Opinion
Criminal Number H-03-217-01.
September 22, 2006
MEMORANDUM OPINION
On October 31, 2005, the United States Court of Appeals for the Fifth Circuit affirmed Jamie Olis's conviction for securities fraud, mail and wire fraud, and conspiracy, but vacated his 292-month prison sentence and remanded the case to this court for resentencing. United States v. Olis, 429 F.3d 540 (5th Cir. 2005). The government argues that the sentencing guideline range is 292-365 months and that the court should sentence Olis to 292 months in prison. Olis argues that the guideline range urged by the government does not apply because the loss attributable to Olis "cannot be proved with sufficient reliability and certainty." Olis also argues that if the court adopts the guideline range proposed by the government, the court should impose a non-guideline sentence.
I. Background
Olis, along with Gene Foster and Helen Sharkey, was indicted for conspiracy to commit mail fraud, wire fraud, and securities fraud (count 1), securities fraud (count 2), mail fraud (count 3), and wire fraud (counts 4-6). Foster and Sharkey pleaded guilty to the conspiracy count and cooperated with the government. The jury convicted Olis on all counts. The court found that under the mandatory sentencing guidelines Olis was required to be sentenced within a range of 292-365 months in prison. The court sentenced Olis to 292 months in prison, three years of supervised release, and a $25,000 fine. Olis appealed.
As summarized by the Fifth Circuit, this range applied because
(1) [the court found that] Olis was responsible for an approximately $105 million loss to the University of California Retirement System, which enhanced his base offense by twenty-six levels under the Sentencing Guidelines;
(2) Olis's offense involved sophisticated means, requiring a two-level enhancement;
(3) Olis used a special skill, in a manner that significantly facilitated the commission or concealment of the offense, resulting in another two-level enhancement; and
(4) Olis's scheme included fifty or more victims, requiring a four-level sentencing enhancement.Olis, 429 F.3d at 534.
After the court sentenced Olis, but before the Fifth Circuit issued its decision, the United States Supreme Court issued its decision in United States v. Booker, 125 S.Ct. 738 (2005). The Court held that the sentencing guidelines are no longer mandatory. Instead, the guidelines are merely advisory, and sentences must be "reasonable." See United States v. Johnson, 445 F.3d 793, 797 (5th Cir.), cert. denied, 126 S.Ct. 2884 (2006). In addition to the guidelines, the federal sentencing statute, 18 U.S.C. § 3553, requires courts to consider a number of other relevant factors in determining a sentence.
The Fifth Circuit held that Booker applied, and that Olis must be resentenced because the government had not established "beyond a reasonable doubt that the district court would have sentenced Olis to nearly twenty-five years in prison had it acted under an advisory Sentencing Guidelines scheme as required byBooker." Olis, 429 F.3d at 544. Because Booker also required a sentencing court to "`consider' the guidelines before issuing a `reasonable' sentence," the Fifth Circuit also reviewed this court's sentencing determinations. Id. citing Booker, 125 S.Ct. at 757, 767. The Fifth Circuit held that the court "overstated the loss caused by Olis's crimes." Id. (emphasis in original). Although Olis also challenged the court's application of sentencing enhancements for the use of "sophisticated means," U.S.S.G. § 2B1.1(b)(8)(C), "special skill," U.S.S.G. § 3B1.3, and for there being more than 50 victims of his crime, U.S.S.G. § 2B1.1(b)(2)(B), the Fifth Circuit affirmed the application of these enhancements by concluding that "Olis used his special skills in accounting and tax matters to advance an extremely sophisticated, but fraudulent, scheme[,]" Olis, 429 F.3d at 549, and that the enhancement based on the number of victims "clearly applies" in this case. Id.
In United States v. Tzep-Mejia, F.3d, 2006 WL 2361701 (5th Cir., August 15, 2006), the Fifth Circuit clarified the post-Booker sentencing regime by explaining that the law now recognizes three types of sentences:
(1) a sentence within a properly calculated Guideline range; (2) a sentence that includes an upward or downward departure as allowed by the Guidelines, which sentence is also a Guideline sentence; or (3) a non-Guideline sentence which is either higher or lower than the relevant Guideline sentence.Id. at *2 (citing United States v. Smith, 440 F.3d 704, 707 (5th Cir. 2006)). Before a court may impose
a non-Guideline sentence, it must first calculate the Guideline range and consider the appropriateness of a sentence within that sentencing range to fulfill its duty to consider the Sentencing Guidelines as advisory and as a frame of reference.Id. (citing United States v. Mares, 402 F.3d 511, 518-19 (5th Cir.), cert. denied, 126 S.Ct. 43 (2005)).
Under both Supreme Court and Fifth Circuit precedent, the court must first calculate the guideline range and then determine whether a sentence within that range adequately addresses all of the other statutory sentencing factors. Both parties have submitted evidence and arguments addressing the correct guideline range and the appropriateness of a non-guideline sentence.
II. The Defendant's Guideline Sentencing Range
As recognized by the Fifth Circuit, "[t]he most significant determinant of Olis's sentence is the guidelines loss calculation." Olis, 429 F.3d at 545. The government argues that the actual losses attributable to the defendant's unlawful conducted related to Project Alpha exceed $100 million, and that the intended loss to the United States Treasury was $79 million. The defendant argues that the government has failed to establish beyond a reasonable doubt, by clear and convincing evidence, or to a reasonable certainty that the actual loss attributable to Project Alpha exceeds $100 million. Defendant argues that the government has failed to establish actual loss in accordance with any standard of proof applicable to this case because the declaration submitted by the expert witness engaged to calculate actual loss "fails to show a price increase or price decrease caused by Project Alpha," "understates the confounding causes of decline," and ignores a competing proposition — that Alpha caused no loss — that is consistent with Dynegy's actual stock behavior" over the life of the fraud.
Mr. Olis' Sentencing Memorandum, Docket Entry No. 270, pp. 10-18.
Id. at p. 16.
Id.
Id. at p. 17.
A. Loss
The guideline covering offenses involving fraud provides for a base offense level of 6, graduated increases "[i]f the loss exceeded $5,000," and additional increases if the offense involved more than 10 victims. U.S. Sentencing Guidelines Manual § 2B1.1 (2001). The application notes following § 2B1.1 explain that "loss is the greater of actual loss or intended loss." Id. at cmt. n. 2(A). "`Actual loss' means the reasonably foreseeable pecuniary harm that resulted from the offense." Id. at cmt. n. 2(A)(i). This definition of "actual loss"
incorporates [a] causation standard that, at a minimum, requires factual causation (often called `but for' causation) and provides a rule for legal causation (i.e., guidance to courts regarding how to draw the line as to what losses should be included and excluded from the loss determination).
U.S. Sentencing Guidelines Manual Supplement to Appendix C, Amendment 617 (November 1, 2001). "`Intended loss' (I) means the pecuniary harm that was intended to result from the offense; and (II) includes intended pecuniary harm that would have been impossible or unlikely to occur." U.S. Sentencing Guidelines Manual § 2B1.1 cmt. n. 2(A)(ii) (2001). "`Pecuniary harm' means harm that is monetary or that otherwise is readily measurable in money." Id. at cmt. n. 2(A)(iii). "`[R]easonably foreseeable pecuniary harm' means pecuniary harm that the defendant knew or, under the circumstances, reasonably should have known, was a potential result of the offense." Id. at cmt. n. 2(A)(iv). The guidelines recognize that "[t]he court need only make a reasonable estimate of the loss," and that "[t]he sentencing judge is in a unique position to assess the evidence and estimate the loss based upon that evidence." Id. at cmt. n. 2(C). "The estimate of loss shall be based on available information, taking into account, as appropriate and practicable under the circumstances, factors such as . . . the approximate number of victims multiplied by the average loss to each victim." Id. at cmt. n. 2(C)(iii). The guidelines define "victim" to mean "any person who sustained any part of the actual loss determined under subsection (b)(1)."Id. at cmt. n. 3(A)(ii).
The Fifth Circuit has explained that
[t]he loss guideline is skeletal because it covers dozens of federal property crimes. Some flesh can be added, however, where the gravamen of the offense conduct is securities fraud perpetrated on an established market. Useful guidance appears in the applicable principles for recovery of civil damages for securities fraud. The civil damage measure should be the backdrop for criminal responsibility both because it furnishes the standard of compensable injury for securities fraud victims and because it is attuned to stock market complexities. In civil cases, the principle of loss causation is well established. See Dura Pharmaceuticals, Inc. v. Broudo, ___ U.S. ___, 125 S.Ct. 1627, 1631-32 (2005); see generally Greenberg v. Crossroads Systems, Inc., 364 F.3d 657 (5th Cir. 2004); Private Securities Litigation Reform Act, 15 U.S.C. § 78u-4(b). Thus, there is no loss attributable to a misrepresentation unless and until the truth is subsequently revealed and the price of the stock accordingly declines. Where the value of a security declines for other reasons, however, such decline, or component of the decline, is not a "loss" attributable to the misrepresentation. See also United States v. Grabske, 260 F.Supp.2d 866, 869-871 (N.D. Cal. 2002).Olis, 429 F.3d at 546. The Fifth Circuit has also recognized as noteworthy the following features from cases applying the guidelines to securities fraud loss calculations:
15 U.S.C. Section 78u-4(b)(4) states:
Loss causation: In any private action arising under this chapter, the plaintiff shall have the burden of providing that the act or omission of the defendant alleged to violate this chapter caused the loss for which the plaintiff seeks to recover damages.
Under the PSLRA the general method for calculating loss is the difference between the price paid by the plaintiff and the price after ameliorative information is released to the market, minus any amount of loss that may have been caused by other market factors present during the period of loss. Olis, 429 F.3d at 546.
(1) "[G]iven the time and evidentiary constraints on the sentencing process, the methods adopted in those cases are necessarily less exact than the measure of damage applicable in civil securities litigation." Id. at 247. Nevertheless, the estimated loss must be reasonable in light of reliable and specific evidence and may not be speculation. See United States v. Renick, 273 F.3d 1009, 1025-28 (11th Cir. 2001), United States v. Vitek Supply Corp., 144 F.3d 476, 490-92 (7th Cir. 1998), cert. denied, 119 S.Ct. 1026 (1999).
(2) At least two sentencing courts have "rejected an over simplified market capitalization measure of damages proffered by the Government in favor of a more nuanced approach modeled upon loss causation principles." Olis, 429 F.3d at 546 (citing United States v. Snyder, 291 F.3d 1291, 1295-1296 (11th Cir. 2002), and United States v. Bakhit, 218 F.Supp.2d 1232, 1238 (C.D. Cal. 2002)).
(3) Courts have also "rejected defendants' arguments that attempted to reason away all losses caused by the fraud." Id.
(4) "[T]he factual variations among these cases reflect the importance of thorough analyses grounded in economic reality." Id. (citing Snyder, 291 F.3d at 1295-96, and Grabske, 260 F.Supp.2d at 871-74).
1. Actual Loss
The government's expert witness, Frank C. Graves, has submitted a report concluding that "the aggregate losses to all . . . affected shareholders [attributable to Project Alpha] is quite large, in the range of $161 million to $714 million." Graves arrived at these figures by conducting statistical events studies for the day of or the day after "the dates when it was revealed that prior accounting disclosures about Project Alpha were of concern to the SEC regulators:" April 25, 2002, and May 8, 2002.
Declaration of Frank C. Graves, June 12, 2006, Exhibit A attached to Government's Rebuttal to Defendant's Expert Reports, Docket Entry No. 278, p. 4 ¶ 8.
Id. at p. 9 ¶ 20.
Events studies are a standard technique used in securities litigation analyses to isolate the impact of a particular event or a series of events on the value of a stock. These studies typically account for the contemporaneous movements in the value of all stocks and in the value of other stocks in the sector.See Imperial Credit Indus., Inc. Sec. Litig., 252 F.Supp.2d 1005, 1014 (C.D. Cal. 2003) (describing "[a]n event study [a]s a statistical regression analysis that examines the effect of an event on a dependent variable, such as a corporation's stock price," and identifying events studies as an accepted means of distinguishing between fraud-related and non-fraud-related changes in a stock's behavior). Graves' events studies had three parts: (1) a linear regression model of Dynegy's stock price as a function of industry and market indices represented by two peer groups, (2) a forecast using that model for calculating abnormal returns on event days, and (3) a determination of whether any abnormal returns were statistically significant. Graves' stated purpose for conducting his events studies was to determine "whether Dynegy's stock price behaved in a different way (i.e., moving up or down more than expected), on the day of or the day after" Project Alpha related disclosures were made.
Id. at p. 11 ¶ 24.
Id.
Id. at p. 9 ¶ 20.
(a) April 25, 2002
On April 25, 2002, Dynegy filed an 8-K with the SEC disclosing that $300 million of cash flow generated by Project Alpha would be reclassified as cash flow from financing instead of cash flow from operations as originally reported, and that Project Alpha was the subject of an informal inquiry by the SEC. The 8-K also disclosed other negative news about the company that Graves characterized as unrelated to Project Alpha, including: (1) a downward change in earnings guidance, (2) an asset write-off totaling $300 million from Dynegy's broadband business, Dynegy Global Communications (DGC), (3) higher than expected losses at DGC, (4) a $13 million charge to Northern Natural Gas (NNG), and (5) a reduction in earnings from Dynegy's Midstream Services. The announcements made in the 8-K caused Moody's to issue a warning of possible downgrade to Dynegy's credit rating at approximately 3:00 p.m. EDT on April 25. Downgrades in analysts' stock recommendations followed.
Id. at p. 17 ¶ 42 and p. 20 ¶ 49.
Id. at p. 7 ¶ 13, p. 8 ¶ 16, and pp. 18-20 ¶¶ 43-48.
On April 25, 2002, Dynegy's stock price fell from $27.06 to $19.04 (a 30% drop). On April 26 Dynegy's stock price fell further to $14.77 per share. Graves concluded that
Id. at p. 8 ¶ 16.
[t]he total drop in Dynegy's stock price from the closing price of $27.06 per share on April 24, 2002 to the closing price of $14.77 per share on April 26, 2002 was $12.29 per share. Of this amount, my event study analysis showed that approximately $1 per share was attributable to market and industry trends. The rest of the drop, between $11.13 and $11.54 per share, was a result of the events on April 25 and 26 (the "April 25 Events"). Given the approximately 367 million shares outstanding, the resulting reduction in Dynegy's market capitalization attributable to the April 25 information was between $4.1 and $4.2 billion.
Id. at p. 16 ¶ 36.
From this information Graves concluded that "[t]he April 25 event, takes place over two days — April 25 and April 26, 2002." Graves also concluded that
Id. at p. 15 ¶ 34.
at least some of the downward price movement caused by the April 25 information was attributable to Project Alpha, because there was subsequent downward movement on May 8 in response to a disclosure that day which simply clarified and strengthened the bad news previously revealed about Project Alpha on April 25 (i.e., that an informal SEC investigation of Project Alpha was now a formal one). This means the earlier news had to also be worrisome to investors.
Id. at p. 16 ¶ 37.
Recognizing, however, that not all of the bad news disclosed on April 25 was related to Project Alpha, Graves undertook "a structural analysis of what the various components of the news . . . were worth to the shares of Dynegy."
This analysis was conducted by estimating ranges for how much each of the pieces of information revealed on April 25 can reasonably be deemed to have contributed to this overall decline in value, and then using the high end of these ranges to quantify the activities whose value is distinct from Project Alpha (the "Non-Alpha Events"). The balance represents losses attributable, in part, to Project Alpha and its fallout for creditworthiness and equity recommendations (the "Alpha-Related Events").
Id. at p. 16 ¶ 38.
Id. at p. 17 ¶ 39.
To calculate the impact that the reduction in earnings guidance had on Dynegy's market capitalization, Graves calculated a price-to-earnings (P/E) ratio "based on the analysts' previous 2002 EPS projection and the April 24, 2002 Dynegy closing stock price. This P/E ratio was multiplied by the analysts' downward adjustment (i.e., old minus new projection) to their respective projections of Dynegy's 2002 EPS." Graves concluded that "[t]he resulting range of potential market value reductions [attributable to Dynegy's reduction in earnings guidance] is from $1.09 billion to $1.37 billion."
Id. at p. 21 ¶ 50.
Id.
Assuming that the impact caused by the charge to NNG, the losses reported for DGC, and the losses reported for Dynegy's midstream services were already reflected by the drop in market capitalization attributed to the reduction in earnings guidance, Graves neither calculated the impact attributable to each of these announcements individually nor attributed any additional drop in market capitalization to them.
Id. at p. 22 ¶¶ 52-54.
To calculate the impact that the $300 million DGC asset write-off had on Dynegy's market capitalization, Graves borrowed asset multiples used by a Deutsche Bank analyst in a report on other telecommunications providers to calculate DGC's total market value (equity plus enterprise) and applied these factors (0.3x to 1.6x) to the amount of DGC assets as presented in Dynegy's 2001 annual report. Then applying an impairment factor "based on the percent of the DGC value that was eliminated in the $300 million write-off," Graves calculated a "possible market value for DGC of between $138 million and $1.4 billion." Deducting the $300 million DGC write-off from these amounts, Graves concluded that "[t]he resulting value of Dynegy's announcement of broadband write-offs outside of its reduction in 2002 earnings was between negative $162 million and positive $1.1 billion." Graves explained that "this negative amount occurs because, if the low valuation reflected the market's valuation of DGC's assets, the valuation of the EPS [earnings per share] guidance impact already more than offsets the value of the DGC assets written-off." Accordingly, Graves concluded that the "total reduction in Dynegy's market capitalization as a result of `Non-Alpha Events' is $930 million to $2.5 billion." Subtracting these amounts from the $4.1-$4.2 billion loss in market capitalization that he attributed to the April 25 announcements, Graves concluded that "between $1.6 billion and $3.3 billion (or between 40% and 78%) of the ARs [abnormal returns] on April 25 and 26, 2002, . . . [were] attributable to Project Alpha." Graves then used the low end of this range (40%) to conclude that the per share price drop attributable to Project Alpha was 40% of $11.13, i.e., $4.45.
Id. at p. 23 ¶ 55.
Id.
Id.
Id. at p. 24 ¶ 56 (citing Exhibit FCG-5C attached thereto).
Id.
Id. at p. 24 ¶ 57 (citing Exhibit FCG-5A attached thereto).
Id. at ¶ 58.
Id. at p. 35 ¶ 91.
(b) May 8, 2002
On May 8, 2002, Dynegy filed an 8-K with the SEC disclosing "that Project Alpha was now the subject of a formal investigation by the SEC and the ensuing SP [Standard and Poors] downgrade." "Dynegy's stock fell from $12.15 to $11.05 on May 8, 2002 (a 9% drop in price)." Although the stock price fell only $1.10 on May 8, Graves concluded that "[i]nvestors buying after the May 8, 2002 correction to the disclosures of Project Alpha would not be affected by it." In other words, Graves concluded that the inflation attributable to Project Alpha
Id. at p. 9 ¶ 20.
Id. at p. 8 ¶ 17.
Id. at ¶ 18.
is $0 on May 8, 2002 and the true price is equal to the actual price. However, the day before, on May 7, the actual price of Dynegy was $12.15 while the predicted price was $10.88 (predicted by the model using the 2001 Peer Group), a difference of $1.27. Since the Dynegy news revealed on May 8 was the fact that the SEC was pursuing a formal investigation of Project Alpha, and no other confounding information also occurs then, this is per se the cause of that difference and therefore is artificial inflation. That amount of inflation persists until the prior revelation on April 25, 2002.
Id. at p. 15 ¶ 33. See also id. at p. 14 ¶ 30, and p. 35 ¶ 89.
(c) Government's Final Loss Calculation
Using the per share price drop attributable to Project Alpha-related disclosures made on April 25 ($4.45), and the per share inflation attributable to Project Alpha calculated for May 7 ($1.27), Graves presented two scenarios for estimating the actual loss caused by the Project Alpha-related disclosures: Scenario 1 based solely on the May 8 event, and Scenario 2 based on both the May 8 and the April 25 events. Asserting that "the amount of the stock price inflation varies over time and by scenario," Graves concluded that
Id. at p. 26 ¶ 62.
[b]ased on the "2001 Peer Group" results, the Dynegy stock price was inflated by $1.27 over August 14, 2001 through May 7, 2002 time period due to Project Alpha if only the May 8, 2002 event is included in the analysis. If, in addition to the May 8, 2002 event, 40% of the April 25, 2002, event attributable to Project Alpha is included, the Dynegy stock price is inflated by $5.72 between August 14, 2001 and April 24, 200[2]; by $2.81 on April 25, 200[2]; by $1.27 from April 26, 2002 through May 7, 2002; and by $0 on May 8, 2002.
Id. (citing Exhibit FCG-6 attached thereto).
As the final step of his analysis, Graves estimated the number of damaged shares. In this case Graves made "direct calculations for specific institutional investors as well as probabilistic calculations for non-specific shareholders based on damaged shares models." Graves divided his analysis into two parts: the 50 largest shareholders, and all other shareholders.
Id. at p. 27 ¶ 64.
Id. at ¶ 66.
Having only end-of-month shareholding information for the 50 largest investors, Graves used approximations to identify damaged shares. "In the case of Scenario 1 where only the May event is considered in estimating inflation, any share purchased after August 31, 2001 and held through May 31, 2002 is considered damaged." Using this approximation Graves concluded that there were "between 153 and 204 million shares held by these large investors . . . on average throughout the fraud period, but only around 47 to 57 million of those shares are damaged." Graves then concluded that "the total damage estimate for the top 50 institutional shareholders ranges from approximately $60 million under Scenario 1 (May 8 inflation only) to about $284 million under Scenario 2 (May 8 plus April 25 inflation)."
Id. at p. 28 ¶ 70.
Id. at p. 30 ¶ 73 (citing Exhibit FCG-7A attached thereto).
Id.
Lacking specific information for the shareholdings of non-institutional investors, Graves applied two types of damaged share models: the proportional trader model (PTM) and the two-trader model (TTM). Based on these models Graves estimated the actual loss incurred by the non-institutional shareholders under Scenario 1 (May 8 event only) to range from "$100 million (under the TTM, with 78.9 million damaged shares) to $121 million (under the PTM, based on 95 million damaged shares)," and under Scenario 2 (May 8 and April 25 events) to range from "$430 million using the TTM to $509 million using the PTM."
Id. at p. 34 ¶ 83.
Id.
In sum, Graves estimated the total amount of actual loss caused by Project Alpha to range from $161 million based on 126.1 million damaged shares under Scenario 1 to $714 million with 182.4 million damaged shares under Scenario 2. Graves explained that
Id. at ¶ 84.
[t]he fact that these damages are large is not surprising, because Dynegy was a very large corporation, and the misrepresentations of Project Alpha affected its stock value for almost a year before being corrected. That is long enough to affect a large portion of the shares. It had about 250 million shares outstanding throughout the duration of Project Alpha's misrepresentations, and the stock fell in value by $16 between market-closing on April 24 and May 8, when Project Alpha was being revealed. Although not all of this price drop was due to Project Alpha, roughly half of the shares were damaged by up to about one third of the total fall in value per share. This inevitably results in large damages.
Id. at pp. 34-35 ¶ 86.
(d) Analysis
Although Graves conducted a thorough analysis of all the relevant factors, a number of confounding negative announcements made on each of the days for which he identified abnormal returns, i.e., April 25 and May 8, 2002, required him to base each of the numerous calculations that underlie his estimate of actual loss on unprovable and often unexplained assumptions. Examples of these include his assumption that only $1 per share of the April 25-26 price drop was attributable to market and industry trends, that announcements concerning the $13 million charge to NNG, and the unexpected DGC and midstream losses did not need to be individually calculated because any impact they had on the drop in market capitalization was already reflected by the drop attributed to the reduction in earnings guidance, that borrowing asset multiples used by a Deutsche Bank analyst in a report on other telecommunications providers constituted a reasonable means for calculating DGC's total market value, and that the inflation in Dynegy's share price attributable to Project Alpha fell to zero on May 8. Moreover, the resulting calculations are additionally confounded by the threat to Dynegy's credit rating announced by Moody's late in the day on April 25 that was followed by downgrades in analyst recommendations. Although Graves recognized that these external announcements were caused by the internal, i.e., company, announcements of April 25, he did not attempt to unbundle the fraudulent from the non-fraudulent causes for them but, instead, interpreted them as wholly related to Project Alpha. Although Graves supports this interpretation with cites to analyst reports, each of the cited reports makes it clear that Project Alpha was only one of many analyst concerns.
Id. at p. 16 ¶ 36.
Id. at p. 22 ¶¶ 52-54.
Id. at pp. 23-24 ¶¶ 55-56.
Id. at p. 15 ¶ 33.
The multiplicity of confounding announcements made both by and about Dynegy on April 25-26, and the lack of reliable or specific information about the number of damaged shares held by either institutional or non-institutional investors posed methodological difficulties that required Graves to rely on unprovable assumptions at every step of the equation used to estimate the loss attributable to corrective information about Project Alpha. Absent guidance from the sentencing commission or the court of appeals on how to decide when results produced by methodologies that are necessarily based on a series of unprovable assumptions that yield speculative results are nevertheless reasonably certain estimates of actual loss caused by a defendant's unlawful conduct, the court is compelled to conclude that the confounding announcements and the unprovable assumptions on which Graves necessarily relied in reaching his estimate of actual loss demonstrate that it is not possible to estimate with any degree of reasonable certainty the actual loss to shareholders caused by the corrective disclosures about Project Alpha made on April 25, 2002.
Although Graves concluded that the corrective disclosures about Project Alpha made on May 8, 2002, were not confounded by unrelated disclosures, the defendant argues that the May 8 disclosures were, in fact, confounded by disclosures of other negative news. Negative disclosures that the defendant argues impacted Dynegy's share price on May 8 include information that the Federal Energy Regulatory Commission (FERC) had decided to broaden its investigation into the California energy crisis to include Dynegy, and that Dynegy had used round-trip trades to inflate its volume figures. While the number of confounding announcements that could have impacted Dynegy's share price on May 8 are both disputed and fewer than those made on April 25, it is undisputed that between the Project Alpha-related announcements of April 25 and May 8, Dynegy remained in the news and its stock price remained volatile. On April 30 following news that Dynegy had successfully renewed a $900 million line of credit, its stock price rose 30% from $13.73 to $17.84. However, when Dynegy's credit ratings failed to improve, the stock price fell to $16.55 on May 1, and fell again the next day to $14.95. On May 7 the price fell 17.5% from $14.72 to $12.15, and on May 8 it fell 9% from $12.15 to $11.05. Nor is it disputed that during the first nine days of May 2002 the drop in Dynegy's stock price outstripped that of the peer group stocks that Graves used to measure abnormal returns.
See Exhibit FCG-3A attached to Graves' Declaration.
Although Graves concluded that the drop in Dynegy's stock price on May 7 was caused solely by the announcement that investigations into the California energy crisis were being broadened to include Dynegy, and that the price drop on May 8 was caused solely by news that Project Alpha would be subject to a formal SEC investigation, the court is concerned that Graves' assignation of each day's price drop to a specific announcement is neat, but overly simplistic, because it fails to attribute any amount of the price drops or the loss of market confidence that caused them either to news that Dynegy had engaged in round-trip trades, or to the cumulative effect that the steadily increasing number of negative announcements on a steadily increasing number of different issues was likely to have had on the company's stock price. Accordingly, for the reasons that the court has already concluded that it is not possible to estimate with any degree of reasonable certainty the actual loss to shareholders caused by the corrective disclosures about Project Alpha made on April 25, 2002, the court concludes that it is not possible to estimate with any degree of reasonable certainty the actual loss to shareholders attributable to the corrective disclosures about Project Alpha made on May 8, 2002.
The court's conclusion that it is not possible to estimate with reasonable certainty the actual loss to shareholders attributable to corrective disclosures about Project Alpha is based on the facts of this case; it is not a conclusion that such estimates are never possible. Graves' regression analysis showed that only two abnormal return events — April 25-26 and May 8, 2002 — could possibly be linked to disclosures about Project Alpha. Since the disclosures about Project Alpha made during these events were confounded with other, non-fraud related bad news about Dynegy, Graves' estimate of the actual loss to shareholders caused by the defendant's unlawful conduct required more than a gross calculation of actual loss. See United States v. Ebbers, 458 F.3d 110, 128 (2d Cir. 2006) (reviewing types of complications that would "undoubtedly appear in a case where more than the grossest calculation [of actual loss] is needed," but concluding that was "not such a case."). Although Graves attempted to narrow his estimates as much as accepted methodologies would allow, the problems described above, and in particular the over four-fold difference between his low estimate of $161 and his high estimate of $714 million, persuade the court that the government has not shown to a reasonable degree of certainty the actual loss to shareholders caused by the Project Alpha-related disclosures made on April 25 and May 8, 2002.
2. Intended Loss
The government argues that "[i]f the court decides that the actual loss [to shareholders] caused by Olis' criminal acts is . . . indeterminable, the appropriate sentence must then be based upon the `intended loss' measure under U.S. Sentencing Guidelines Manual Application Note 2(A) (2001 ed.)." The government asserts that
Government's Rebuttal to Defendant's Expert Reports, Docket Entry No. 278, p. 9.
[i]t is uncontroverted that, by manipulating the cash flow treatment for Project Alpha, Olis sought to deprive the National Treasury of $79 million in taxes. Thus . . . this intended harm of $79 million to the United States would be the `loss' that determines his guidelines sentence.
Id.
Without disputing the $79 million amount of the intended loss argued by the government, the defendant argues only that "no intended tax loss has been proved." The court disagrees.
Mr. Olis' Brief on Intended Loss, Docket Entry No. 291, p. 1.
At defendant's trial, Gene Foster offered uncontroverted testimony that the tax benefit to Dynegy arising from Project Alpha could have been higher than the approximately $79 million that the company reported, but that he and Olis purposely reduced it so that the resulting "effective tax rate" would remain high enough to avoid scrutiny that could have revealed the truth about Project Alpha. Since Foster's uncontroverted testimony establishes that Olis expected the fraudulent accounting treatment accorded to Project Alpha to reduce the taxes that Dynegy would otherwise have owed to the United States Treasury by $79 million, the court concludes that $79 million is a reasonably certain estimate of the pecuniary harm that the defendant intended to result from his offense. See U.S. Sentencing Guidelines Manual § 2B1.1 cmt. n. 2(A)(ii) (2001).
See Government's Memorandum on Sentencing Loss, Docket Entry No. 194, p. 18 n. 28 (citing Trial Testimony of Gene Foster, Transcript Day Six, Docket Entry No. 103, pp. 218-24 and Government Trial Exhibit 793 quantifies the Project Alpha related tax benefit as $79 million). See also form 8-K filed by Dynegy, Inc. on April 25, 2002, stating that Project Alpha allowed Dynegy to realize a tax benefit of approximately $80 million in 2001.
B. Multiple Victim Enhancement
The parties agree that if the court uses the intended loss to the United States, rather than the actual loss, to calculate the defendant's offense level, the four-level enhancement for more than 50 victims pursuant to § 2B1.1(b)(2)(B) does not apply.See United States v. Leach, 417 F.3d 1099, 1106-07 (10th Cir. 2005). The parties also agree that because this issue has become newly relevant upon resentencing, the mandate rule does not preclude the court from considering it. See United States v. Lee, 358 F.3d 315, 320 (5th Cir. 2004). Because the court has concluded that intended loss, rather than actual loss, is more clearly established by the evidence, the court will not apply the multiple victim enhancement.
Application note 3(A)(ii) to section 2B1.1 defines "victim" as "(I) any person who sustained any part of the actual loss determined under subsection (b)(1); . . ."
C. Conclusion
For the reasons explained above, the court concludes the actual loss to shareholders caused by the corrective disclosures made about Project Alpha on April 25 and May 8, 2002, cannot be reasonably calculated based on the evidence now before the court. Since, however, the uncontroverted trial testimony of Gene Foster establishes that Olis expected the fraudulent accounting treatment accorded to Project Alpha to reduce the taxes that Dynegy would otherwise have owed to the United States Treasury by $79 million, the court concludes that the defendant's guideline sentencing range should be calculated on the basis of an intended loss to the United States Treasury of $79 million. Accordingly, the court concludes that the defendant's total offense level would be 34 yielding a guideline sentencing range of 151 to 188 months based on the following calculation:
Base Level 6 Intended Loss of $79 Million 24 Use of Special Skill 2 Use of Sophisticated Means 2 Number of Victims 0
Total Offense Level 34
III. The Reasonableness of a Guideline Sentence
As explained above, the court's flexibility to decide a fair sentence has significantly changed since Olis's first sentencing. Although the court must still consider the sentencing guidelines, since Booker the guidelines are "merely one sentencing factor among many, and the calculated guideline range must be considered in conjunction with the other § 3553(a) factors." United States v. Reinhart, 442 F.3d 857, 864 (5th Cir. 2006), pet. for cert. granted June 5, 2006 (No. 05-11431) (citing Booker, 125 S.Ct. at 756-757). The guidelines do not have "quasi-mandatory status."Id.
The federal sentencing statute states:
The court shall impose a sentence sufficient, but not greater than necessary, to comply with the purposes set forth in paragraph 2 of this subsection. The court, in determining the particular sentence to be imposed, shall consider—
(1) the nature and circumstances of the offense and the history and characteristics of the defendant;
(2) the need for the sentence imposed —
(A) to reflect the seriousness of the offense, to promote respect for the law, and to provide just punishment for the offense;
(B) to afford adequate deterrence to criminal conduct;
(C) to protect the public from further crimes of the defendant; and
(D) to provide the defendant with needed educational or vocational training, medical care, or other correctional treatment in the most effective manner;
(3) the kinds of sentences available;
(4) the kinds of sentence and the sentencing range established for . . . the applicable category of offense committed by the applicable category of defendant as set forth in the guidelines . . .;
(5) any pertinent policy statement . . . issued by the Sentencing Commission . . .;
(6) the need to avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct; and
(7) the need to provide restitution to any victims of the offense.18 U.S.C. § 3553(a).
To comply with the Fifth Circuit's instructions in United States v. Mares, 402 F.3d 511 (5th Cir.), cert. denied, 126 S.Ct. 43 (2005), the court, if it elects to impose a non-guideline sentence,
should carefully articulate the reasons [the court] concludes that the sentence [it] has selected is appropriate for that defendant. These reasons should be fact specific and include, for example, aggravating or mitigating circumstances relating to personal characteristics of the defendant, his offense conduct, his criminal history, relevant conduct or other facts specific to the case at hand which led the court to conclude that the sentence imposed was fair and reasonable.Mares, 402 F.3d at 519.
In part II of this opinion the court has calculated the defendant's guideline range as 151 to 188 months. This range is driven primarily by the $79,000,000, loss, which accounts for 24 of the 34 points used to calculate the defendant's total offense level under the guidelines. The court will now address the other statutory sentencing factors.
A. The Nature and Circumstances of the Offense
A district court is in a good position to assess the nature and circumstances of the offense and the defendant's history and characteristics. See Reinhart, 442 F.3d at 864 (relying in part on these factors to support a non-guidelines sentence). This observation is particularly apt in the present case for a number of reasons. Because the case was tried to a jury, the court heard evidence of the facts surrounding the defendant's role in the offense. The court has received a wealth of information about the defendant, his upbringing, education, employment, and family. Moreover, the court presided over two related civil actions arising primarily out of Project Alpha. See In re Dynegy, Inc. ERISA Litigation, 309 F.Supp.2d 861 (S.D. Tex. 2004), and In re Dynegy, Inc. Securities Litigation, 339 F.Supp.2d 804 (S.D. Tex. 2004). Both of those actions were ultimately settled with sizable payments to those who suffered losses because of Project Alpha.
In the ERISA action the settlement was $30,750,000. In the Securities action the settlement was $474,050,000 in cash and stock, most of which was paid by the Dynegy defendants.
Olis was found guilty of serious charges — securities fraud, mail and wire fraud, and conspiracy. The evidence at trial showed that Olis, along with two indicted and several unindicted co-conspirators, sought to address the stock market's concern about Dynegy's meager cash flow from operations by structuring a transaction — Project Alpha — that would appear to the company's outside auditors and to the investing public as if it resulted in income from operations when in fact it was a disguised loan. The conspiracy was intended to decrease Dynegy's tax liability, increase its cash flow, and enhance the views of outside investors about Dynegy's earnings, thereby resulting in a greater market price for its stock.
Although Olis was intimately involved in the conspiracy and in planning Project Alpha, he did not have the ultimate authority at Dynegy to approve Project Alpha, nor was he responsible for drafting the documents by which the conspiracy was carried out and concealed. Moreover, unlike some other recently publicized corporate fraud cases, the purpose of this conspiracy was not to defraud Dynegy or to enrich Olis. Nor did the conspiracy cause Dynegy to file for bankruptcy. Although Dynegy suffered a loss in its market capitalization after the true facts of Project Alpha became public and paid large amounts to settle the resulting civil cases, the company remains a viable entity. The initial success of Project Alpha brought Olis a promotion and stock options, but it did not result in substantial pecuniary benefits to him. Although these facts do not detract from the seriousness of the crime for which Olis was convicted, they mitigate against the type of harsh sentence that may be deserved in cases where the defendant's conduct enriched him at the company's detriment or brought about the downfall of the company.
B. The History and Characteristics of the Defendant
Olis was born in Korea, the son of a Korean woman and an American soldier, who abandoned Olis's mother before Olis was born. Olis moved with his mother to the United States when he was five years old. His mother struggled to support Olis and his sister with menial jobs until William Olis, a retired American soldier, began supporting the family, and later married Olis's mother and adopted Olis and his sister. Olis adjusted to the difficulties of his early life. He learned English, became a United States citizen, graduated from high school, and received a scholarship to St. Mary's University, from which he graduated with a degree in accounting. He later received a law degree, while also working as an accountant. Olis began working for Dynegy in 1998. Olis has no prior arrests or convictions, and according to the many letters the court has received from friends and former co-workers, has led an exemplary life, except for his participation in Project Alpha.
C. The Need for the Sentence to Reflect the Seriousness of the Offense, to Promote Respect for the Law, and to Provide Just Punishment for the Offense
Although a significant sentence of imprisonment is required to reflect the seriousness of this offense, to promote respect for law, and to provide just punishment, those goals can be achieved by a sentence that is less severe than one within the guideline range.
D. General Deterrence
After November 1, 2001, the guideline ranges for defendants convicted of serious fraud and related white-collar crimes were substantially increased. Since then the media has reported on lengthy sentences received by a number of white-collar defendants. The court is persuaded that the public in general, and potential white-collar criminals in particular, have become aware of the substantial risk of imprisonment for lengthy periods if they commit crimes of this nature. In cases where a corporate officer enriched himself at the expense of the corporation or where his conduct resulted in the demise of the corporation, a lengthy sentence of the magnitude contemplated under the guidelines might be necessary to provide general deterrence. In this case, however, such a lengthy sentence is not necessary to provide general deterrence.
E. Specific Deterrence
A lengthy sentence is not needed to deter Olis from future criminal conduct. Once Olis returns to his family, the chastening effect of years in prison, the attendant negative publicity, the loss of his job and accounting and law licenses, and the need to provide support for his family will provide adequate deterrence against any potential future criminal conduct.
F. Other Statutory Sentencing Factors
The factors set out in § 3553(a)(2)(D), (6) and (7) are not relevant in this case. Olis does not need educational or vocational training or medical care. Neither a lengthy sentence nor a shorter, non-guideline sentence is needed to avoid unwarranted sentencing disparities. The other two indicted co-conspirators, Foster and Sharkey, received lower, non-guideline sentences because they pleaded guilty and cooperated with the government. The disparity between these sentences and Olis's sentence is not "unwarranted" because lower sentences based on cooperation with the government were intended by Congress and the Sentencing Commission, United States v. Duhon, 440 F.3d 711, 720-21 (5th Cir. 2006). Other members of the conspiracy have not been indicted, and according to the government's Sentencing Memorandum, it is unlikely that they will be indicted. There is therefore no one with a similar record who engaged in similar conduct with whom to compare in sentencing Olis.
G. Conclusion
Having considered all of the factors mandated by Congress, including the sentencing guidelines and policy statements, the court concludes that a sentence within the applicable guideline range would not be reasonable, and that a non-guideline sentence of 72 months in prison is appropriate.
IV. Conclusion and Order
Pursuant to the Sentencing Reform Act of 1984, it is the judgment of the court that the defendant, Jamie Olis, is hereby committed to the custody of the Bureau of Prisons to be imprisoned for a term of seventy-two (72) months. To achieve this punishment the court imposes a term of sixty (60) months as to each of Counts 1, 3, 4, 5, and 6 and a term of imprisonment of seventy-two (72) months as to Count 2, all such terms to run concurrently, for a total term of imprisonment of seventy-two (72) months. The terms and conditions of supervised release, fine, and special assessment of the defendant's original sentence will remain the same.