consider a downward departure. We permitted the departure because the combined effect of substantially overlapping offense level adjustments and the increased extent of enhancement of the applicable sentencing range that occurs at the higher end of the sentencing table resulted in the overlapping adjustments adding significantly more time to the sentence ranges than would have occurred if the adjustments had been applied at the lower end of the sentencing table.1 See United States v. Lauersen, 348 F.3d 329, 342-44 (2d Cir. 2003); United States v. Jackson, 346 F.3d 22, 26 (2d Cir.2003)2. Because we had ruled on our own motion that such "cumulative effects" departures were available and the parties had not briefed the issue on the original appeals, we granted the Government's petitions for rehearing, afforded time for briefs, and heard oral argument. We now adhere to our original rulings for the reasons set forth in this opinion.
The facts are detailed in our prior opinions. We recapitulate only the essential facts that bear on the departure issue.
Lauersen. Niels Lauersen, an obstetrician/gynecologist, was convicted of health care fraud. He submitted false claims to insurance companies by misrepresenting fertility treatments, which were not covered by his patients' insurance, as other procedures that were covered.
The District Court calculated his adjusted offense level under the Sentencing Guidelines to be 29. Starting from a base offense level of 6, see U.S.S.G. § 2F1.1(a) (2000),3 the Court added 13 levels for the amount of the loss, see id. § 2F1.1(b)(1)(N); 2 levels for more than minimal planning, see id. § 2F1.1(b)(2)(A); 4 levels for leadership role, id. § 3B1.1(a); 2 levels for abuse of trust, see id. § 3B1.3; and 2 levels for obstruction of justice, id. § 3C1.1. The adjusted offense level of 29 in Criminal History Category I yielded a sentencing range of 87 to 108 months. See id. ch. 5, pt. A (Sentencing Table). The Court imposed a sentence of 87 months (seven years and three months), the minimum of the applicable sentencing range.
On the original appeal, we upheld the Government's contention, raised on its cross-appeal, that the District Court had erred in not increasing the adjusted offense level by an additional 4 levels pursuant to U.S.S.G. § 2F1.1(b)(8)(B). Lauersen, 348 F.3d at 343. That enhancement applies if the offense "affected a financial institution and the defendant derived more than $1,000,000 in gross receipts from the offense."4 U.S.S.G. § 2F1.1(b)(8)(B). The revised adjusted offense level of 33, which we required to be used on remand, yielded a sentencing range of 135 to 168 months. See id. ch. 5, pt. A (Sentencing Table). The additional adjustment increased the minimum of Lauersen's sentencing range by 48 months (four years).
Before remanding, we ruled that the additional enhancement "creates a new circumstance that might well justify a downward departure." Lauersen, 348 F.3d at 343. As we explained, the new enhancement substantially overlapped with the existing 13-level enhancement for the amount of the loss because "the large amount of money involved in the fraud significantly triggers both of them." Id. at 344. We also noted that in Lauersen's case, the added adjustment increased the minimum of his sentencing range by four years, a much larger amount than would have resulted if his adjusted offense level had not placed him at such a high level in the sentencing table. Although recognizing that the two enhancements did not constitute impermissible double-counting, we concluded "that the cumulation of such substantially overlapping enhancements, when imposed upon a defendant whose adjusted offense level translates to a high sentencing range, presents a circumstance that is present `to a degree' not adequately considered by the Commission." Id. (quoting 18 U.S.C. § 3553(b)(1)). For that reason we permitted the District Court to "exercise discretion to mitigate the effect of the enhancement by making a downward departure." Id. (footnote omitted).
Jackson. James Jackson was convicted of fraud offenses involving identity theft employed to purchase valuable items and charge them to his victims' credit card accounts.
The District Court calculated his adjusted offense level under the Sentencing Guidelines to be 21. Starting from a base offense level of 6, see U.S.S.G. § 2F1.1(a), the Court added 10 levels for the amount of the loss, see id. § 2F1.1(b)(1)(K); 2 levels for more than minimal planning, see id. § 2F1.1(b)(2)(A); 2 levels for use of sophisticated means, see id. § 2F1.1(b)(5)(C); and 4 levels for leadership of an extensive criminal activity, see id. § 3B1.1(a). With a 3-level reduction for acceptance of responsibility, see id. § 3E1.1(a), (b), the resulting adjusted offense level of 21 in Criminal History Category VI yielded a sentencing range of 77 to 96 months. See id. ch. 5, pt. A (Sentencing Table). The Court imposed a sentence of 96 months, the maximum of the applicable sentencing range.
On the original appeal, we affirmed, ruling that the challenged adjustments, although overlapping, did not constitute impermissible double-counting. Jackson, 346 F.3d at 25-26. Nevertheless, we concluded that, as in Lauersen, the cumulative effects of the overlapping enhancements for the amount of the loss, more than minimal planning, sophisticated means, and leadership of an extensive criminal activity, combined with the significant impact of these enhancements at the higher end of the sentencing table permitted consideration of a departure by the District Court on remand. See id. at 26. As we pointed out, "Most fraud schemes that obtain more than one half million dollars involve careful planning, some sophisticated techniques, and are extensive." Id.
The Government's petitions for rehearing challenge the propriety of the "cumulative effects" departure that we authorized the sentencing judges to consider in Lauersen and Jackson. The core of the Government's argument is that each of the two factors that contributed to our allowance of such a departure — (1) the aggregation of overlapping enhancements and (2) the increased sentencing effect of an enhancement upon a defendant at the higher end of the sentencing table compared to the effect at a lower level — were expressly prescribed by the Sentencing Commission. Therefore, the Government's argument continues, they cannot constitute a circumstance present (in a particular case) to a degree not adequately considered by the Commission, such as would permit consideration of a departure under 18 U.S.C. § 3553(b) and U.S.S.G. § 5K2.0. We agree with the Government's premises, but not its conclusion.
There is no doubt, as the Government contends, that the Commission prescribed that all applicable offense-level enhancements are to be aggregated. The Commission specified in an introductory section of the Guidelines that offense level adjustments are "cumulative (added together)," whether adjustments are "from more than one specific offense characteristic within an offense guideline" or "from different guideline sections."5 U.S.S.G. § 1B1.1, comment. (n.4). We have regularly recognized that all applicable offense-level adjustments are to be aggregated, at least in the absence of impermissible double-counting. See, e.g., United States v. Martin, 78 F.3d 808, 813 (2d Cir.1996); United States v. Howard, 998 F.2d 42, 48 (2d Cir.1993). Indeed, that principle underlay our agreement with the Government's claim on its cross-appeal for an additional enhancement in Lauersen and our rejection of the defendant's challenges to his enhancements on appeal in Jackson.
There is also no doubt, as the Government further contends, that the Commission deliberately constructed the sentencing table so that each enhancement results in an increase in the sentencing range that becomes larger as the adjusted offense level increases. For example, in Criminal History Category I, a 4-level enhancement added at offense level 17 increases the sentencing range minimum by 13 months; at offense level 27, by 38 months; and at offense level 37, by 114 months.6 Instead of causing the sentencing range minimum to increase by the same number of months whenever an enhancement increases the adjusted offense levels, the sentencing table causes the sentencing range minimum to increase by approximately the same percentage. With each 4-level enhancement, the increase is approximately 50 percent. For example, a 4-level enhancement added at adjusted offense level 17 increases the sentencing range minimum by 54 percent; at level 23, by 52 percent; and at level 29, by 55 percent.7 Because the sentencing table is structured to have this percentage proportionality effect, the Government argues, the fact that Lauersen's 4-level enhancement and Jackson's various enhancements increase their sentencing range minimums by much more than would have occurred if their unadjusted offense levels were lower could not be a circumstance not adequately considered by the Commission.
Although the Guidelines contemplate the aggregation of applicable enhancements and provide increasing sentencing range increments as the adjusted offense level increases, we continue to believe that when the addition of substantially overlapping enhancements results in a significant increase in the sentencing range minimum (as it does at the higher end of the sentencing table), a departure may be considered. What is present to a degree not adequately considered by the Commission is the combined effect of the aggregation of substantially overlapping enhancements and the large increase in the sentencing range minimum at the higher end of the sentencing table. We do not doubt that in Lauersen's case, for example, the Commission expected the 4-level financial institution enhancement to be added to other applicable enhancements. Nor do we doubt that the Commission expected, in the general run of cases, that an additional 4-level enhancement would increase the sentencing range minimum by many more months for a defendant with a high unadjusted offense level than for a defendant with a low unadjusted offense level.
Nevertheless, the Commission is unlikely to have considered the combined effect upon sentencing range minimums of every group of enhancements that could be imposed under the Guidelines. The Guidelines provide for scores of possible enhancements. Even if there were only 50 possible enhancements, there would be 1,225 possible pairings of any two such enhancements.8 Where, as in the pending cases, applicable enhancements substantially overlap, we see no indication that the Commission adequately considered how serious an effect on the sentencing range minimum would result for a defendant whose unadjusted offense level is at the higher end of the sentencing table. For Lauersen, the addition of the financial institution enhancement (on top of his enhancement for the amount of loss) increases his sentencing range minimum by four years — from seven years and three months to eleven years and three months.9 For Jackson, the addition of his three overlapping enhancements (on top of his enhancement for the amount of loss) increases his sentencing range minimum by three years and eight months, from two years and nine months to six years and five months.10
The Commission itself has recognized that uniform application of the percentage proportionality structure of the sentencing table is not always warranted. In explaining the percentage proportionality aspect of the sentencing table, the Commission noted that it had "discovered that proportional (percentage) adjustments to sentence length are frequently appropriate." United States Sentencing Commission, Supplementary Report on the Initial Sentencing Guidelines and Policy Statements 15 (1987). The plain implication is that the percentage increase in sentences normally resulting when enhancements are added to an offense level are sometimes not appropriate. An aggregation of substantially overlapping enhancements at the higher end of the sentencing table would seem to be the sort of circumstance that prompted the Commission to deem the percentage proportionality of the sentencing table to be appropriate "frequently" but not always.
The Government argues that the Commission must have intended all combinations of enhancements, whether or not substantially overlapping, to have whatever effect results from the structure of the sentencing table because the Commission expressly set a limit on the increase in adjusted offense level that can result from a few specific pairings of enhancements.11 However, the fact that the Commission chose to put a ceiling on the combined effect of a small number of pairings of enhancements does not mean that it adequately considered all of the vast number of combinations of enhancements that are possible under the Guidelines and intended to preclude a sentencing judge from considering a departure where a combination of substantially overlapping enhancements has a significant effect on sentencing ranges at the higher end of the sentencing table.
The Government also argues that a "cumulative effects" departure, available for adjusted offense levels at the higher end of the sentencing table, perversely "rewards a defendant for particularly egregious wrongdoing...." Br. for Appellee at 20. This is an odd use of the concept of a "reward." Defendants like Lauersen and Jackson are being given added punishment because of the imposition of overlapping enhancements, and the amount of the added punishment is much longer in months than would have occurred if their unadjusted offense levels had been lower. The "cumulative effects" departure simply permits a sentencing judge to mitigate the extent of that added punishment, leaving them still subject to more punishment than they would have received if their enhancements had been applied to a lower unadjusted offense level. Even if they receive a departure in the sentencing judge's discretion, it is unlikely that they will consider a slight reduction in an additional punishment to be a "reward."
Finally, the Government contends that a "cumulative effects" departure for overlapping enhancements at the higher end of the sentencing table "threatens to erode the entire Guideline system and that system's core goal: national uniformity among similarly situated defendants." Br. for Appellee at 24. We think this contention substantially overstates the impact of this departure. Preliminarily, we note our doubt that national uniformity is being achieved under the Sentencing Guidelines. Although each guideline is applied to all defendants within its coverage, mechanisms such as departures authorized by the Guidelines and various other circumstances are markedly reducing whatever uniformity of results might have been expected.
For example, as the Commission itself has reported based on the most recent data (2001), downward departure rates vary from a low of 1.4 percent in the Eastern District of Kentucky to a high of 62.6 percent in the District of Arizona. See United States Sentencing Commission, Report to the Congress: Downward Departures from the Federal Sentencing Guidelines 32 (2003) ("2003 Report"). Moreover, the Government itself is responsible for much of the lack of uniformity. Substantial assistance departures, available only upon the Government's motion, see U.S.S.G. § 5K1.1, accounted for 49 percent of all downward departures, 2003 Report at 60, and the rate of such departures varied from above 35 percent in five districts to below seven percent in five districts, id. at 68. In addition, apart from substantial assistance departures, Government initiated departures accounted for more than one third of all other downward departures. Id. at 60.
Furthermore, the Government's charging decisions and presentations of facts contribute to many sentences that are markedly not uniform. To cite just one recent example within our Circuit, one defendant convicted of viewing child pornography on the Internet, who had not attempted to approach any children, received a Guidelines sentence of ten years, see United States v. Sofsky, 287 F.3d 122, 124 (2d Cir.2002), and another defendant convicted of distributing child pornography over the Internet, who had solicited for sexual purposes a person he thought was a 13-year-old girl, received a Guidelines sentence of just under four years, see United States v. Maneri, 353 F.3d 165, 167-68 (2d Cir.2003). See also United States v. Bonnet-Grullon, 212 F.3d 692, 706-08 (2d Cir.2000) (rejecting departure despite lack of nationwide uniformity attributable to regional variations in prosecutors' plea-bargaining practices).
Wholly apart from these and other anomalies that pervade Guidelines sentencing, we think the Government's apprehension about the consequences of permitting a "cumulative effects" departure for overlapping enhancements at the higher end of the sentencing table is considerably overstated. In the first place, not many combinations of enhancements will be substantially overlapping, although the two enhancements at issue in Lauersen will normally be an overlapping pair. Second, the departure is available only at the higher end of the sentencing table. Third, the extent of the departure can only diminish, and not eliminate, the added punishment caused by the overlapping enhancements. Finally, the decision whether to depart, and the extent of the departure, if made at all, remain within the discretion of the sentencing judge. Under these circumstances, there is no realistic "threat to erode the entire Guideline system."
We have previously recognized that, in some limited circumstances, a combination of enhancements can permit consideration of a downward departure. See United States v. Gigante, 94 F.3d 53, 56 (2d Cir. 1996) (downward departure authorized where substantially enhanced sentence range results from a series of enhancements proven only by a preponderance of the evidence), amending 39 F.3d 42, 48 (2d Cir.1994); cf. United States v. Concepcion, 983 F.2d 369, 387-89 (2d Cir.1992) (downward departure authorized where enhancement based on acquitted conduct resulted in substantial increase in sentence); United States v. Rahman, 189 F.3d 88, 157-58 (2d Cir.1999) (downward departure authorized where sentence was lengthened by prosecutor's decision to add duplicative counts). In addition, we have approved, at the Government's urging, an upward departure in circumstances where a combination of enhancements did not fully reflect the degree of harm inflicted upon a victim. See United States v. Kaye, 23 F.3d 50, 53-54 (2d Cir.1994).12
Although departure authority is not a device for subverting the Guidelines, see Bonnet-Grullon, 212 F.3d at 709, its restrained use in limited circumstances can provide appropriate flexibility in an elaborate sentencing regime that, however thoughtfully constructed, could not possibly anticipate all of the circumstances that might arise in its application. That is why the governing statute and the Guidelines themselves permit departures, even when a circumstance has been considered by the Commission, as long as that circumstance is present "to a degree" not adequately considered by the Commission. 18 U.S.C. § 3553(b); U.S.S.G. § 5K2.0. We continue to believe that this limited leeway permits consideration of departures in the two pending cases, and we therefore adhere to our prior decisions.
The sentencing table is constructed with the low adjusted offense levels and the corresponding low sentencing ranges in the top rows of the table and the high adjusted offense levels and the corresponding high sentencing ranges in the bottom rows of the table. By using the terms "higher end" and "lower end" of the sentencing table, we mean the portions of the sentencing table at which sentencing ranges are high and low, respectively
AlthoughLauersen is reported in the Federal Reporter after Jackson, it was decided two weeks earlier. The later placement results from the fact that the original opinion, filed on Sept. 15, 2003, was amended on Nov. 25, 2003, after the Jackson opinion was filed on Oct. 1, 2003, and amended on Oct. 22, 2003.
All references are to the 2000 version of the Guidelines, which applies to both of the pending cases
The financial institution adjustment, after an amendment in 2001,see U.S.S.G.App. C, vol. II, amend. 617, provides an increase of only 2 levels, see id. § 2B1.1(b)(12)(A) (2001). Lauersen cannot benefit from this reduction because the amended Guidelines, if applied to all aspects of his guideline calculation, would increase his adjusted offense level from 33 to 36, and the "one-book" rule precludes application of only one beneficial change from a subsequent version of the Guidelines. See United States v. Keller, 58 F.3d 884, 889 (2d Cir.1995).
The introductory section provides an exception to cumulation where "the guidelines specifies that only the greater (or greatest) [of the adjustments] is to be used." U.S.S.G. § 1B1.1, comment. (n.4)
The Commission's policy decision to impose more punishment for the same enhancement upon defendants with higher adjusted offense levels than upon those with lower adjusted offense levels would seem to have stronger justification for some enhancements than for others. For example, the 4-level enhancement for being the leader of a criminal activity with five or more participants,see U.S.S.G. § 3B1.1(a), should result in a larger increment of punishment for a defendant who is the leader of an organization selling large quantities of narcotics than for a defendant who is the leader of an organization selling small quantities of narcotics. On the other hand, if two narcotics sellers recklessly endangered another person while fleeing from the police, it not so clear that the 2-level enhancement for creating such danger, see U.S.S.G. § 3C1.2, should result in a larger increment of punishment for the seller of large quantities of narcotics than for the seller of small quantities of narcotics, after their respective narcotics quantities had already differentiated their adjusted offense levels.
In general, "[a] change of six levels roughly doubles the sentence irrespective of the level at which one starts." U.S.S.G. ch. 1, pt. A, ¶ 4(h). As the Government points out, "theonly way to achieve `percentage proportionality' — as was the aim of the Commission — is to increase sentencing ranges by a disproportionately greater number of months as offense levels rise." Br. for Appellee at 16 n.*.
The formula for the number of combinations of p selections from a group of n items is n!/(n-p)! × p!
At oral argument on rehearing, the Government disputed that the enhancement increased Lauersen's sentencing range minimum by four years. The argument is based on the Commission's recommended sequencing of adjustments in arriving at an adjusted offense level. The Guidelines instruct that specific offense characteristics, such as the financial institution adjustment, are to be applied, in the order listed, immediately after the base offense level has been determined,see U.S.S.G. § 1B1.1(b), i.e., prior to other adjustments. Thus, the Government argues, the financial institution adjustment should have been added after the sentencing judge determined the base offense level of 6 and added 13 levels for the amount of loss and 2 levels for more than minimal planning. Therefore, the argument continues, the addition, at that point in the calculation, of 4 levels for the financial institution enhancement would have increased the adjusted offense level from 21 to 25, which would have increased Lauersen's sentencing range minimum by 20 months, from 37 to 57.
The argument is arithmetically correct in reflecting the proper sequence of determining the final adjusted offense level, but incorrect in its ultimate conclusion. The incontrovertible facts are that without the financial institution enhancement, Lauersen's sentence would have been (in fact, was) seven years and three months; with the enhancement, it is eleven years and three months; and the difference is four years. Perhaps what the Government means is that the financial institution enhancement accomplishes a four-year increase only because it is imposed in combination with other enhancements.
It is instructive to compare what the effect of each of Jackson's three overlapping enhancements would have been if each one had been added without the other two. With a base offense level of 6, a 10-level adjustment for amount of loss, and a 3-level decrease for acceptance of responsibility, Jackson's adjusted base offense level was 13 before any of the three overlapping adjustments was added. Each of his 2-level adjustments alone would have increased his adjusted offense level to 15, causing his sentencing range minimum to increase by 8 months — from 33 months to 41 months (in Criminal History Category VI). His 4-level adjustment alone would have increased his adjusted offense level to 17, causing his sentencing range minimum to increase by 18 months — from 33 months to 51 months
However, with either one of the two 2-level adjustments and the 4-level adjustment already increasing the adjusted offense level from 13 to 19, the addition of the second 2-level adjustment increased the adjusted offense level to 21, causing his sentencing range minimum to increase by 14 months — from 63 months to 77 months. Thus, the last 2-level adjustment, imposed with the other overlapping adjustments, nearly doubled the increase in his sentencing range minimum compared to its effect if added alone (14 months versus 8 months).
The cumulative effect of the 4-level enhancement can be similarly analyzed. With each of the two 2-level adjustments already increasing the adjusted offense level from 13 to 17, the addition of the 4-level adjustment increased the adjusted offense level to 21, causing his sentencing range minimum to increase by 26 months — from 51 months to 77 months. Thus, the 4-level adjustment, imposed with the other overlapping adjustments, increased his sentencing range minimum 44 percent more than would have occurred if this adjustment had been added alone (26 months versus 18 months).
For example, for the offense conduct of aggravated assault, the maximum enhancement for use of a firearm is 5 levels, U.S.S.G. § 2A2.2(b)(2)(A), and the maximum enhancement for injury to a victim is 6 levels,id. § 2A2.2(b)(3)(C), but cumulative adjustments under both provisions are capped at 9, id. Similar capping is prescribed for a combination of weapon and injury enhancements with respect to robbery, id. § 2B3.1(b)(3), and extortion, id. § 2B3.2(b)(4).
The Government relies onUnited States v. Kuhn, 345 F.3d 431, 436-38 (6th Cir.2003), which rejected a downward departure based on the relationship between enhancements for abuse of trust, U.S.S.G. § 3B1.3, and leadership role, id. § 3B1.1. Those enhancements, although not entirely dissimilar, did not involve a substantial overlap.