Chatham-FINAL-SMF by xiuliliaofz


									  Playing with Post-Booker Fire: The Dangers of Increased
   Judicial Discretion in Federal White Collar Sentencing
                                           Daniel A. Chatham

   I. INTRODUCTION ........................................................................................................ 619
  II. BACKGROUND ......................................................................................................... 620
     A. History of the United States Sentencing Guidelines ........................................... 620
     B. Mandatory Minimum Sentences in the Federal System ...................................... 623
     C. Impact of Blakely and Booker Generally........................................................... 624
     D. The Impact of Booker on White Collar Crime Sentencing ................................. 626
     E. The Recent High-Profile White Collar Crime Boom .......................................... 627
          1. Martha Stewart .............................................................................................. 628
          2. HealthSouth ................................................................................................... 629
     F. Game Theory Principles..................................................................................... 632
 III. THE POST-BOOKER GAME ........................................................................................ 635
     A. Applying Game Theory to Post-Booker White Collar Sentencing and the
          Congressional Response ................................................................................. 635
     B. Explaining Results Outside the Scope of the Game ............................................ 636
IV. RECOMMENDATION ................................................................................................. 637
 V. CONCLUSION ........................................................................................................... 638

                                                  I. INTRODUCTION

     Federal criminal sentencing is currently at a crossroads. Following the Supreme
Court’s decision to render the federal sentencing guidelines merely advisory rather than
mandatory in United States v. Booker, 1 sentencing reform is once again a possibility. This
reform will most likely come in areas of great sentencing disparity or where the public
and Congress feel offenders are not punished harshly enough. This Note analyzes, in
game theory terms, 2 the possibility of federal sentencing reform in white collar crime
cases, specifically, the imposition of mandatory minimums for certain corporate crimes.
     This Note first discusses the implementation of mandatory sentencing guidelines,

      1. United States v. Booker and United States v. Fanfan, 543 U.S. 220 (2005) (holding that mandatory
sentencing guidelines violate the Sixth Amendment right to a jury and rendering the United States Sentencing
Guidelines merely advisory).
      2. Game theory is an economic model that examines how and why certain actors make decisions. See
infra Part II.F.
620                               The Journal of Corporation Law                                      [Spring

abolishment of parole, and popularity of federal mandatory minimum sentences as a
response to growing public concern over sentencing disparity and the perception that
violent criminals and drug criminals were not punished harshly enough or uniformly. Part
II also discusses the recent increase in high profile corporate crime and the subsequent
government crackdown on those offenses and draws parallels between the current white
collar crime atmosphere and the “war on drugs” of the 1980s. Part III applies a two-
period model of game theory to analyze the possibility of Congress imposing mandatory
minimum sentences for white collar criminals to compensate for increasing sentencing
disparity and the return to low sentences for corporate criminals. In one outcome of the
“game,” the increased judicial discretion following the Booker decision leads to reduced
white collar sentences, provoking Congress to respond by passing mandatory minimum
laws. In another outcome, judges forego using their increased discretion to decrease white
collar sentences except in extreme cases, essentially leaving the guideline system in
place, and Congress declines the option to impose mandatory minimum sentences for
white collar crimes.
      This Note recommends that federal judges opposed to sentencing guidelines should
not use their newfound Booker discretion to give drastically greater sentencing departures
or lower, non-guidelines sentences in non-extraordinary cases. This Note recommends
this approach for two primary reasons. First, appellate courts have frequently remanded
cases in which district courts have given huge departures in “non-extraordinary” cases for
resentencing. 3 Thus, the district court eventually, though begrudgingly, must sentence a
defendant to somewhere within or significantly closer to the guideline range in order to
survive “reasonableness” review. 4 Second, by routinely departing downward or giving
non-guidelines sentences in white collar cases, judges run the very real risk of provoking
a response from Congress in the form of mandatory minimum sentences for white collar
crimes, which would further decrease judicial sentencing discretion. Instead, this Note
recommends that the best way for federal judges to keep the increased discretion offered
by an advisory guideline regime is to exercise that discretion sparingly, making sure to
cite specific reasons for exercising that discretion when it is employed.

                                             II. BACKGROUND

                      A. History of the United States Sentencing Guidelines
      Until the passage of the Sentencing Reform Act (SRA) in 1984, federal judges had
relatively wide discretion in sentencing federal offenders up to the statutory maximum. 5

      3. The commentary notes to U.S. SENTENCING GUIDELINES MANUAL § 5K2.0 (2004) define
“extraordinary” cases as those that fall outside the “heartland” of cases covered by the different provisions of
the Guidelines. Factors that place a particular case outside the “heartland” of cases, so long as they are not
factors that the Commission has expressly forbidden a sentencing judge to consider, may be used as the basis
for a departure. Id.; see also Koon v. United States, 518 U.S. 81, 92-96 (1996) (describing the procedure for
departures based on “factors [that] the Commission has not been able to take into account fully in formulating
the guidelines”).
      4. See infra Parts II.C, II.E.2.
      5. See Kate Stith & Steve Y. Koh, The Politics of Sentencing Reform: The Legislative History of the
Federal Sentencing Guidelines, 28 WAKE FOREST L. REV. 223, 225 (1993) [hereinafter Stith & Koh]
2007]                               Playing with Post-Booker Fire                                             621

This judicial discretion led to disparity in the sentences of similarly situated offenders, a
problem decried by both Democrats and Republicans in Congress, particularly in white
collar cases where defendants typically received probation or a fine. 6 In the decade prior
to 1984, Congress considered several proposals for sentencing reform, including some
that proposed advisory guideline systems to be considered by sentencing judges as
merely one sentencing factor. 7 In 1984, Congress passed the SRA as part of the
Comprehensive Crime Control Act. 8 The SRA created an independent judicial agency,
the United States Sentencing Commission, to promulgate and oversee the new federal
sentencing guidelines. 9
      The sentencing guidelines established by the Commission are designed to: [(1)]
      incorporate the purposes of sentencing (i.e., just punishment, deterrence,
      incapacitation, and rehabilitation); [(2)] provide certainty and fairness in
      meeting the purposes of sentencing by avoiding unwarranted disparity among
      offenders with similar characteristics convicted of similar criminal conduct,
      while permitting sufficient judicial flexibility to take into account relevant
      aggravating and mitigating factors; [(3)] reflect, to the extent practicable,
      advancement in the knowledge of human behavior as it relates to the criminal
      justice process. 10
The SRA also abolished the indeterminate sentencing system as well as federal parole,
“eliminat[ing] the need for federal judges to [predict] future actions of the Parole
Commission.” 11 Among the principal targets for more serious penalties under the new
guideline system were white collar and violent repeat offenders. 12 One important feature
of the guidelines system adopted by Congress was its mandatory nature, which decreased
and structured the judiciary’s discretion within bounds set by Congress. 13 The first

(describing the evolution of federal criminal sentencing and parole prior to 1984); see also U.S. SENTENCING
COMM’N, AN OVERVIEW OF THE UNITED STATES SENTENCING COMMISSION (2005), available at (stating that “[b]efore guidelines were developed,
judges could give a defendant a sentence that ranged anywhere from probation to the maximum penalty for the
       6. See Stith & Koh, supra note 5, at 261-62 (noting that Edward Kennedy, D—Mass., one of the driving
forces behind federal sentencing reform, thought that the sentencing disparity in the federal system was a
“national disgrace,” while Strom Thurmond, R—S.Car., disliked that federal judges necessarily approached
each offense and offender “from the lenient perspective”).
       7. See id. at 244-45 (citing S. 1437, 95th Cong. (1977) as a proposal advocating an advisory guidelines
system); see also United States v. Booker and United States v. Fanfan, 543 U.S. 220, 293 n.12 (2005) (Stevens,
J., dissenting) (stating that “[i]ncidentally, the original version of S. 1437 looked much like the regime that the
Court has mandated today—it directed the sentencing judge to consider a variety of factors, only one of which
was the sentencing range established by the Guidelines”).
       8. Comprehensive Crime Control Act of 1984, Pub. L. No. 98-473, 98 Stat. 1837 (1984).
       9. U.S. SENTENCING COMM’N, supra note 5, at 1.
     10. Id.
     11. U.S. SENTENCING COMM’N, Special Report to the Congress: Mandatory Minimum Penalties in the
Federal Criminal Justice System at 16 (Aug. 1991), available at
[hereinafter Mandatory Minimum Penalties].
     12. U.S. SENTENCING COMM’N, supra note 5, at 2.
     13. See supra note 7 (noting that Justice Stevens’ dissent in Booker recognized that Congress rejected an
advisory guidelines scheme when formulating the Guidelines).
622                              The Journal of Corporation Law                                    [Spring

United States Sentencing Guidelines went into effect November 1, 1987. 14
     The two main components of the federal guidelines were the seriousness of the
offense and the defendant’s criminal history. 15 In applying the guidelines in each case,
the sentencing court first determined the appropriate base offense level for the offense of
conviction under Chapter Two (Offense Conduct). 16 The court then made any
adjustments to that level as warranted by factors detailed in Chapters Two and Three
(Adjustments). 17 Next, the court determined the defendant’s criminal history category
under Chapter Four (Criminal History and Criminal Livelihood). 18 Based on the total
calculated offense level after adjustments and the criminal history category, the court
determined the corresponding guideline range on the guideline range chart listed in Part
A in Chapter Five. 19 If, after determining the applicable guideline range, the court
believed that range did not adequately reflect the proper punishment for the specific
defendant, the court could depart upward or downward from the guideline range only for
reasons listed in Chapter Five, Section K. 20 The court then determined where within the
guideline range to sentence the defendant. However, if the guideline range determined by

     14. 28 U.S.C. § 994 (1988).
     15. U.S. SENTENCING GUIDELINES MANUAL § 1B1.1 (2004).
     16. Id. The “offense conduct” consists of the acts constituting the offense for which the defendant was
convicted, also known as “relevant conduct.” See id. § 1B1.3. The guidelines contained in Chapter Two of the
United States Sentencing Guidelines Manual provide scores for the defendant’s instant offense of conviction,
and correspond to the relevant criminal statutes contained in Appendix A. See generally id. § 1B1.2.
     17. Id. § 1B1.1(b), (c). Chapter Three provides for adjustments related to victims and based on the
defendant’s role in the offense and acceptance of responsibility. U.S. SENTENCING GUIDELINES MANUAL §
     18. Id. § 1B1.1(f). Criminal History points are generally assessed as follows: three points for a prior
sentence of imprisonment exceeding thirteen months; two points for a prior sentence of imprisonment
exceeding sixty days; and one point for each prior sentence not otherwise counted (maximum of four points
under the “catchall” provision). See id. § 4A1.1(a)-(c). Additional points may be added if the defendant
committed the instant offense within two years of a prior sentence or while on probation or parole. Id. §
4A1.1(d), (e). The sum of these points determines the defendant’s criminal history category on the Sentencing
Table located in Chapter Five. See id. § 4A1.1. The categories range from Category One (zero to one criminal
history point) to Category Six (thirteen or more criminal history points). See U.S. SENTENCING GUIDELINES
     19. Id. § 1B1.1(g). The following is an example of a portion of the Sentencing Table:

                              ILLUSTRATION 1: SENTENCING TABLE (EXCERPT)
                                   Criminal History Category (Criminal History Points)

              Offense          I           II         III        IV          V             VI
              Level         (0 or 1)    (2 or 3)    (4,5,6)    (7,8,9)   (10,11,12)      (13+)

                  10          6-12        8-14      10-16      15-21        21-27        24-30
                  11          8-14       10-16      12-18      18-24        24-30        27-33
                  12         10-16       12-18      15-21      21-27        27-33        30-37

    20. Id. §§ 1B1.1(h), 5G1.1(c). Chapter Five, Section K provides the valid bases under which a district
court may depart (upward or downward) from the guideline range. The most frequently used departure is the
substantial assistance motion listed under section 5K1.1, see infra note 91, but section 5K also provides for
several other departures, including a downward departure for a victim’s conduct in provoking the behavior
(5K2.10), and an upward departure for damage or loss not taken into account by relevant conduct (5K.2.5).
2007]                                Playing with Post-Booker Fire                                             623

the court was lower than the mandatory minimum or higher than the statutory maximum,
the statutorily authorized maximum or minimum sentence was the guideline sentence. 21
     While the sentencing guidelines were a major revision to the federal criminal system
in the 1980s, they were only a part of the federal sentencing overhaul. Around the same
time Congress passed and the Sentencing Commission implemented the Guidelines,
Congress began passing mandatory minimum sentence laws for certain violent and drug
offenses. 22 The next Part discusses federal mandatory minimum sentences and their
interaction with the United States Sentencing Guidelines.

                     B. Mandatory Minimum Sentences in the Federal System
      Mandatory minimum sentences have a long history in federal sentencing law,
though until the 1980s mandatory minimum sentences were quite low and typically
aimed at specific crimes, rather than an entire class of offenses. 23 The move for
sentencing reform in Congress coincided with the 1980s public outcry over drug crimes
and the “war on drugs.” 24 In response to this outcry, Congress passed several bills
implementing mandatory minimum sentences for drug and violent crimes, and proposed
many others, including mandatory minimums for some white collar offenses. 25 Many
federal judges abhorred the new strictures on their sentencing discretion and expressed
outrage with the Guidelines and mandatory minimum scheme. 26 There is little to suggest
that federal judges’ attitudes have changed toward the Guidelines in the ensuing years. 27

     21. Id. § 5G1.1(a), (b). Following a guilty plea or guilty verdict, the United States Probation Office
conducts a presentence investigation, which culminates in the preparation and presentation to the district court
of a presentence report. 18 U.S.C. § 3552 (2004); see also FED. R. CRIM. P. 32(d) (2004) (mandating that the
presentence report “identify all applicable guidelines, . . . calculate the defendant’s offense level and criminal
history category, . . . state the resulting sentencing range and types of sentences available,” and identify other
relevant factors and bases for departure from the applicable sentencing range). The facts in the presentence
report, in the absence of objections from either party, form the basis for the district court’s sentencing decision.
     22. See Mandatory Minimum Penalties, supra note 11, at 9 (stating that “[b]eginning in 1984, and every
two years thereafter, Congress enacted an array of mandatory minimum penalties specifically targeted at drugs
and violent crime”).
     23. See id. at 6-11 (discussing the history of mandatory minimums in the federal system since 1790,
including Congress’s retreat from comprehensive application of mandatory minimum sentences for drug crimes
in the Comprehensive Drug Abuse Prevention and Control Act of 1970). Appendix A of the Report provides a
historical list of federal mandatory minimum sentences. Notably, many offenses for which Congress imposed
mandatory minimum sentences prior to the 1980s were white collar offenses (e.g., securities violations, bribery,
embezzlement, fraud, and false entries by banking officer). Id. at app. A.
     24. See Franklin E. Zimring, Penal Policy and Penal Legislation in Recent American Experience, 58
STAN. L. REV. 323, 331-32 (2005) (describing penal drug legislation during the mid-1980s in response to the
proliferation of crack cocaine in inner cities).
     25. See Mandatory Minimum Penalties, supra note 11, at 9. One proposed mandatory minimum for a
white collar-type offense was H.R. 2090, which provided for “mandatory life imprisonment for an officer or
employee of a depository institution that conducts or attempts to conduct transaction(s) to launder drug money.”
Id. at B-94.
     26. See Don J. Benedictis, The Verdict is In: Throw Out Mandatory Minimum Sentences, Judges Tell ABA
Journal Poll, 79 A.B.A. J. 78 (1993) (reporting the results of a poll of judges as to the effectiveness of the
United States Sentencing Guidelines and mandatory minimums). In a 1993 poll conducted by the ABA, 90% of
federal judges believed that mandatory minimum sentences for drug cases were a bad idea, and 45% of federal
judges believed that the federal sentencing guidelines should be “scrapped.” Id.
     27. See generally Richard T. Boylan, Do the Sentencing Guidelines Influence the Retirement Decisions of
624                               The Journal of Corporation Law                                       [Spring

The next Part discusses various attempts to return some sentencing discretion to judges
through attacks on the constitutionality of the Guidelines.

                            C. Impact of Blakely and Booker Generally
     The constitutionality of the SRA has been challenged several times since its passage
in 1984. 28 The first major challenge came in Mistretta v. United States, 29 when the
Supreme Court upheld the constitutionality of the Commission against a challenge on the
basis of excessive legislative delegation and violation of separation of powers. In 2000,
however, the Supreme Court held in Apprendi v. New Jersey 30 that, “[o]ther than the fact
of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed
statutory maximum must be submitted to a jury, and proved beyond a reasonable
doubt.” 31 Because Apprendi involved the New Jersey state sentencing guidelines, the
Court, of course, did not speak directly to the federal sentencing guidelines in that
opinion. But the parallels between the federal guidelines and the New Jersey guidelines
caused many to believe that the decision could lead to the federal sentencing guidelines
being declared unconstitutional. 32 The Supreme Court, however, declined to extend
Apprendi until 2004 in Blakely v. Washington. 33
     In June 2004, the Supreme Court again considered the constitutionality of
sentencing guidelines in Blakely. Blakely involved a Washington state defendant
convicted of first degree kidnapping, whose state guideline sentencing range was forty-
nine to fifty-three months. 34 The sentencing judge, however, found that Blakely acted
with “‘deliberate cruelty,’ a statutorily enumerated ground for departure in domestic-
violence cases,” and departed upward to sentence Blakely to ninety months
imprisonment. 35 The Supreme Court held that sentencing a defendant above the
mandated guideline range on the basis of judge-found facts violated the defendant’s Sixth
Amendment right to trial by jury. 36 The Court interpreted the “statutory maximum”
language in Apprendi to mean “the maximum sentence a judge may impose solely on the
basis of facts reflected in the jury verdict or admitted by the defendant;” thus, the relevant
statutory maximum was the guideline maximum, absent upward departures based solely

Federal Judges?, 33 J. LEGAL STUD. 231 (2004) (noting that, since the passage of the Sentencing Reform Act
and mandatory minimums, federal judges have, on average, taken senior status 0.4 years after becoming eligible
for retirement, down from a three-year average prior to the crackdown on judicial sentencing discretion).
     28. See, e.g., Mistretta v. United States, 488 U.S. 361, 412 (1989) (rejecting a challenge to the Guidelines
and Sentencing Commission on non-delegation and separation-of-powers grounds); Apprendi v. New Jersey,
530 U.S. 466, 490 (2000) (holding that the enhancement of a defendant’s sentence above the statutory
maximum based solely on judge-found facts violates the Sixth Amendment).
     29. Mistretta, 488 U.S. at 412.
     30. Apprendi, 530 U.S. 466.
     31. Id. at 490.
     32. See generally Kyron Huigens, Solving the Apprendi Puzzle, 90 GEO. L.J. 387, 434-49 (2002)
(analyzing the post-Apprendi constitutional validity of the federal sentencing guidelines).
     33. Blakely v. Washington, 542 U.S. 296 (2004) (holding that the Washington state sentencing guidelines
violated a defendant’s Sixth Amendment right to counsel).
     34. Id. at 299.
     35. Id. at 300.
     36. Id. at 313.
2007]                               Playing with Post-Booker Fire                                             625

on judge-found facts. 37
     Because the Washington state sentencing guidelines were structurally very similar to
the federal sentencing guidelines, most believed that Blakely sounded the death knell for
upward departures under the United States Sentencing Guidelines as well, and perhaps
the Guidelines as a whole. 38 After Blakely, federal district courts began imposing
alternative sentences, a standard guidelines sentence if the federal sentencing guidelines
were found to be constitutional and a separate sentence for if the guidelines were found to
be unconstitutional. 39 While most district courts imposed identical sentences based on
the guidelines and 18 U.S.C. § 3553(a), 40 some judges took the opportunity to impose
significantly lighter alternative sentences based on the statutory factors. 41
     In January 2005, the Supreme Court addressed the constitutionality of the federal
sentencing guidelines in United States v. Booker and United States v. Fanfan. 42 The
guidelines range for the defendant in Booker, based on facts found by the jury at
defendant’s trial for possession of crack cocaine, was 210 to 262 months in prison; but at
sentencing the judge found, by a preponderance of the evidence, that the defendant
possessed a higher amount of crack cocaine and increased defendant’s guidelines range to
360 months to life imprisonment. 43 Justice Stevens, delivering the opinion for the Court
in part, held that the sentencing guidelines are subject to the jury trial requirements of the
Sixth Amendment, as construed in Blakely. 44 However, in the second part of the Court’s
opinion, authored by Justice Breyer, the Court held that the proper remedy for this
constitutional error was to sever the SRA’s provisions that made the guidelines
mandatory, rendering the federal sentencing guidelines effectively advisory. 45 After
Booker, district courts are required to consider the guideline range established in the same
fashion as before, but only as a factor along with the other sentencing goals listed in 18
U.S.C. § 3553(a). 46 Appellate courts are to review sentences for “unreasonableness.” 47

    37. Id. at 303 (emphasis omitted).
    38. See generally Alan Vinegrad & Jonathan Sack, ‘Blakely’: The End of the Sentencing Guidelines?,
N.Y. L.J., July 6, 2004, at 4 (noting that “the entire federal guidelines scheme is on precarious constitutional
    39. See, e.g., United States v. Hammoud, 378 F.3d 426 (4th Cir. 2004) (recommending, pending a
definitive ruling by the Supreme Court on the constitutionality of the federal sentencing guidelines, that district
courts announce at sentencing an alternative sentence pursuant to the statutory factors listed in 18 U.S.C. §
3553(a), treating the guidelines as advisory only).
    40. 18 U.S.C. § 3553(a) instructs sentencing judges, in crafting a defendant’s sentence, to consider, among
other things:
       (2) the need for the sentence imposed — (A) To reflect 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 from the defendant;
       and (D) To provide the defendant with needed educational and vocational training, medical care,
       or other correctional treatment in the most effective manner.
     41. See, e.g., United States v. Owens, 142 F. App’x 697, 698 (4th Cir. 2005) (vacating a district court’s
alternative, non-guidelines sentence of fifty months where the bottom of the properly calculated guideline range
was 108 months).
     42. United States v. Booker, 543 U.S. 220 (2005).
     43. Id. at 227.
     44. Id. at 231-33.
     45. Id. at 245, 258-61.
     46. Id. at 259-60.
626                              The Journal of Corporation Law                                    [Spring

      Though Booker rendered the sentencing guidelines merely advisory, the guidelines
still play a near-identical role in sentencing as they did pre-Booker. 48 One added wrinkle
to the sentencing picture is the emergence of the guideline “variance” (also known as a
non-guidelines sentence) in which the court finds that no Chapter Five departures 49 apply
to the defendant, but relying solely on the factors listed in 18 U.S.C. § 3553, sentences
the defendant to a non-guidelines term of imprisonment. 50 In determining any non-
guidelines sentence, the court must consider the advisory guideline range (along with any
applicable departures) in making its determination, but if it finds a compelling reason
under section 3553 to impose a sentence above or below the guidelines, it may now do
so. 51 According to the United States Sentencing Commission’s Special Post-Booker
Coding Project, as of February 22, 2006, 10.6% of post-Booker cases have involved
sentencing variances (1.3% above and 9.3% below the guideline range). 52 More
significant for purposes of this Note is the fact that 10.8% of post-Booker white collar
cases (cases sentenced under section 2B1.1—Theft and Fraud) have involved sentences
“otherwise below the range,” meaning that the district court granted a Booker variance or
some other non-government sponsored or guideline-authorized downward departure from
the guideline range. 53 This is the second-highest percentage of downward variances of
any guideline, with only the gun guideline being higher. 54 Moreover, the percentage by
which the court varied from the white collar guidelines was on average much higher than
other guidelines. 55

                  D. The Impact of Booker on White Collar Crime Sentencing
     In addition to violent and drug crimes, one of the primary areas of federal criminal
regulation, and the focus of this Note, is white collar crime. 56 “White collar crime”
includes approximately twenty-five crimes, but most notable for federal sentencing
guidelines purposes are fraud, embezzlement, tax evasion, antitrust offenses, and insider
trading. 57 One of the motivating factors behind Congress’s passage of the SRA was the

    47. Booker, 543 U.S. at 260-64 (finding the unreasonableness standard of review implicit in section
    48. See, e.g., United States v. Crosby, 397 F.3d 103, 113 (2d Cir. 2005) (holding that, in sentencing
defendants after Booker, a district court must still calculate the applicable Guidelines range, then decide
whether to impose a “non-Guidelines sentence” based on statutory factors).
    49. See supra note 20.
    50. See text of 18 U.S.C. § 3553(a), supra note 40.
    51. See, e.g., United States v. Wilson, 355 F. Supp. 2d 1269, 1272 (D. Utah 2005) (detailing the district
court’s approach to post-Booker sentencing variances).
    52. U.S. SENTENCING COMM’N, Final Report on the Impact of United States v. Booker on Federal
Sentencing, at D-5 (Feb. 22, 2006), available at
[hereinafter Final Booker Report].
    53. Id.
    54. Id. 9.7% of defendants sentenced under section 2D1.1 (drug trafficking) have received downward
variances, and 11.1% of defendants sentenced under section 2K2.1 (firearms) have received downward
variances. Id.
    55. Id. at D-24.
    56. See supra note 6 and accompanying text (explaining that judicial discretion has led to disparate
sentences among offenders who have committed comparable crimes); infra note 58 (same).
    57. The primary guidelines dealing with white collar offenses, and the focus of this Note, are U.S.
2007]                              Playing with Post-Booker Fire                                          627

light sentences given to white collar offenders, who typically received sentences of
probation or a fine. 58
     Approximately 11% of the sentences handed out after Booker have been under white
collar crime guidelines. 59 For the guidelines as a whole, sentences within the guideline
range have dropped from 69.4% in 2003 to 62.2% post-Booker. 60 Though departure
sentencing data for individual offenses are not specifically set out in the study, it is
probable that most of the downward departures have come in drug cases, as the drug
guidelines are much higher than white collar guidelines and are more universally decried
by district court judges. 61 However, with the recent burst of high-profile white collar
cases (e.g. Martha Stewart and HealthSouth), 62 along with the history of federal judges’
leniency toward white collar offenses, it is possible that judges may be more willing to
consider downward departures or even variances in white collar cases as well. 63

                     E. The Recent High-Profile White Collar Crime Boom
     White collar crime has been thrust into the news more frequently in the past five
years, with several major scandals making headlines. 64 The notoriety of these cases
places white collar crime on the public’s and Congress’s radar.65 The following

SENTENCING GUIDELINES MANUAL §§ 2B1.1 (fraud and embezzlement), 2B1.4 (insider trading), 2S1.1 (money
laundering), and 2S1.3 (structuring transactions to evade reporting requirements, etc.). Additionally, many
potential white collar defendants (see, e.g., Martha Stewart, infra Part II.E.1) are prosecuted for perjury or
obstruction of justice for making false statements to law enforcement or agency officials during the
investigation of the alleged wrongdoing and sentenced under U.S. SENTENCING GUIDELINES MANUAL § 2J1.2
(obstruction of justice) or § 2J1.3 (perjury).
     58. See Allan Ellis, Representing White Collar Clients in a Post-Booker World, CHAMPION, Sept.-Oct.
2005, at 12 (noting that the Sentencing Commission saw the probationary sentences for white-collar offenders
as a “‘problem,’ which it ‘solved’ with ‘guidelines that classify as serious many offenses for which probation
previously was frequently given and provide for at least a short period of imprisonment in such cases’”); see
also Steven Breyer, The Federal Sentencing Guidelines and the Key Compromises Upon Which They Rest, 17
HOFSTRA L. REV. 1, 22 (1988) (noting that the Sentencing Commission (of which Breyer was a member)
“considered present sentencing practices, where white-collar criminals receive probation more often than other
offenders who committed crimes of comparable severity, to be unfair”).
     59. Final Booker Report, supra note 52, at D-6.
     60. Id. at D-10.
     61. For an example of a district court judge’s animosity towards the drug and gun guidelines, see United
States v. Haack, 403 F.3d 997, 1001 (8th Cir. 2005) (“[W]hen are people going to wake up and do something
about these ridiculous United States Sentencing Guidelines?”) (quoting Chief United States District Court Judge
for the Northern District of Iowa, Mark W. Bennett).
     62. See infra Part II.E (discussing the Martha Stewart obstruction of justice case and the billion-dollar
HealthSouth fraud); see also United States v. Stewart, 433 F.3d 273 (2d Cir. 2006) (upholding Martha Stewart’s
conviction and sentence); United States v. Martin, 135 F. App’x 411, 412 (11th Cir. 2005) (unpublished
opinion) (reversing, as unjustified and unreasonable, a departure from a guidelines range of 108 to 135 months
to a sentence of probation for a former HealthSouth CFO); United States v. Botts, 135 F. App’x 416, 417 (11th
Cir. 2005) (unpublished opinion) (reversing, as unreasonable, a departure from a guidelines sentence of sixty
months to a sentence of probation for a former HealthSouth Senior Vice President for Tax).
     63. See infra Part II.E.2 (discussing the HealthSouth scandal and sentencings).
     64. See infra Parts II.E.1, II.E.2.
     65. See Corporate Fraud Task Force, First Year Report to the President by the Corporate Fraud Task
Force, 1478 PRAC. L. INST./CORP. 613, 624, 637-38 (2005) (stating that President Bush “[d]irected the
Sentencing Commission to review sentencing in white collar crime, obstruction of justice, securities,
628                               The Journal of Corporation Law                                    [Spring

discussion will look at two illustrations of how the federal courts have handled these high
profile cases.

                                           1. Martha Stewart
     Martha Stewart, chairperson of the board of directors and CEO of Martha Stewart
Living Omnimedia, was at the center of perhaps the highest profile of the recent white
collar scandals, though certainly not the most egregious. 66 In December 2001, Stewart
held almost 4,000 shares of stock in ImClone, a drug company founded by her friend
Sam Waksal. 67 ImClone had developed a cancer treatment drug called Erbitux that was
being considered for approval by the Food and Drug Administration (FDA). 68 On
December 27, 2001, Stewart sold all of her ImClone shares. 69 The next day, the FDA
announced its rejection of ImClone’s application for Erbitux, sending ImClone’s stock on
a sharp decline. 70 When prosecutors questioned Stewart regarding her timely stock sale,
she told them that she had a standing agreement with her stock broker, Peter Bacanovic,
to sell the stock once it reached a certain price. 71 Due to Stewart’s and Bacanovic’s
conflicting accounts of the transaction, federal prosecutors indicted Bacanovic and
Stewart for securities fraud, obstruction of justice, and conspiracy. 72 On February 27,
2004, the district court dismissed the securities fraud charges against Stewart, but on
March 5, 2004, a jury convicted Stewart of all the remaining charges and Baconovic of
four of the five counts of the indictment in which he was named. 73
     On July 16, 2004, Judge Miriam Goldman Cedarbaum sentenced Stewart under the
federal sentencing guidelines. 74 Stewart’s base offense level was twelve, 75 and no other
Chapter Two or Chapter Three adjustments applied. Stewart had no criminal history,
placing her in Criminal History Category I, for a guideline range of zero to sixteen
months’ imprisonment. 76 Despite the government’s recommendation that Stewart be
sentenced at the high end of the guideline range, Judge Cedarbaum sentenced Stewart to

accounting, and pension fraud cases,” and noting the subsequent increase in maximum penalties for white collar
crimes under the Sarbanes-Oxley Act).
     66. See Stewart Convicted on All Charges, CNN MONEY, Mar. 5, 2004, available at
     67. Id. (providing a timeline of the major occurrences in the Martha Stewart case).
     68. Id.
     69. Id.
     70. Id.
     71. See Stewart Convicted on All Charges, supra note 66.
     72. See generally Superseding Indictment, United States v. Martha Stewart & Peter Bacanovic, No. S1 03
Cr.             717             (MGC)              (S.D.N.Y.           2003),            available          at Stewart was not charged with insider
     73. See Stewart Convicted on All Charges, supra note 66; see also United States v. Stewart, 433 F.3d 273
(2nd Cir. 2006) (upholding Stewart’s conviction and sentence).
     74. See Stewart Convicted All Charges, supra note 66. Stewart’s sentencing took place after Blakely, but
before Booker, so the district court sentenced her under a mandatory guidelines scheme.
     75. See U.S. SENTENCING GUIDELINES MANUAL § 2J1.2(a) (2004).
     76. See id. ch. 5, pt. A (SENTENCING TABLE) (2004). The guideline range fell within Zone C, meaning that
Stewart was eligible to serve up to one-half of the minimum term of imprisonment in community confinement.
See id. § 5C1.1(c).
2007]                             Playing with Post-Booker Fire                                        629

the minimum sentence of imprisonment, ten months, with five months to be served in
prison and five months’ community confinement, two years’ supervised release, and a
$30,000 fine. 77 Due to the uncertainty of the guidelines system at that time, Stewart’s
attorneys requested that Stewart be given a non-guidelines sentence of probation, a
request the district court denied. 78 The Second Circuit later remanded the case to the
district court to determine whether it would have imposed a lower sentence under an
advisory guidelines scheme; however, the district court declined the invitation to modify
Stewart’s sentence. 79
      Martha Stewart’s case is an example of a low sentence under a straightforward
application of the guidelines. The light sentence reflects the nature of the charges of
which she was convicted and the relative harm caused by her conduct. 80 The next Part
discusses an example of individuals engaged in years of egregious conduct and a district
court with a much different view of the sentencing guidelines.

                                            2. HealthSouth
     Another recent major white collar case, though not as high profile as Martha
Stewart’s, exemplifies what has historically been judges’ attitudes toward white collar
criminals. 81 From 1996 to March 2003, a group of fourteen officers of the large health
care provider HealthSouth “conspired to artificially inflate HealthSouth’s reported
earnings, earnings per share, and overall financial condition.” 82 Overall, the accounting
fraud cost the corporation and its investors approximately $2.6 billion. 83
     Former Chief Financial Officer (CFO) Michael Martin and Senior Vice President for
Tax Richard Botts were among the participants in the fraud conspiracy.84 Martin falsified
earnings numbers to meet expectations and provided inaccurate reports to the SEC and
HealthSouth investors, and Botts knowingly submitted tax forms containing material,
false information about HealthSouth’s taxable income. 85 Martin resigned from
HealthSouth in 2000, after what he described as futile attempts to talk HealthSouth
founder and CEO Richard Scrushy into ending the fraudulent conduct.86 Both Martin and
Botts cooperated with the government, providing assistance “in helping the United States,

    77. See Brooke A. Masters, Martha Stewart Sentenced to Prison, WASH. POST, July 17, 2004, at A1.
    78. Id.
    79. See Memorandum Opinion at 2, United States v. Martha Stewart & Peter Baconovic, No. 03 Cr. 717
(MGC) (S.D.N.Y. 2005), available at
(noting Judge Cedarbaum was satisfied that “if the Sentencing Guidelines had been advisory at the time of
sentencing, [she] would have imposed the same sentence”).
    80. See Masters, supra note 77 (noting that Judge Cedarbaum stated at Stewart’s sentencing that the
“public interest objective in this prosecution has been served by the jury verdict”).
    81. See supra note 6 and accompanying text.
    82. United States v. Martin, 135 F. App’x 411, 412 (11th Cir. 2005) (remanding Martin’s case for re-
    83. See Michael Tomberlin, Ex-CFO to Serve Week in Prison, BIRMINGHAM NEWS, Sept. 21, 2005, at 1E.
    84. Id.
    85. Martin, 135 F. App’x at 412; United States v. Botts, 135 F. App’x 416, 417 (11th Cir. 2005).
    86. Martin, 135 F. App’x at 413. Former CEO and Chairman of the Board Richard Scrushy was charged
in an 85-count indictment. See In re HealthSouth Corp., 308 F. Supp. 2d 1253, 1259 (2005). On June 28, 2005,
a jury acquitted Scrushy on all counts. See Simon Romero & Kyle Whitmire, Former Chief of HealthSouth
Acquitted in $2.7 Billion Fraud, N.Y. TIMES, June 29, 2005, at A1.
630                               The Journal of Corporation Law                                       [Spring

HealthSouth, and the forensic auditor discover, in an expeditious manner, the varied
ways” in which the officers conducted the fraud. 87
      Martin pled guilty to: one count of conspiracy to commit securities fraud, mail
fraud, and falsification of books and records; one count of falsifying books, records, and
accounts; and one forfeiture count. 88 At the June 19, 2004, sentencing hearing in the U.S.
District Court for the Northern District of Alabama, Judge U.W. Clemon 89 adopted the
presentence report’s findings of a total offense level of thirty-one, Criminal History
Category I, resulting in a guidelines range of 108 to 135 months’ imprisonment. 90 Due to
Martin’s cooperation with the government, the government made a substantial assistance
motion under United States Sentencing Guidelines Manual section 5K1.1, and
recommended a downward departure to level twenty-five, resulting in a recommended
sentence of sixty-two months’ imprisonment. 91 The district court sentenced Martin to
sixty months’ probation with a special condition of six months’ home detention on
Counts One and Two, to run concurrently. 92
      Botts similarly pled guilty to one count of conspiracy to commit securities fraud,
falsification of books and records, and mail fraud, as well as to one forfeiture count. 93
The counts to which Botts pled guilty carried a statutory maximum of five years. 94 At
Botts’s June 2, 2004 sentencing hearing, Judge Clemon adopted the presentence report’s
findings of a total offense level of thirty-four, Criminal History Category I, for a United
States Sentencing Guidelines range of 151 to 188 months’ imprisonment; however, due
to the statutory maximum, the guideline range became sixty months. 95 Due to Botts’s
assistance in the investigation, the government made a substantial assistance motion
under section 5K1.1, recommending a departure to forty months’ imprisonment. 96 As he
did in Martin’s case, Judge Clemon utilized the discretion afforded him by the 5K1.1

     87. Martin, 135 F. App’x at 412; Botts, 135 F. App’x at 418. The Eleventh Circuit used nearly identical
language in remanding both cases (decided on the same day) back to the district court for resentencing
consistent with the circuit court’s opinion and Booker.
     88. Martin, 135 F. App’x at 412. Pursuant to the forfeiture count, Martin forfeited nearly $2.4 million. See
Tomberlin, supra note 83.
     89. Tomberlin, supra note 83.
     90. Martin, 135 F. App’x at 413.
     91. Id.; see also U.S. SENTENCING GUIDELINES MANUAL § 5K1.1 (2004) (allowing the district court, upon
motion by the government, to depart from the guidelines based on a defendant’s “substantial assistance in the
investigation or prosecution of another person who has committed an offense”). After the government makes a
section 5K1.1 motion, the district court has discretion to depart, but must do so based solely on factors
enumerated in section 5K1.1, and must take into consideration the government’s recommendation as to an
appropriate departure. Id. A motion under section 5K1.1, however, does not grant the district court the power to
depart below the statutory minimum. Id. For the judge to have the ability to depart below the statutory
minimum, the government must make, and the judge must grant, an additional motion for substantial assistance
under 18 U.S.C. § 3553(e).
     92. Martin, 135 F. App’x at 414.
     93. United States v. Botts, 135 F. App’x 416, 417 (11th Cir. 2005).
     94. See 18 U.S.C. § 371 (2005) (providing a statutory maximum of five years for conspiracy to defraud
the United States conviction).
     95. See Botts, 135 F. App’x at 417-18; U.S. SENTENCING GUIDELINES MANUAL § 5G1.1(a) (“Where the
statutorily authorized maximum sentence is less than the minimum of the applicable guideline range, the
statutorily authorized maximum sentence shall be the guideline sentence.”).
     96. Botts, 135 F. App’x at 418.
2007]                               Playing with Post-Booker Fire                                            631

motion to depart down to sixty months’ probation. 97
     The Eleventh Circuit reversed both sentences and remanded both cases for
resentencing because the district court failed to set out adequate reasons under section
5K1.1 for such an extraordinary departure. 98 On remand, Judge Clemon declined to
revise Botts’ sentence; 99 however, he reformed Martin’s sentence to seven days’ jail, six
months’ probation (with credit for the six months Martin had already served), along with
the previously-paid $50,000 fine and almost $2.4 million forfeiture. 100 In justifying the
99.8% departure from the bottom of Martin’s guideline range, Judge Clemon noted that,
while “[s]ome period of incarceration is necessary to reflect the seriousness of the crime,”
he did not feel that a longer prison term would be fair in light of Scrushy’s acquittal. 101
The government again appealed Martin’s sentence, and the Eleventh Circuit again
reversed, finding the sentence imposed was “patently unreasonable.” 102 Holding it was
“likely” that Judge Clemon “would have difficulty putting his previous views and
feelings aside,” the Eleventh Circuit ordered that the case be reassigned to a different
judge on remand. 103
     Overall, of the first seventeen officers charged with crimes in the $2.7 billion
HealthSouth scandal, only four were initially sentenced to jail time, the longest sentence
being twenty-seven months served by former CFO Weston Smith, who, not
coincidentally, was not sentenced by Judge Clemon, but by Judge Robert Propst.104 At
Smith’s sentencing, Judge Propst noted that others involved in the fraud had not received
as much prison time as Smith, which he attributed to other individuals being “wrongfully
acquitted or sentenced too lightly.” 105
     The low sentences handed out by Judge Clemon, and, more importantly, the judge’s
reasoning behind the sentences, indicate that not much has changed in at least some
federal judges’ views of white collar crime and criminals. 106 Some judges still see white
collar criminals as less worthy of jail time, despite Congress’s ratcheting up the
punishments for white collar crimes. 107 Low sentences still must pass reasonableness
(similar to “abuse of discretion”) review on appeal, and circuit courts have routinely
overturned huge departures by district courts that, in the appellate court’s view, did not

     97. Id.
     98. Id. at 420-21; United States v. Martin, 135 F. App’x 411, 415-16 (11th Cir. 2005).
     99. See Tomberlin, supra note 83, at 1.
   100. Id.
   101. Id.
   102. United States v. Martin (Martin II), 455 F.3d 1227, 1237 (11th Cir. 2006).
   103. Id. at 1242 (quoting United States v. Torkington, 874 F.2d 1441, 1447 (11th Cir. 1989)).
   104. Id. Strangely enough, Weston Smith was the final CFO during the fraud, and the first person to blow
the whistle on the fraud to authorities. See Michael Tomberlin, HealthSouth Whistleblower Smith Gets OK to
Start Prison During Appeal, BIRMINGHAM NEWS, Oct. 25, 2005, at 1A. The government made a 5K1.1
substantial assistance motion for Smith, and the district court departed to a sentence of twenty-seven months. Id.
   105. Id.
   106. See supra Part II.A (discussing the historically lenient attitude of federal judges toward white collar
   107. See United States v. Martin, 135 F. App’x 411 (11th Cir. 2005) (stating that the district court’s failure
to specify the reasons for extraordinary departures regarding prison terms necessitated remand for sentencing);
United States v. Botts, 135 F. App’x 416 (11th Cir. 2005) (stating a similar holding).
632                               The Journal of Corporation Law                                     [Spring

provide sufficient reasons for such large departures. 108 Part II.F of this Note discusses a
model that will help analyze the possible consequences of different approaches to white
collar sentencing by federal judges.

                                      F. Game Theory Principles
     To assess the possible effect of increasing judicial discretion in white collar
sentencing, this Note will apply a game theory analysis. Game theory is an economic
model that examines how and why certain actors make decisions, and is typically
described in its basic form with the “Prisoner’s Dilemma” scenario: 109
      Two prisoners are held in separate cells and they are not allowed to
      communicate with each other. Each is given the choice of confessing (and
      implicating the other prisoner) or refusing to confess. If both prisoners refuse to
      confess, they both go free. Otherwise, the prisoner who confesses first gets a
      short prison sentence and the nonconfessing (or second-to-confess) prisoner
      gets a lengthy prison sentence. Thus, even though each individual prisoner has
      an incentive to defect (to be the first to confess), and do it quickly, both
      prisoners are better off if they cooperate and refuse to confess. While the first
      to defect benefits himself, he benefits himself more by cooperating from the
      beginning with the other prisoner. Thus, mutual cooperation is the best strategy
      for both prisoners. 110
     This particular scenario of game theory assumes that the decision-makers are
rational, self-interested, and “take into account what they expect other rational self-
interested decisionmakers to do” in making their decision. 111 As noted by Pamela H.
Bucy, “[t]his assumption leads to a key concept in game theory, the Nash equilibrium,”
which “advises players in any game that, when choosing their strategies, they should take
into account what other players are likely to do.” 112 To analyze the possible outcomes of
the game as well as the likelihood of each outcome, this Note uses a similar methodology
to that applied by Linda S. Beres to civil contemnors: the two-period model with
“perfect” information, meaning that each actor in the game is familiar with the other
actor. 113

   108. See, e.g., United States v. Coyle, 429 F.3d 1192, 1192-94 (8th Cir. 2005) (noting, in the context of a
section 5K1.1 downward departure motion in a drug case, that a seventy-three percent sentence reduction for
what the circuit court viewed as non-extraordinary assistance to the government left little room for greater
departures in cases where defendants put themselves in danger to help the government, and therefore failed to
serve the goal of reducing sentence disparity between similarly situated defendants).
   109. See Pamela H. Bucy, Game Theory and the Civil False Claims Act: Iterated Games and Close-Knit
Groups, 35 LOY. U. CHI. L.J. 1021, 1027-28 (2004) (describing game theory, an economic model that
“simplifies social situations and offers insights from the simplification”).
   110. Id. at 1047 n.46 (internal citations omitted) (citing AVINASH K. DIXIT & BARRY J. NALEBUFF,
   111. Bucy, supra note 109, at 1047 n.46.
   112. Id. at 1028. The Nash equilibrium is a game theory concept developed by John Nash, the subject of the
Oscar-winning film, A BEAUTIFUL MIND (Universal Pictures 2001).
   113. See generally Linda S. Beres, Games Civil Contemnors Play, 18 HARV. J.L. & PUB. POL’Y 795 (1995)
(applying a two-period model of game theory to the scenario of a judge ordering an unwilling witness to testify
before a federal grand jury).
2007]                                 Playing with Post-Booker Fire                                              633

     The actors in this game are “Judges” and “Congress,” and in each iterance of the
game, Judges will act first and Congress second. 114 In fact, because the participants in
this game have specific prior experience in dealing with one another in a similar
situation—the passage of the SRA and mandatory minimum laws in the 1980s—the
actors have as perfect information as seems reasonably possible, making the outcomes
more predictable. 115 The predictability increases because the order of the game does not
change—judges will always move first, with Congress reacting. 116
     This game makes four basic assumptions: (1) federal judges do not like the
decreased discretion inherent in the sentencing guidelines system; 117 (2) Congress will
not act to curb judicial discretion unless it sees the need to do so because of an increase in
departures or variances; (3) if Congress acts to curb judicial discretion in the wake of
Booker, it will do so through mandatory minimums; and (4) if Judges deviate (use their
discretion to make huge departures or impose low, non-guidelines sentences in white
collar cases), Congress will react by imposing mandatory minimums. 118 Due to those
assumptions, there are two remaining possible outcomes for the game in this model:
      (1) Apply Guidelines/No Mandatory Minimums Imposed - Judges apply the
      guidelines in white collar cases under an advisory guidelines system largely in
      the same way as they did under a mandatory guidelines system and Congress
      does not intervene (represented by “AN”); and

    114. By referring to one of the game’s actors as “Judges” this author does not mean to imply that federal
judges are a homogenous group; however, this Note makes the assumption that those judges willing to grant, in
Congress’s view, unnecessarily large departures from the guidelines or downward variances in white collar
cases will be Congress’s target in any attempt to curb judicial discretion. Cf. H.R. REP. NO. 108-528, at 4, 7
(2004), as reprinted in 2004 U.S.C.C.A.N. 779, 780-83 (listing, in House Judiciary Committee report on statute
imposing a mandatory two-year consecutive sentence for identity theft, several cases in which identity thieves
received light sentences and noting Crime Subcommittee Chairman Howard Coble’s remark that “opponents of
mandatory minimums would have a more compelling case if they could assure the Congress that the judges
were faithfully following the Federal Sentencing Guidelines. And I think, sadly, there’s evidence that doesn’t
support that”). This Note assumes that since there is no other way, short of impeachment of individual
“deviating” judges, to avoid sentencing disparity in an advisory guidelines system in which some judges
constantly depart and others consistently follow the advisory guidelines, Congress will impose mandatory
minimums to regain at least some uniformity in white collar sentencing and to prevent the appearance that a
defendant’s sentence in the federal system depends entirely upon the district in which the defendant lives.
Therefore, in this game, the moniker “Judges” refers to those judges seeking to free themselves of the
constraints of the sentencing guidelines system and impose sentences they feel are reasonable based on the
circumstances of the specific defendant, and who would be the cause of any Congressional backlash.
    115. Beres, supra note 113, at 812. An additional factor in determining whether the participants in the game
have “perfect information” is the media, which provides a common information system in which each
participant’s preferences can be made known to the other. Bucy, supra note 109, at 1031-32.
    116. See Beres, supra note 113, at 812 (illustrating the effect of perfect information on the two-period
    117. See Benedictis, supra note 26; Boylan, supra note 27 (both commenting on federal judges’ responses
to the federal sentencing guidelines).
    118. Assumptions two and four are based on the model theory that the actors are rational and have “perfect
information.” As previously noted, the two-period model used in this Note assumes perfect information and that
the rational actors will only act in their own self-interest. If the actors do not have perfect information or act
other than in their own self-interest (irrationally), they fall outside the model. See infra Part III.A (discussing the
reasons for federal judges acting outside the scope of the model used in this Note).
634                               The Journal of Corporation Law                                      [Spring

      (2) Depart or Vary/Mandatory Minimums Imposed - Judges depart or vary
      from the guidelines frequently in white collar cases and Congress imposes
      mandatory minimums for white collar crimes (represented by “DM”). 119
It is also worthwhile to note the outcomes eliminated by the assumptions made in this
game that are now outside the scope of this model:
      (3) Apply Guidelines/Mandatory Minimums Imposed - Judges apply the
      guidelines in white collar cases under an advisory guidelines system largely in
      the same way as they did under a mandatory guidelines system but Congress
      still imposes mandatory minimums (represented by “AM”); 120 and
      (4) Depart or Vary/No Mandatory Minimums Imposed - Judges depart or vary
      from the guidelines frequently in white collar cases but Congress does not
      impose mandatory minimums (represented by “DN”). 121
      Though they are outside the scope of the game, results AM and DN are useful to
show the utility of every possible outcome to each actor in the game. The table in
Illustration 2 summarizes each actor’s preferences and assigns each possible outcome a
“utility score” for each actor. A utility of three represents each party’s best outcome, two
represents its second-best outcome, one represents its second-worst outcome, and zero
represents its worst outcome.
                                        ILLUSTRATION 2 122
                                        Judges                 Congress

                   Preferences        DN>AN>DM>AM                   AN>AM>DM>DN

                        AN                        2                            3

                       (AM)                       0                            2

                        DM                        1                            1

                       (DN)                       3                            0

   119. See Beres, supra note 113, at 810-11 (discussing the classification of witnesses and judges in a two-
period model of civil contempt).
   120. Outcome AM is outside the scope of the model because judges applying the guidelines would not be
able to avoid the undesired result of Congress passing mandatory minimums for white collar crimes. In game
theory terms, outcome AM would be the result of an irrational act by Congress (as Congress presumably would
not go to the expense of overhauling an advisory system that functions almost exactly the same fashion as its
intended mandatory system), so it falls outside the scope of the model. See infra note 138 and accompanying
text (discussing irrationality in the game theory context).
   121. Outcome DN is also outside the scope of the model because it would also involve an irrational act by
Congress. This outcome is irrational because it violates Congress’s stated goals of sentencing uniformity and
higher white collar sentences. See supra Part II.A (discussing the history of the United States sentencing
guidelines); see also supra note 6 and accompanying text (discussing the disparity in federal judges’ sentencing
   122. The format of this table follows that of the two-period model used in Beres, supra note 113, at 811.
2007]                             Playing with Post-Booker Fire                                         635

     Therefore, the outcomes most and least preferred by Judges are outside the scope of
this model because of assumptions two and four. The next Part analyzes the expected
outcome of the game. It also discusses the implications of the outcomes outside the scope
of the game and posits reasons for such outcomes.

                                    III. THE POST-BOOKER GAME

        A. Applying Game Theory to Post-Booker White Collar Sentencing and the
                              Congressional Response
     Despite the return of greater discretion in sentencing after Booker, most judges have
continued to sentence defendants within the guideline range as they did before Booker. 123
Under the two-period model with perfect information, this is the expected outcome. The
following matrix depicts the game as set out in the previous sections [Judges, Congress]:

                                             ILLUSTRATION 3 124
                                        No Mandatory Minimums                 Mandatory Minimums
                                              Imposed                              Imposed

 Judges Apply Guidelines 125                          2,3                                   -- 126

        Judges Deviate 127                              -- 128                            1,1

      Thus, both actors derive the greatest possible utility by Judges applying the
guidelines in largely the same fashion as before Booker. Similarly, both actors get the
least utility if the judges deviate, or vary. Due to the assumptions of the game, the only
outcome if judges apply the guidelines is the maintenance of the status quo of judicial
discretion, and the only outcome if judges deviate is the imposition of mandatory
minimums and the further deterioration of judicial sentencing discretion. Thus, the game
shows that the only rational choice for judges wanting to maintain their Booker discretion
in white collar cases is to continue to apply the guidelines as they have done since their
inception. Despite the model’s conclusion that the best outcome is for judges to apply the
guidelines as they did before Booker, judges continue to grant huge departures and use

   123. See Final Booker Report, supra note 52, at E-1 (noting 83% of defendants sentenced under the fraud
guideline (§ 2B1.1) have been sentenced within the guideline range or received departures based on government
   124. The format of this table follows that of the two-period model used in Beres, supra note 113, at 812.
   125. Judges apply the guidelines largely in the same fashion as before Booker.
   126. Outcome “DN” (Depart or Vary/No Mandatory Minimums Imposed) is outside the scope of the
model. See supra Part II.F (discussing game theory principles).
   127. Judges exercise their discretion to depart downward to a great degree or vary downward.
   128. Outcome “AM” (Apply Guidelines/Mandatory Minimums Imposed) is outside the scope of the model.
See supra Part II.F (discussing game theory principles).
636                               The Journal of Corporation Law                                     [Spring

sentencing variances in white collar cases. 129 Reasons for these seemingly anomalous
results are discussed in the following Part.

                      B. Explaining Results Outside the Scope of the Game
     Sentencing enhancements for fraud and other white collar crimes are based on the
amount of loss caused by the defendant or the conspiracy and the number of victims
harmed by the defendant’s conduct. 130 By designing the white collar guidelines in this
fashion, Congress clearly intended for white collar defendants who cause the most
financial harm to spend the most time in prison. 131 Despite that clear intent, however,
some judges continue to sentence white collar defendants involved in fraudulent conduct
causing massive amounts of financial damage to probation or only a few months in
prison. 132 With the very real danger of a Congressional backlash it is difficult to
understand why judges would risk their sentencing discretion in all white collar cases by
departing or granting downward variances in nearly every white collar case on their
docket. Using the terms of the above game (which assumes rational, self-interested
actors) there are two explanations for judges who deviate from the acceptable outcomes
of that game: (1) lack of information and (2) irrationality. 133
     Lack of information is the first possible, though unlikely, reason for an individual
judge’s deviation from the game. 134 The “perfect information” assumed in this game is
premised on the idea that judges and Congress have been down a similar road in the
recent past, with resulting mandatory minimums for drug and gun crimes. 135 Thus, any
judge who was not a judge or did not practice in the federal system prior to
implementation of the sentencing guidelines and mandatory minimums may not know the
ground on which they tread in imposing light sentences for white collar defendants. This
reason seems improbable, however, as every federal judge knows that the cornerstone of
determinate sentencing is decreasing sentencing disparity for similarly situated
defendants. 136 Individual judges departing or varying from the guidelines due to their
own preferences is exactly the evil aimed at by Congress in implementing the

   129. See supra Part II.E.2 (discussing sentencings in the HealthSouth scandal).
   130. See U.S. SENTENCING GUIDELINES MANUAL § 2B1.1(b)(1) (2004) (LOSS TABLE). The Loss Table
provides for increases in offense level based on amount of loss greater than $5000. Id. For example, if a
defendant caused more than $5000 but less than $10,000 loss, that defendant would receive a two-level
increase. Id. The most a defendant’s sentence may be increased solely based on the amount of loss is a 30-level
increase for causing more than $400,000,000 loss. Id.
   131. For further discussion of the deterrence and retributive merits of a white collar sentencing scheme
driven by amount of loss, see generally Frank O. Bowman, III, Coping With “Loss:” A Re-Examination of
Sentencing Federal Economic Crimes Under the Guidelines, 51 VAND. L. REV. 461 (1998) (discussing the
“Role of Sentencing Purpose in Sentencing the Economic Criminal”).
   132. See supra Part II.E.2 (discussing the HealthSouth scandal).
   133. See Beres, supra note 113, at 817-18 (discussing the two-period model with perfect information).
   134. A game with imperfect information is vividly illustrated in the “Prisoner’s Dilemma” paradigm. The
prisoners have the dilemma because they do not have perfect information—neither knows whether the other
prisoner decided to cooperate with police. See supra note 115 and accompanying text (discussing the model’s
assumption of perfect information).
   135. See id.
   136. See supra note 6 and accompanying text (noting that reducing sentencing disparity was one of the
primary reasons for Congress adopting the sentencing guidelines system).
2007]                               Playing with Post-Booker Fire                                             637

guidelines. 137 Therefore, the more likely reason for judges deviating is irrationality.
     Irrationality in this game does not mean that the judge is not a rational person; it
merely means that the judge is not acting self-interestedly. 138 Thus, “self-interest” refers
merely to the judge wanting to maintain post-Booker discretion, not that the judge has
some personal interest in the sentence for any given defendant. This problem is most
likely a product of short-sighted reasoning: the judge sees a defendant before the court
whom the judge feels does not deserve to spend time in prison. However, in utilizing (or
in Congress’s view, abusing) the discretion bestowed by Booker, the judge threatens the
very existence of that discretion. As the game shows, the only way for rational, self-
interested judges to keep their discretion to sentence some low-guideline white collar
defendants to probation is to exercise some restraint in sentencing higher-guideline
defendants. 139 Judges who fail to see the long-term implications in sentencing individual
defendants do not act “rationally” and may antagonize Congress into imposing
mandatory minimum sentences for white collar crimes to once again restore sentencing
uniformity in the federal system.

                                          IV. RECOMMENDATION

     Federal district courts sentencing white collar defendants in the post-Booker world
potentially have the power to control the destiny of their discretion. 140 As shown by the
application of game theory principles in this context, if both Congress and federal judges
act rationally and in their own self-interest, the only post-Booker outcome is the
maintenance of the status quo—an advisory guideline system in which judges may give a
reasonable non-guidelines sentence in situations where it was not possible under a
mandatory guidelines system. 141 Sentences like those imposed upon Michael Martin and
Richard Botts, 142 which included guidelines imprisonment departures of 99.8 and 100%,
respectively, are the type of sentences that are sure to attract the attention of Congress,
particularly because it sends the message that some federal judges harbor the same ideas
that prompted the initial imposition of mandatory guidelines. 143 Because mandatory
guidelines are no longer an available avenue to curb judicial discretion, the most logical
direction for Congress to turn is toward an already-popular alternative method for

   137. See id.
   138. See George Norman & Joel P. Trachtman, The Customary International Law Game, 99 AM. J. INT’L L.
541, 541 (2005) (stating that “[i]n a rationalist model, behavior is assumed to be motivated by self-interest”);
Adrian Vermuele, The Constitutional Law of Congressional Procedure, 71 U. CHI. L. REV. 361, 370 (2004)
(noting that actors may act based on “rational self-interest or irrational passions”).
   139. See supra Part III.A, ILLUSTRATION TWO.
   140. See supra Part III.A.
   141. Id.
   142. See supra Part II.E.2. Had there been some mandatory minimum sentence in place (for instance, two
years) at the time of the HealthSouth sentencings, Judge Clemon would not have been able to depart down to
probation for Martin and Botts, even though the government made a substantial assistance motion under U.S.
SENTENCING GUIDELINES MANUAL § 5K1.1 (2004). See supra note 91. By requiring an additional motion by
the government for the judge to have the discretion to depart below mandatory minimum sentences in white
collar cases (as is currently the case in drug and gun crimes), Congress would give nearly all sentencing
discretion to the prosecutor in determining departures for substantial assistance in white collar crimes, as it has
done for drug and violent crimes with mandatory minimums. Id.
   143. See supra note 6.
638                             The Journal of Corporation Law                                   [Spring

decreasing judicial discretion—mandatory minimum sentences.
     A better alternative is for district courts to utilize their discretion to vary from the
sentencing guidelines sparingly, saving variances and large departures for cases in which
they are best able to justify to a reviewing court that a variance was reasonable in light of
the statutory sentencing factors. 144 For example, it was possible for the district court in
the HealthSouth cases to depart lower than the sentences recommended by the
government without departing all the way to probation. 145 Under this “tempered
discretion” approach, Judge Clemon could have voiced his displeasure with the
government’s recommended departure by sentencing Martin to thirty months, half of the
sentence recommended by the government. 146 This sentence would have been roughly
equal to the sentence imposed upon a similarly situated defendant, Winston Smith, and
probably would have survived Booker reasonableness review. 147 Moreover, by
maintaining sentence uniformity with immediately similarly situated defendants, the
court would have a lower chance of drawing attention to “problems” with white collar
sentencing that might require a Congressional fix. This “tempered discretion” alternative
avoids incurring the wrath of Congress and the further deterioration of judicial sentencing
discretion in white collar cases.
     The approach taken by Judge Cedarbaum in Martha Stewart’s sentencing and
remand is an example of a judge using the advisory guidelines in a similar fashion as the
mandatory guidelines, but still giving a low sentence. 148 By relying on the guidelines
sentence, Judge Cedarbaum maintained the sentencing uniformity so craved by Congress,
while giving Stewart a relatively light sentence that included community confinement
rather than prison for half of the sentence. 149 Even though Stewart’s guidelines range was
very low, had Judge Cedarbaum varied downward and sentenced Stewart to probation in
such a high-profile case, Congress would have gained momentum for a push toward
mandatory minimums in white collar cases to show the public that high-profile white
collar criminals are not above the law. As the application of game theory shows, federal
judges abusing their limited Booker discretion to sentence white collar criminals to
disproportionately light sentences only serves to eliminate nearly all judicial sentencing
discretion in the long run. By applying the guidelines largely in the same fashion as
before Booker, federal judges dictate the best possible outcome for all actors in the game
and avoid painting themselves into a mandatory minimum corner.

                                           V. CONCLUSION

     The Supreme Court’s decision in United States v. Booker opened new avenues for
judicial discretion in federal criminal sentencing. However, in light of the original
purposes for the federal sentencing guidelines and mandatory minimum statutes of the
late 1980s and early 1990s—in particular, reduction of sentencing disparity caused by

  144. See supra Part II.C and note 40 (listing the statutory sentencing factors in 18 U.S.C. § 3553).
  145. See supra note 91 (discussing the sentencing judge’s discretion after granting a downward departure
  146. See supra Part II.E.2 (discussing the HealthSouth scandal).
  147. See supra Part II.C and note 47 (discussing the Booker reasonableness review standard).
  148. See supra Part II.E.1 (discussing the Martha Stewart trial and sentencing).
  149. Id.
2007]                        Playing with Post-Booker Fire                             639

near-total judicial discretion—Congress is wary of wide judicial sentencing discretion.
Though there are presently no mandatory minimum sentences for federal white collar
crimes, Congress need only point to a few instances of district courts returning to the
view that white collar defendants do not deserve jail time to justify imposing mandatory
minimums for white collar crimes as well. The recent wave of high-profile white collar
cases serves to draw even greater attention to the issue. Sentences of probation for
defendants convicted of participating in multi-billion dollar frauds may be just the poster
children Congress needs to start down the mandatory minimum road. The only way for
judges to achieve their best outcome in the game is to realize that they are playing—
before it is too late.

To top