Inter-judge sentencing disparity after Booker: A first look

Stanford Law Review

Volumn 63, Issue 1, 2010, Pages 1-67

  Scott, Ryan W ^a  

^a INDIANA UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 78650410159     PISSN: 00389765     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Note

References (268)

1. 543 U.S. 220 (2005).
2. 543 U.S. 220 (2005) at 245 (Breyer, J., delivering the opinion of the Court in part).
3. 552 U.S. 38 (2007).
4. 552 U.S. 38 (2007) at 52-53.
5. 552 U.S. 85 (2007).
6. 552 U.S. 85 (2007) at 101 (internal quotation marks omitted).
7. Memorandum from Jonathan J. Wroblewski, Dir., Office of Policy & Legislation, U.S. Dep't of Justice, to Hon. William K. Sessions III, U.S. Sentencing Comm'n 2 (June 28, 2010) [hereinafter Wroblewski Memorandum], available at http://sentencing.typepad.com/ files/annual_letter_2010_final_062810.pdf.
8. Eric Holder, U.S. Att'y Gen., Remarks for the Charles Hamilton Houston Institute for Race and Justice and Congressional Black Caucus Symposium: Rethinking Federal Sentencing Policy, 25th Anniversary of the Sentencing Reform Act (June 24, 2009) (transcript available at http://www.usdoj.gov/ag/speeches/2009/ag-speech-0906241.html).
9. See infra notes 113-117 and accompanying text.
10. Patrick J. Fitzgerald, U.S. Att'y, N. Dist. Ill., Statement Before the U.S. Sentencing Commission in the Regional Hearing on the State of Federal Sentencing 3 (Sept. 10, 2009) [hereinafter Fitzgerald Statement] (transcript available at http://www.ussc.gov/ AGENDAS/20090909/Fitzgeraldtestimony.pdf).
11. Karin J. Immergut, U.S. Att'y, Dist. Or., Statement Before the U.S. Sentencing Commission in the Regional Hearing on the State of Federal Sentencing 12 (May 27, 2009) [hereinafter Immergut Statement] (transcript available at http://www.ussc.gov/AGENDAS/20090527/Immergut_testimony.pdf).
12. Frank O. Bowman, III, Debacle: How the Supreme Court Has Mangled American Sentencing Law and How It Might yet Be Mended, 77 U. CHI. L. REV. 367, 368 (2010).
13. Amir Efrati, Looser Rules on Sentencing Stir Concerns About Equity, WALL ST. J., Nov. 5, 2009, at A15.
14. See infra text accompanying note 175.
15. See infra Figure 5 and accompanying text.
16. See infra notes 180-182 and accompanying text.
17. See infra Table 3 and accompanying text.
18. KATE STITH & JOSÉ A. CABRANES, FEAR OF JUDGING: SENTENCING GUIDELINES IN THE FEDERAL COURTS 9-11 (1998). Parole boards added an additional layer of indeterminacy to federal sentences. The Sentencing Reform Act of 1984 abolished parole in the federal system.
19. KATE STITH & JOSÉ A. CABRANES, FEAR OF JUDGING: SENTENCING GUIDELINES IN THE FEDERAL COURTS 9-11 (1998). Parole boards added an additional layer of indeterminacy to federal sentences. The Sentencing Reform Act of 1984 abolished parole in the federal system. at 11. The federal bank robbery statute, for example, provided that an offender "shall be fined not more than $5,000 or imprisoned not more than twenty years, or both." Bank Robbery Act of 1934, Pub. L. No. 73-235, § 2(a), 48 Stat. 783, 783 (current version at 18 U.S.C. § 2113 (2006)); see also Jerome v. United States, 318 U.S. 101, 101-02 (1943).
20. Kevin R. Reitz, Sentencing, in THE HANDBOOK OF CRIME AND PUNISHMENT 542, 543 (Michael Tonry ed., 1998).
21. STITH & CABRANES, supra note 18, at 9 & 197 n.3.
22. United States v. Mueffelman, 327 F. Supp. 2d 79, 83 (D. Mass. 2004) (Gertner, J.); see also Douglas A. Berman, Conceptualizing Booker, 38 ARIZ. ST. L.J. 387, 389 (2006).
23. See James M. Anderson et al., Measuring Interjudge Sentencing Disparity: Before and After the Federal Sentencing Guidelines, 42 J.L. & ECON. 271, 274 (1999) (defining "disparity" as "solely that variation caused by the identity of the decision maker"). What counts as a "legitimate" difference between cases justifying a higher or lower sentence is, of course, heavily contested and dependent on some underlying theory of punishment. See Kevin Cole, The Empty Idea of Sentencing Disparity, 91 NW. U. L. REV. 1336, 1337 (1997); Michael M. O'Hear, The Original Intent of Uniformity in Federal Sentencing, 74 U. CIN. L. REV. 749, 749-50 (2006). But Congress concluded that inter-judge disparity, driven by judicial preferences and biases rather than offense and offender characteristics, is unwarranted. See 18 U.S.C. § 3553(a)(6) (2006) (directing judges to impose sentences so as "to avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct"); S. REP. NO. 98-225, at 45 (1983) ("Sentencing disparities that are not justified by differences among offenses or offenders are unfair both to offenders and to the public.").
24. E.g., Norval Morris, Towards Principled Sentencing, 37 MD. L. REV. 267, 274 (1977).
25. See ANTHONY PARTRIDGE & WILLIAM B. ELDRIDGE, FED. JUDICIAL CTR., THE SECOND CIRCUIT SENTENCING STUDY: A REPORT TO THE JUDGES OF THE SECOND CIRCUIT 36 (1974); Kevin Clancy et al., Sentence Decisionmaking: The Logic of Sentence Decisions and the Extent and Sources of Sentence Disparity, 72 J. CRIM. L. & CRIMINOLOGY 524, 525-26 (1981); Shari Seidman Diamond & Hans Zeisel, Sentencing Councils: A Study of Sentence Disparity and Its Reduction, 43 U. CHI. L. REV. 109, 119-24 (1975) (analyzing the recommendations of "sentencing councils" in which panels of judges not assigned to a case would review the file and choose a sentence independently, then consult with the sentencing judge).
26. MARVIN E. FRANKEL, CRIMINAL SENTENCES: LAW WITHOUT ORDER 5 (1973).
27. MARVIN E. FRANKEL, CRIMINAL SENTENCES: LAW WITHOUT ORDER 5 (1973). at 10.
28. Wroblewski Memorandum, supra note 7, at 2.
29. Diamond & Zeisel, supra note 25, at 110-11 (quoting J. BENNETT, OF PRISONS AND JUSTICE, S. DOC. NO. 88-70, at 319 (1964)).
30. Stephen J. Schulhofer & Ilene H. Nagel, Negotiated Pleas Under the Federal Sentencing Guidelines: The First Fifteen Months, 27 AM. CRIM. L. REV. 231, 237 (1989) ("These disparities not only fostered undue optimism among offenders who hoped to 'beat the rap,' they also undermined deterrence and crime control objectives.").
31. Pub. L. No. 98-473, 98 Stat. 1837 (codified as amended in scattered sections of 18 and 28 U.S.C.).
32. U.S. SENTENCING COMM'N, SENTENCING GUIDELINES AND POLICY STATEMENTS 1.2 (1987); Stephen Breyer, The Federal Sentencing Guidelines and the Key Compromises upon Which They Rest, 17 HOFSTRA L. REV. 1, 4 (1988); see also Susan R. Klein & Jordan M. Steiker, The Search for Equality in Criminal Sentencing, 2002 SUP. CT. REV. 223, 232-33 (describing "the reduction of unwarranted disparity in sentencing" as "Congress's stated goal" in sentencing reform); Ilene H. Nagel, Structuring Sentencing Discretion: The New Federal Sentencing Guidelines, 80 J. CRIM. L. & CRIMINOLOGY 883, 895-99 (1990).
33. U.S. SENTENCING GUIDELINES MANUAL § 1A1.3 introductory cmt. (1987) ("Congress sought reasonable uniformity in sentencing by narrowing the wide disparity in sentences imposed for similar criminal offenses committed by similar offenders.").
34. STITH & CABRANES, supra note 18, at 38-48.
35. The Act was co-sponsored by strange bedfellows in the Senate: Ted Kennedy and Strom Thurmond. Id. at 38-39.
36. 28 U.S.C. § 991(a) (2006); see also id. §§ 994, 995(a)(1). The Commission's composition and location "in the judicial branch" are unusual, and scores of federal courts struck down the Act as unconstitutional before the Supreme Court rejected a separation of powers challenge in Mistretta v. United States, 488 U.S. 361, 380-412 (1989).
37. 28 U.S.C. § 994(a) (2006).
38. 28 U.S.C. § 994(a) (2006). § 994(p).
39. U.S. SENTENCING GUIDELINES MANUAL § 5K1.1 (2009); see also 28 U.S.C. § 994(n) (2006).
40. U.S. SENTENCING GUIDELINES MANUAL § 5K2.0 (2009).
41. 18 U.S.C. § 3553(b)(1) (2006). This was one of the provisions excised by the remedial opinion in Booker. See infra Part I.B.
42. 18 U.S.C. § 3553(c)(1)-(2) (2006).
43. 18 U.S.C. § 3553(c)(1)-(2) (2006). § 3742(a)-(b). It was not until 1996 that the Supreme Court clarified that the standard of appellate review was "abuse of discretion." See Koon v. United States, 518 U.S. 81, 92-100 (1996). The Act's appellate review provision was excised in Booker. See infra Part I.B.
44. See STITH & CABRANES, supra note 18, at 59-64; Paul G. Cassell, Too Severe? A Defense of the Federal Sentencing Guidelines (and a Critique of Federal Mandatory Minimums), 56 STAN. L. REV. 1017, 1018 (2004).
45. See, e.g., Vincent L. Broderick, The Importance of Flexibility in Sentencing, 78 JUDICATURE 182, 182 (1995); Daniel J. Freed, Federal Sentencing in the Wake of Guidelines: Unacceptable Limits on the Discretion of Sentencers, 101 YALE L.J. 1681, 1719-20, 1725-27 (1992); Gerald Heaney, No End to Disparity, 28 AM. CRIM. L. REV. 161 (1991); Marc L. Miller, Domination & Dissatisfaction: Prosecutors as Sentencers, 56 STAN. L. REV. 1211, 1236 (2004); Daniel Zlotnick, The Future of Federal Sentencing Policy: Learning Lessons from Republican Judicial Appointees, 79 U. COLO. L. REV. 1, 27-28 (2008).
46. Kate Stith, The Arc of the Pendulum: Judges, Prosecutors, and the Exercise of Discretion, 117 YALE L.J. 1420, 1430 (2008).
47. See Anderson et al., supra note 23, at 303; Paul J. Hofer et al., The Effect of the Federal Sentencing Guidelines on Inter-Judge Sentencing Disparity, 90 J. CRIM. L. & CRIMINOLOGY 239, 241, 291, 296 (1999).
48. Anderson et al., supra note 23, at 291; Hofer et al., supra note 47, at 282.
49. Anderson et al., supra note 23, at 290 tbl.2.
50. Anderson et al., supra note 23, at 290 tbl.2. at 303.
51. Hofer et al., supra note 47, at 284.
52. Hofer et al., supra note 47, at 287. The percentages reported are derived from R-squared, a regression statistic that measures the fraction of variation in a dependent variable that is explained by the independent variable(s).
53. Hofer et al., supra note 47. at 293-94.
54. Hofer et al., supra note 47 at 298.
55. See, e.g., A. Abigail Payne, Does Inter Judge Disparity Really Matter? An Analysis of the Effects of Sentencing Reforms in Three Federal District Courts, 17 INT'L REV. L. & ECON. 337 (1997) (using data from 1980 to 1991 for select types of cases in three federal district courts); Joel Waldfogel, Aggregate Inter-Judge Disparity in Federal Sentencing: Evidence from Three Districts, 4 FED. SENT'G REP. 151 (1991) (using data from three different district courts from 1984 to 1987); Joel Waldfogel, Does Inter-Judge Disparity Justify Empirically Based Sentencing Guidelines?, 18 INT'L REV. L. & ECON. 293 (1998) [hereinafter Waldfogel, Empirically Based Sentencing] (employing a natural experiment regression analysis using data from ten judges in San Francisco from 1984 to 1987).
56. See Anderson et al., supra note 23, at 302; Hofer et al., supra note 47, at 299-302.
57. See Anderson et al., supra note 23, at 299.
58. See infra notes 154-155 and accompanying text.
59. Pub. L. No. 108-21, § 401, 117 Stat. 650, 667-76 (codified as amended in scattered sections of 18, 28, and 42 U.S.C.).
60. See H.R. REP. NO. 108-66, at 58 (2003) (Conf. Rep.) (announcing an intention to address "the longstanding problem of downward departures from the Federal Sentencing Guidelines"); Stith, supra note 46, at 1465.
61. U.S. SENTENCING COMM'N, DOWNWARD DEPARTURES FROM THE SENTENCING GUIDELINES, at iv-v, 59-60 (2003); Stith, supra note 46, at 1465 (describing the numbers before Congress in 2003 as "powerful," showing "persistent increases in the rate of noncooperation downward departures during the 1990s-especially after the Koon decision was handed down in 1996"); see also Miller, supra note 45, at 1228 fig.1.
62. See infra notes 71-72 and accompanying text.
63. PROTECT Act § 401(d)(2).
64. PROTECT Act § 401(d)(2) § 401(m)(2)(A).
65. PROTECT Act § 401(d)(2) § 401(j)(2).
66. PROTECT Act § 401(d)(2) § 401(l)(1)(A).
67. PROTECT Act § 401(d)(2) § 401(i).
68. See Noelle Tsigounis Valentine, Note, An Exploration of the Feeney Amendment: The Legislation that Prompted the Supreme Court to Undo Twenty Years of Sentencing Reform, 55 SYRACUSE L. REV. 619, 628-29 (2005).
69. Alan Vinegrad, The New Federal Sentencing Law, 15 FED. SENT'G REP. 310, 313 (2003).
70. See Letter from Leonidas Ralph Mecham, Sec'y, Judicial Conference of the U.S., to Senator Orrin G. Hatch, Chairman, Comm. on the Judiciary 3 (Apr. 3, 2003), available at http://www.nacdl.org/public.nsf/2cdd02b415ea3a64852566d6000daa79/departures/$FILE/judconf_feeney.pdf.
71. Max Schanzenbach, Have Federal Judges Changed Their Sentencing Practices? The Shaky Empirical Foundations of the Feeney Amendment, 2 J. EMPIRICAL L. STUD. 1, 1 (2005); Stith, supra note 46, at 1464-65.
72. U.S. SENTENCING COMM'N, supra note 61, at iv.
73. 543 U.S. 220 (2005).
74. PROTECT Act § 401(d)(2) at 226-27, 243-44.
75. The case prompted six separate opinions, including two principal majorities and two principal dissents. Id. at 225.
76. 530 U.S. 466 (2000).
77. 542 U.S. 296 (2004).
78. Booker, 543 U.S. at 244 (Stevens, J., delivering the opinion of the Court in part).
79. Booker, 543 U.S. at 245 (Breyer, J., delivering the opinion of the Court in part).
80. Booker, 543 U.S. at 245 (Breyer, J., delivering the opinion of the Court in part).
81. Booker, 543 U.S at 245, 259-60.
82. See Ryan Scott Reynolds, Note, Equal Justice Under Law: Post-Booker, Should Federal Judges Be Able to Depart from the Federal Sentencing Guidelines to Remedy Disparity Between Codefendants' Sentences?, 109 COLUM. L. REV. 538, 560, 563-64 (2009) (noting that both cases expanded district courts' discretion and that some courts of appeals have responded by reconsidering their treatment of particular sentencing factors); The Supreme Court, 2007 Term-Leading Cases, 122 HARV. L. REV. 276, 333 (2008) (asserting that Kimbrough and Gall "appear to loosen the hold of the Guidelines").
83. 552 U.S. 38 (2007).
84. 552 U.S. 38 (2007) at 47.
85. 552 U.S. 38 (2007) at 51-52.
86. 552 U.S. 85 (2007).
87. 552 U.S. 85 (2007) at 109-10.
88. 552 U.S. 85 (2007) at 101-02 (quoting Brief for Respondent at 16, Kimbrough, 552 U.S. 85 (No. 06-6330)) (internal quotation marks and alteration omitted).
89. 552 U.S. 85 (2007) at 109.
90. See, e.g., United States v. Herrera-Zuniga, 571 F.3d 568, 584-85 (6th Cir. 2009) (interpreting Kimbrough as recognizing "the broad authority of sentencing judges" to "categorically reject the sentencing range prescribed by the Guidelines" (quoting Spears v. United States, 129 S. Ct. 840, 844 (2009))).
91. For a collection of initial reactions, see Erik Luna & Barton Poulson, Restorative Justice in Federal Sentencing: An Unexpected Benefit of Booker?, 37 MCGEORGE L. REV. 787, 787-88 (2006). Two federal judges in the District of Massachusetts publicly praised Booker shortly after it was announced. See Shelley Murphy, Two Boston Jurists Hail Return of Discretion, BOSTON GLOBE, Jan. 13, 2005, at A20.
92. U.S. SENTENCING COMM'N, FINAL REPORT ON THE IMPACT OF UNITED STATES V. BOOKER ON FEDERAL SENTENCING, at vii (2006) [hereinafter FINAL REPORT], available at http://www.ussc.gov/booker_report/Booker_Report.pdf (documenting a modest increase in average sentence length and concluding that "[t]he severity of sentences imposed has not changed substantially").
93. Max M. Schanzenbach & Emerson H. Tiller, Reviewing the Sentencing Guidelines: Judicial Politics, Empirical Evidence, and Reform, 75 U. CHI. L. REV. 715, 739 (2008); see also Douglas A. Berman, Tweaking Booker: Advisory Guidelines in the Federal System, 43 HOUS. L. REV. 341, 349 (2006) ("[D]ata on post-Booker sentencing outcomes released by the Commission reveal only relatively small changes in the patterns of sentencing outcomes.").
94. See Schanzenbach & Tiller, supra note 93, at 739 (noting that "most observers" believe "the fundamentals of sentencing changed little post-Booker"); see also Berman, supra note 93, at 348 ("[T]he Booker decision does not appear to have radically transformed either basic practices or typical outcomes in the federal sentencing system."); Frank O. Bowman, III, The Year of Jubilee . . . or Maybe Not: Some Preliminary Observations About the Operation of the Federal Sentencing System After Booker, 43 HOUS. L. REV. 279, 319 (2006) (calling the changes "strikingly modest"); D. Michael Fisher, Striking a Balance: The Need To Temper Judicial Discretion Against a Background of Legislative Interest in Federal Sentencing, 46 DUQ. L. REV. 65, 77-78 (2007) ("While the change is noticeable, it does not reflect the fear of some post-Booker commentators that judges, now invested with a new kind of discretion, would ignore the Guidelines and sentence defendants however they saw fit."); Jeffrey S. Hurd, Federal Sentencing and the Uncertain Future of Reasonableness Review, 84 DEN. U. L. REV. 835, 860 (2007); Michael M. O'Hear, The Duty To Avoid Disparity: Implementing 18 U.S.C. § 3553(a)(6) After Booker, 37 MCGEORGE L. REV. 627, 645 (2006); Zlotnick, supra note 45, at 15.
95. The data for this Figure comes from table 13 in the 2000-2009 editions of the U.S. Sentencing Commission's Sourcebook of Federal Sentencing Statistics and table 19 in the 2006-2009 editions of the Commission's 4th Quarter Preliminary Data Reports. The fiscal year 2004 period ends on June 24, 2004, the date of the Supreme Court's decision in Blakely v. Washington, 542 U.S. 296 (2004). FINAL REPORT, supra note 92, at 71 tbl.3.
96. The data for this Figure come from figure G and table N in the 2000-2009 editions of the U.S. Sentencing Commission's Sourcebook of Federal Sentencing Statistics and table 1 in the 2006-2009 editions of the Commission's 4th Quarter Preliminary Data Reports. See also FINAL REPORT, supra note 92, app. E-1. Percentages for all post-Booker periods combine traditional departures with nonguideline sentences based on the § 3553(a) factors (sometimes called "variances").
97. U.S. SENTENCING COMM'N, PRELIMINARY QUARTERLY DATA REPORT 1 tbl.1 (2010), available at http://www.ussc.gov/sc_cases/USSC_2010_Quarter_Report_2nd.pdf.
98. See supra note 61 and accompanying text.
99. See Biographical Directory of Federal Judges, FED. JUD. CENTER, http://www.fjc.gov/history/home.nsf/page/judges.html (last visited Oct. 11, 2008). There are 1016 sitting federal district court judges, including 651 judges in active status. Of them, 593 judges (58%), including 506 in active status (78%), were appointed between the effective date of the first Sentencing Guidelines on November 1, 1987 and the Booker decision on January 12, 2005.
100. See, e.g., Nancy Gertner, Supporting Advisory Guidelines, 3 HARV. L. & POL'Y REV. 261, 270 (2009) (describing continued guideline sentencing as the result of "the habits ingrained during twenty years of mandatory Guideline sentencing," and noting that "after the SRA, judges were trained only in the Guidelines"); Stith, supra note 46, at 1496-97 (concluding that "the gravitational pull of the Guidelines on the pendulum of sentencing practice remains strong" based, in part, on the "reluctan[ce]" of "incumbent sentencing decision makers" who were obliged to follow the Guidelines for two decades).
101. United States v. Booker, 543 U.S. 220, 263-64 (2004) (Breyer, J., delivering the opinion of the Court in part).
102. 551 U.S. 338 (2007).
103. 551 U.S. 338 (2007) at 347.
104. Nancy Gertner, What Yogi Berra Teaches About Post-Booker Sentencing, 115 YALE L.J. POCKET PART 137, 140 (2006), http://www.thepocketpart.org/images/pdfs/50.pdf(describing decisions of appellate courts that reinforce the Guidelines and reporting that "[d]istrict judges have gotten the message"); Jack King, Up, Down or Lazy? Panelists Discuss Federal Sentencing After Rita, CHAMPION, Sept.-Oct. 2007, at 8-9; see also Kevin R. Reitz, The Enforceability of Sentencing Guidelines, 58 STAN. L. REV. 155, 171 (2005) (concluding that the post-Booker Guidelines "remain as restrictive of judicial sentencing discretion as any system in the United States"). The incentive against departure from the Guidelines was at least as strong before Booker. See Stephanos Bibas, The Feeney Amendment and the Continuing Rise of Prosecutorial Power to Plea Bargain, 94 J. CRIM. L. & CRIMINOLOGY 295, 302 (2004).
105. Gertner, supra note 100, at 270; see also Cass R. Sunstein, Behavioral Analysis of Law, 64 U. CHI. L. REV. 1175, 1188 (1997); Amos Tversky & Daniel Kahneman, Judgment Under Uncertainty: Heuristics and Biases, 185 SCIENCE 1124, 1128-30 (1974).
106. See Birte Englich et al., Playing Dice with Criminal Sentences: The Influence of Irrelevant Anchors on Experts' Judicial Decision Making, 32 PERSONALITY & SOC. PSYCHOL. BULL. 188, 194 (2006).
107. Gall v. United States, 552 U.S. 38, 49 (2007).
108. Gertner, supra note 104, at 138; Stith, supra note 46, at 1496.
109. See Zlotnick, supra note 45, at 14-15; Daniel A. Chatham, Note, Playing with Post-Booker Fire: The Dangers of Increased Judicial Discretion in Federal White-Collar Sentencing, 32 J. CORP. L. 619, 637-38 (2007) (recommending this approach).
110. Cf. Jack B. Weinstein, The Role of Judges in a Government of, by, and for the People: Notes for the Fifty-Eighth Cardozo Lecture, 30 CARDOZO L. REV. 1, 211 (2008). On this theory, the change in party control of Congress in 2006 and the White House in 2008 could embolden district court judges to depart more frequently.
111. Cf. Jack B. Weinstein, The Role of Judges in a Government of, by, and for the People: Notes for the Fifty-Eighth Cardozo Lecture, 30 CARDOZO L. REV. 1, 211 (2008). On this theory, the change in party control of Congress in 2006 and the White House in 2008 could embolden district court judges to depart more frequently.
112. King, supra note 104, at 9 (quoting Judge Myron Thompson).
113. Fitzgerald Statement, supra note 10, at 3.
114. Immergut Statement, supra note 11, at 12 (warning that "the signs point to increasing sentencing disparity-including disparity based on differing judicial philosophies among judges working in the same courthouse").
115. Benton J. Campbell, U.S. Att'y, E. Dist. N.Y., Statement Before the U.S. Sentencing Commission in the Regional Hearing on the State of Federal Sentencing 8 (July 9, 2009) (transcript available at http://www.ussc.gov/AGENDAS/20090709/Campbell_testimony.pdf).
116. Immergut Statement, supra note 11, at 6 (reporting that some judges "continue to follow the advisory guideline sentence in the majority of cases" while "other judges routinely decline to impose a guideline sentence").
117. Dana Boente, U.S. Att'y, E. Dist. Va., Statement Before the U.S. Sentencing Commission in the Regional Hearing on the State of Federal Sentencing 3 (July 9, 2009) (transcript available at http://www.ussc.gov/AGENDAS/20090709/Boente_testimony.pdf).
118. Wroblewski Memorandum, supra note 7, at 2.
119. See supra notes 32-35 and accompanying text.
120. For an overview of proposed reforms, see Berman, supra note 93, at 356-71.
121. For an overview of proposed reforms, see Berman, supra note 93 at 364-65 (discussing proposals to "Blakely-ize" the Guidelines); see also United States v. Booker, 543 U.S. 220, 276-79, 284-85 (2005) (Stevens, J., dissenting in part).
122. E.g., KAN. STAT. ANN. § 21-4716(b) (West 2010).
123. See MODEL PENAL CODE: SENTENCING § 7.07B(2) (Tentative Draft No. 1, 2007).
124. Holder, supra note 8.
125. See 28 U.S.C. § 995(a)(12)-(16) (2006).
126. U.S. SENTENCING COMM'N, GUIDE TO PUBLICATIONS & RESOURCES 2007-2008, at 45 (2007), available at http://www.ussc.gov/publicat/Cat2005.pdf ("Pursuant to the policy on public access to Sentencing Commission documents and data, all case and defendant identifiers have been removed from the data." (internal citation omitted)).
127. The Feeney Amendment authorized Congress or the Justice Department to request data that include the identity of the sentencing judge, but did not provide for public dissemination of that information. See 28 U.S.C. § 994(w)(3)-(4) (2006).
128. See Anderson et al., supra note 23, at 287.
129. See Public Access to Sentencing Commission Documents and Data, 54 Fed. Reg. 51,279, 51,282 (Dec. 13, 1989).
130. Mark H. Bergstrom & Joseph S. Mistick, The Pennsylvania Experience: Public Release of Judge-Specific Sentencing Data, 16 FED. SENT'G REP. 57, 63 (2003) (noting that Pennsylvania now releases judge-identifying information and that "[m]any of the negative outcomes predicted during the development of the policy have not materialized"); Marc L. Miller, A Map of Sentencing and a Compass for Judges: Sentencing Information Systems, Transparency, and the Next Generation of Reform, 105 COLUM. L. REV. 1351, 1356 n.19, 1385 (2005).
131. See, e.g., Richard G. Kopf, A Brief and Modest Proposal, SENT'G L. & POL'Y(July 28, 2010, 5:19 PM), http://sentencing.typepad.com/sentencing_law_and_policy/2010/07/a-brief-and-modest-proposal-an-original-essay-from-us-district-judeg-richard-kopf-.html.
132. See Miller, supra note 130, at 1356 & n.19; Marc L. Miller, Sentencing Reform "Reform" Through Sentencing Information Systems, in THE FUTURE OF IMPRISONMENT 121, 146-48 (Michael Tonry ed., 2004); Schanzenbach & Tiller, supra note 93, at 741-42; Steven L. Chanenson, Write On!, 115 YALE L.J. POCKET PART 146, 147 (2006), http://www.thepocketpart.org/images/pdfs/1.pdf.
133. Schanzenbach & Tiller, supra note 93, at 729-30.
134. Specifically, the Commission no longer reports the date of sentencing, but instead reports only the month and year, greatly increasing the chance that multiple cases in the Commission's data will match publicly available docket information for a given case. See infra note 242.
135. United States v. Green, 346 F. Supp. 2d 259, 277 n.66 (D. Mass. 2004) (Young, C.J.) (citing Minutes of the Court Meeting of the District of Massachusetts 4 (Sept. 4, 2001)). Judges may order the Statement of Reasons sealed for case-specific reasons, id., such as the protection of a defendant who cooperated with authorities. In practice, judges rarely order the Statement of Reasons sealed, minimizing the risk of selection bias. I encountered fewer than five cases (out of more than 2200 total coded) in which the Statement of Reasons was unavailable.
136. The documents are also a gold mine of qualitative data. Many judges attach transcripts from the sentencing hearing or write narrative descriptions of their reasons, offering a rare glimpse of how judges are sentencing-on a day-to-day basis in ordinary, unreported cases-after Booker.
137. See JUDICIAL CONFERENCE OF THE U.S., REPORT OF THE PROCEEDINGS OF THEJUDICIAL CONFERENCE OF THE UNITED STATES 17 (2001).
138. United States v. Kandirakis, 441 F. Supp. 2d 282, 332 n.76 (D. Mass. 2006) (Young, J.).
139. Details of the case-matching technique are set forth in Part A.2 of the Appendix.
140. See supra notes 26-30 and accompanying text.
141. See Letter from Leonidas Ralph Mecham to Senator Orrin G. Hatch, supra note 70, at 3.
142. Anderson et al., supra note 23, at 290-91; Hofer et al., supra note 47, at 282.
143. Specifically, sentences were excluded if the sentencing judge was on pace to impose fewer than twenty-five sentences in a two-year period. Cf. Anderson et al., supra note 23, at 288 (using a cutoff of thirty cases, including jurisdictional transfers and acquittals, in a two-year period).
144. D. MASS. LOCAL R. 40.1(B)(3) (2008).
145. See infra notes 245-248 and accompanying text.
146. For a detailed breakdown of the sentence count for each judge, see Table A2.
147. See infra note 241 & Table A1 and accompanying text.
148. See infra Table A3 and accompanying text.
149. See Anderson et al., supra note 23, at 281; Hofer et al., supra note 47, technical app. at 307-09; Waldfogel, Empirically Based Sentencing, supra note 55, at 294. Consistent with the Sentencing Commission's convention, sentences of probation are coded as zeromonths of imprisonment. See Hofer et al., supra note 47, technical app. at 307-09; see also STITH & CABRANES, supra note 18, at 62 (following the same convention).
150. Sentence length is measured as a term of imprisonment in months. Following the Sentencing Commission, a sentence of probation is coded as zero months of imprisonment.
151. See infra notes 259-261 and accompanying text.
152. Because the Commission has not yet released sentencing data for fiscal year 2009 and beyond, the Kimbrough/Gall period, of necessity, is shorter than the other periods. For a full discussion of period selection issues, see infra notes 232-237 and accompanying text.
153. Hofer et al., supra note 47, at 275 n.103.
154. Hofer et al., supra note 47 at 240-41; see also STITH & CABRANES, supra note 18, at 119.
155. Hofer et al., supra note 47, at 240-41; see also STITH & CABRANES, supra note 18, at 119 (acknowledging "[a] possibility that comparing each judge's average sentence masks considerable variability within each set of sentences"); Hofer et al., supra note 47, at 296 (calling judge-to-judge disparity in average sentence length "the tip of the disparity iceberg").
156. See Bowman, supra note 94, at 296.
157. See supra notes 113-117 and accompanying text.
158. Average distance from the guideline range is calculated using all of the judge's sentences, treating within-range sentences as zero months. See infra note 261 and accompanying text.
159. For an explanation of the cutoff dates for each period, see infra notes 232-237 and accompanying text.
160. See infra notes 249-256 and accompanying text.
161. See, e.g., Nancy Gertner, Circumventing Juries, Undermining Justice: Lessons from Criminal Trials and Sentencing, 32 SUFFOLK U. L. REV. 419 (1999); Nancy Gertner, From Omnipotence to Impotence: American Judges and Sentencing, 4 OHIO ST. CRIM. L.J. 523 (2007) [hereinafter Gertner, Omnipotence to Impotence]; Nancy Gertner, Rita Needs Gall-How To Make the Sentencing Guidelines Advisory, 85 DEN. U. L. REV. 63 (2007); Gertner, supra note 100; Gertner, supra note 104; Nancy Gertner, Women Offenders and the Sentencing Guidelines, 14 YALE J.L. & FEMINISM 291 (2002); Nancy Gertner, Federal Sentencing Guidelines: A View from the Bench, HUM. RTS., Spring 2002, at 6.
162. See, e.g., William G. Young, An Open Letter to U.S. District Judges, 50 FED. LAW., July 2003, at 30. Judge Young's remarkable 177-page decision in United States v. Green, 346 F. Supp. 2d 259 (D. Mass. 2004), not only anticipated the invalidation of the Guidelines on Sixth Amendment grounds, but contains one of the most comprehensive critiques of the Guidelines ever assembled.
163. See Patti B. Saris, Below the Radar Screens: Have the Sentencing Guidelines Eliminated Disparity? One Judge's Perspective, 30 SUFFOLK U. L. REV. 1027 (1997).
164. Hofer et al., supra note 47, at 279.
165. 2008 Presidential Race: Massachusetts, N.Y. TIMES, Nov. 4, 2008, http://elections.nytimes.com/2008/president/states/massachusetts (noting that over the last ten Presidential elections, Massachusetts has been the most solidly Democratic state in the country); Lydia Saad, Political Ideology: "Conservative" Label Prevails in the South, GALLUP (Aug. 14, 2009), http://www.gallup.com/poll/122333/political-ideology-conservative-label-prevails-south.aspx (using Gallup poll results to show that Massachusetts is the most liberal state in the nation, trailing only the District of Columbia).
166. See Biographical Directory of Federal Judges, supra note 99. Party affiliation is an imperfect proxy for ideology, and Republican judges in a politically liberal state likeMassachusetts probably skew more liberal than their Republican colleagues nationwide. Nonetheless, a study of Boston judges does not involve any greater risk of party and ideology effects than past studies of San Francisco, see Waldfogel, Empirically Based Sentencing, supra note 55, at 294, or New York City and Philadelphia, see Payne, supra note 55, at 337. Even the large-scale national studies have ensured a random distribution of cases by limiting their data set to district offices where multiple judges shared a single case wheel, which necessarily oversamples sentences in cities with disproportionately liberal and Democratic populations.
167. The data supporting this comparison can be found in appendix B of the 2002-2008 editions of the U.S. Sentencing Commission's Sourcebook of Federal Sentencing Statistics. The gap between the national and Massachusetts figures for guideline sentencing is partially attributable to "fast-track" programs for immigration offenses, see U.S. SENTENCING GUIDELINES MANUAL § 5K3.1 (2009) (authorizing departures for early disposition programs), which account for 7.9% of sentences nationwide, according to the fiscal year 2008 data. Fast-track programs ease a crushing burden on courts and prosecutors in border districts, but they are controversial because they must be authorized by the Attorney General and are not available in all districts, injecting obvious regional disparity into sentencing outcomes. Fast-track programs have the effect of boosting the nationwide rate of governmentsponsored sentences compared with districts, like Massachusetts, that have no fast-track authority.
168. See Ilene H. Nagel & Stephen J. Schulhofer, A Tale of Three Cities: An Empirical Study of Charging and Bargaining Practices Under the Federal Sentencing Guidelines, 66 S. CAL. L. REV. 501, 552-58 (1992) (documenting inter-district disparities driven by prosecutor and defense practices); Daniel Richman, Federal Sentencing in 2007: The Supreme Court Holds-The Center Doesn't, 117 YALE L.J. 1374, 1403-06 (2008) (discussing inter-district disparities driven by differences among federal prosecutors' offices).
169. See Divisions, U.S. ATT'Y'S OFF. DISTRICT MASS. (last visited Sept. 3, 2010), http://www.usdoj.gov/usao/ma/divisons.html. It is possible that prosecutors and defense attorneys in the district change their charging and plea bargaining practices in response to the judge assigned to the case, based on the judge's reputation. Because such changes reflect an assessment of the judge, rather than differences between prosecutors or between defense attorneys, they are properly treated as sources of inter-judge disparity.
170. See Gall v. United States, 552 U.S. 38, 47 (2007) (rejecting the conclusion of some courts of appeals that a significant variance from the Guidelines requires an extraordinary justification); Rita v. United States, 551 U.S. 338, 347 (2007) (affirming the decision of some courts of appeals to apply a presumption of reasonableness when reviewing withinrange sentences on appeal).
171. See Schanzenbach & Tiller, supra note 93, at 735. For foundational research on the influence of party affiliation on courts of appeals, see CASS R. SUNSTEIN ET AL., ARE JUDGES POLITICAL? AN EMPIRICAL ANALYSIS OF THE FEDERAL JUDICIARY (2006).
172. Cases were treated as having no mandatory minimum if the court sentenced below the otherwise-applicable minimum based on the statutory "safety valve," 18 U.S.C. § 3553(f) (2006), or a government "substantial assistance" motion, see 18 U.S.C. § 3553(e) (2006); U.S. SENTENCING GUIDELINES MANUAL § 5K1.1 (2009).
173. For details of these regression models for sentence length, see Table A4.
174. For discussion of the regression models generally, see infra notes 259-261 and accompanying text.
175. For a discussion of period-selection issues, see Appendix A.1.
176. For details of these regression models, see Table A5.
177. The fact that the model for the Kimbrough/Gall period is not significant reinforces the need for caution in interpreting the results for cases not governed by a mandatory minimum. Statistical significance is highly sensitive to sample size, and the Kimbrough/Gall period necessarily has about one-third as many cases as the other periods, even before excluding mandatory minimums. Although the relationship in the Kimbrough/Gall period is strongly positive, the model falls well short of statistical significance.
178. This Article uses letters, rather than names, to identify judges. See supra text accompanying notes 140-141.
179. See Gertner, supra note 104, at 137-38 (using the phrase "free at last" to describe the reaction to Booker among some district court judges).
180. Sentences by Judges E, F, and G also fit this pattern. Judge E's rate of belowrange sentencing approximately tripled since Booker, from 7.3% in the Mandatory Guidelines period and 13.3% in the PROTECT Act period, to 34.0% and 32.7% in the two post-Booker periods, and falling to 21.2% in the Kimbrough/Gall period. Judge F's rate of belowrange sentencing has more than doubled, from 15.0% in the Mandatory Guidelines period and 14.7% in the PROTECT Act period, to 34.1% and 31.6% in the two post-Booker periods. So has Judge G's rate of below-range sentencing, which went from 13.9% in the Mandatory Guidelines period, to 10.5% in the PROTECT Act period, to 33.3%, 34.0%, and 32.4% in the three periods since Booker.
181. Sentences by Judge H fit a similar pattern. Judge H's below-range sentencing rates in the pre-Booker periods (16.5% and 23.9%) are very similar to those in the post-Booker periods (24.4%, 25.5%, and 17.4%). Sentences by Judge I seemed to fit this pattern during the eighteen months after Booker with a rate of 22.0%, compared with 25.9% in the Mandatory Guidelines period and 21.1% under the PROTECT Act. But Judge I's rate of below-range sentencing more than doubled to 47.7% in the Post-Booker II period, and stands at 38.9% in the Kimbrough/Gall period.
182. The sentencing pattern of Judge J is unique and highly volatile. From a belowrange sentencing rate of 24.6% in the Mandatory Guidelines period, it dropped to 11.9% under the PROTECT Act, nearly tripled to 32.1% in the Post-Booker I period, dropped again to 19.1% in the Post-Booker II period, and has nearly doubled again to 31.8% in the Kimbrough/ Gall period.
183. See infra Appendix A.4 (defining "discretionary" sentences and explaining why they are of special relevance in measuring inter-judge disparity in guideline sentencing).
184. See Hofer et al., supra note 47, at 287. Actual months of variance explained were determined by "(1) multiplying the total variance by the portion of the variance accounted for by judges, and (2) finding the square root of the result, thus translating the numbers back into absolute terms." Id. at 287 n.127.
185. For details of these regression models, see Tables A6 and A7.
186. See supra Part I.B.3.
187. See supra Tables 1 & 2 and accompanying text.
188. See Hofer et al., supra note 47, at 287; see also Waldfogel, Empirically Based Sentencing, supra note 55, at 295.
189. Anderson et al., supra note 23, at 294 n.53.
190. That was what one Senator predicted in the immediate aftermath of Booker. Press Release, Senator Chuck Grassley, Supreme Court Decision on Sentencing Guidelines (Jan. 13, 2005), available at 2005 WLNR 2769009.
191. In addition to calling for research on inter-judge sentencing disparity, Attorney General Holder has convened a department-wide Sentencing and Corrections Working Group to consider those issues. See Holder, supra note 8.
192. See supra notes 99-112 and accompanying text.
193. See supra notes 99-100 and accompanying text.
194. Pre-1987 service by the sentencing judge was not a statistically significant predictor of sentence length in any of the three periods.
195. Pre-1987 service by the sentencing judge was not a statistically significant predictor of how far a sentence fell from the guideline range during any period.
196. See supra notes 101-104 and accompanying text.
197. Gall v. United States, 552 U.S. 38, 51 (2007).
198. Kimbrough v. United States, 552 U.S. 85, 101-02 (2007).
199. See supra Tables 1 & 3 and accompanying text.
200. See Gertner, supra note 104, at 140 (concluding, in the period between Booker and the decisions in Kimbrough and Gall, that appellate courts were closely policing sentences on appeal and that "[d]istrict judges have gotten the message").
201. See supra notes 105-108 and accompanying text.
202. Englich et al., supra note 106, at 194.
203. Gall v. United States, 552 U.S. 38, 49 (2007).
204. See supra notes 109-110 and accompanying text.
205. See, e.g., LEE EPSTEIN & JACK KNIGHT, THE CHOICES JUSTICES MAKE 9-18 (1998); THOMAS H. HAMMOND ET AL., STRATEGIC BEHAVIOR AND POLICY CHOICE ON THE U.S. SUPREME COURT 65-248 (2005); JEFFREY A. SEGAL & HAROLD J. SPAETH, THE SUPREME COURT AND THE ATTITUDINAL MODEL REVISITED 97-109 (2002).
206. See supra notes 111-112 and accompanying text.
207. Richard A. Posner, What Do Judges and Justices Maximize? (The Same Thing Everybody Else Does), 3 SUP. CT. ECON. REV. 1, 31 (1993) (describing a "judicial utility function" that includes work time and leisure time).
208. See supra note 181 and accompanying text.
209. See Gertner, Omnipotence to Impotence, supra note 161, at 530, 539 (describing "robust judicial opposition to the Guidelines"); Gertner, supra note 100, at 267 (describing how district court judges "overwhelmingly opposed the Guidelines"); Joseph W. Luby, Reining in the "Junior Varsity Congress": A Call for Meaningful Judicial Review of the Federal Sentencing Guidelines, 77 WASH. U. L.Q. 1199, 1276 (1999) ("[T]he Commission has little legitimacy in the sentencing regime. Its Guidelines are reviled (even though tolerated) by lawyers, judges, and commentators alike."); José A. Cabranes, Sentencing Guidelines: A Dismal Failure, N.Y. L.J., Feb. 11, 1992, at 2 (claiming that "virtually everyone who is associated with the federal justice system" deems the Guidelines a "dismal failure").
210. Naftali Bendavid, Judicial Traitor or Consensus Builder? Breyer's Role as Sentencing Pioneer Still Rankles, LEGAL TIMES, May 16, 1994, at 7.
211. Criticizing Sentencing Rules, U.S. Judge Resigns, N.Y. TIMES, Sept. 30, 1990, § 1, at 22; Don J. DeBenedictis, The Verdict Is In, A.B.A. J., Oct. 1993, at 78.
212. Klein & Steiker, supra note 32, at 232-33; see also Stephanos Bibas, Blakely's Federal Aftermath, 16 FED. SENT'G REP. 333, 342 (2004) ("To put it bluntly, many judges and others hate the Guidelines . . . "); Julie R. O'Sullivan, In Defense of the U.S. Sentencing Guidelines' Modified Real-Offense System, 91 NW. U. L. REV. 1342, 1343-44 (1997) ("[J]udging from the scholarly commentary, virtually everyone loves to hate [the Guidelines].").
213. Blakely v. Washington and the Future of the Sentencing Guidelines: Hearing Before the S. Comm. on the Judiciary, 108th Cong. 3-4 (2004) (statement of Sen. Orrin Hatch, Chairman, S. Comm. on the Judiciary) (stating "I can understand why").
214. U.S. SENTENCING COMM'N, RESULTS OF SURVEY OF UNITED STATES DISTRICT JUDGES JANUARY 2010 THROUGH MARCH 2010 tbl.8 (2010).
215. See id.
216. See For all drug types except crack cocaine, more than 50% of judges responded that the guideline range was generally appropriate. See id.
217. See id. at tbls.8 & 13. For example, although 41% of judges consider vocational skills "ordinarily relevant," 53% consider that factor "not ordinarily relevant," and 6% consider it "never relevant." Id.
218. See LINDA DRAZGA MAXFIELD, U.S. SENTENCING COMM'N, FINAL REPORT SURVEY OF ARTICLE III JUDGES ON THE FEDERAL SENTENCING GUIDELINES, at ES-1 (2003), available at http://www.ussc.gov/Judge_Survey/execsum.pdf.
219. See LINDA DRAZGA MAXFIELD, U.S. SENTENCING COMM'N, FINAL REPORT SURVEY OF ARTICLE III JUDGES ON THE FEDERAL SENTENCING GUIDELINES. at 12 exh. II-11, 15 exh. II-14, 24 exh. II-23, available at http://www.ussc.gov/Judge_Survey/jschap2.pdf.
220. See LINDA DRAZGA MAXFIELD, U.S. SENTENCING COMM'N, FINAL REPORT SURVEY OF ARTICLE III JUDGES ON THE FEDERAL SENTENCING GUIDELINES at 24 exh. II-23.
221. See LINDA DRAZGA MAXFIELD, U.S. SENTENCING COMM'N, FINAL REPORT SURVEY OF ARTICLE III JUDGES ON THE FEDERAL SENTENCING GUIDELINES at 2 exh. II-1, 3 exh. II-3.
222. See LINDA DRAZGA MAXFIELD, U.S. SENTENCING COMM'N, FINAL REPORT SURVEY OF ARTICLE III JUDGES ON THE FEDERAL SENTENCING GUIDELINES at 12 exh. II-11.
223. See LINDA DRAZGA MAXFIELD, U.S. SENTENCING COMM'N, FINAL REPORT SURVEY OF ARTICLE III JUDGES ON THE FEDERAL SENTENCING GUIDELINES at 15 exh. II-14.
224. See Rita v. United States, 551 U.S. 338, 349-50 (2007); United States v. Wilson, 350 F. Supp. 2d 910, 914-15 (D. Utah 2005) (Cassell, J.). For a thoughtful defense of the Commission's work in the controversial context of the child pornography guidelines, see United States v. Cunningham, 680 F. Supp. 2d 844, 862-64 (N.D. Ohio 2010).
225. United States v. Wanning, 354 F. Supp. 2d 1056, 1062 n.9 (D. Neb. 2005); see also Stephanos Bibas et al., Policing Politics at Sentencing, 103 NW. U. L. REV. 1371, 1388 (2009) (stating that "courts lack the institutional competence to make systemic policy choices," but "Congress has established an agency, the Sentencing Commission, to collect data and the views of various constituencies in formulating policies and rules").
226. Kevin R. Reitz, Sentencing Facts: Travesties of Real-Offense Sentencing, 45 STAN. L. REV. 523, 553-55 (1993) (acknowledging a division of opinion among scholars, but reviewing research showing that "despite our faith that we can spot those offenders most likely to recidivate, individualized predictions of future dangerousness are little better than a game of chance"); see also Erica Beecher-Monas & Edgar Garcia-Rill, Danger at the Edge of Chaos: Predicting Violent Behavior in a Post-Daubert World, 24 CARDOZO L. REV. 1845, 1845-47 (2003); Stephen J. Morse, Blame and Danger: An Essay on Preventive Detention, 76 B.U. L. REV. 113, 126 & n.39 (1996); Paul H. Robinson, Commentary, Punishing Dangerousness: Cloaking Preventive Detention as Criminal Justice, 114 HARV. L. REV. 1429, 1450 (2001).
227. United States v. Tabor, 365 F. Supp. 2d 1052, 1061 (D. Neb. 2005).
228. 28 U.S.C. § 994(p) (2006).
229. United States v. Cage, 451 F.3d 585, 593 (10th Cir. 2006) (describing the Guidelines as "an expression of popular political will about sentencing"); Wanning, 354 F. Supp. 2d at 1062 n.9; Wilson, 350 F. Supp. 2d at 915; Bibas et al., supra note 225, at 1388 ("Most importantly, Congress has democratic legitimacy; courts do not.").
230. Luby, supra note 209, at 1202 ("The Commission . . . rarely justifies its guidelines, consistently avoids on-the-record decisionmaking, and operates unencumbered by the procedural safeguards that ensure the political legitimacy of other administrative agencies."); Kate Stith & José A. Cabranes, Judging Under the Federal Sentencing Guidelines, 91 NW. U. L. REV. 1247, 1270-71 (1997) (noting that "the Sentencing Commission almost never explains the reason behind a particular Guidelines rule," and characterizing the Guidelines as a "compilation of administrative diktats").
231. See Frank O. Bowman, III, Mr. Madison Meets a Time Machine: The Political Science of Federal Sentencing Reform, 58 STAN. L. REV. 235, 254-55 (2005).
232. 542 U.S. 296 (2004).
233. The Commission's post-Booker reports have largely ignored the period between Blakely and Booker as well. See, e.g., FINAL REPORT, supra note 92.
234. At its Data and Research Conference in May 2009, the Commission distributed flash drives containing the full set of sentencing data files through fiscal year 2008. The release of fiscal years 2007 and 2008 data ahead of the ordinary schedule was unexpected, and a valuable benefit for participants.
235. The Kimbrough/Gall period is shorter than the other periods because no data are available for fiscal year 2009.
236. See supra notes 59-67 and accompanying text.
237. See infra Tables A4 & A5 and accompanying text.
238. Schanzenbach & Tiller, supra note 93, at 729-30.
239. Schanzenbach & Tiller, supra note 93 at 729. Each of those data points ordinarily appears in the criminal docket, with the exception of ethnicity. Schanzenbach and Tiller presumably determined ethnicity by asking whether the defendant had a Hispanic-sounding name.
240. See id. at 730.
241. PACER's "Reports" tool allows searches by "Case Type," including criminal cases. I included pending and terminated defendants, but excluded cases involving fugitive defendants. I also conducted targeted searches for cases with earlier filing dates that were "closed" during fiscal year 2002, to ensure a comparable percentage of matched cases in each year being studied.
242. The Commission made date matching much more difficult because, beginning in 2004, case records no longer include the exact date of sentencing, but only the month and year. U.S. SENTENCING COMM'N, VARIABLE CODEBOOK FOR INDIVIDUAL OFFENDERS 63 (2009) (noting that after fiscal year 2003, "[t]he date on which the defendant was sentenced" is not available).
243. Cf. Schanzenbach & Tiller, supra note 93, at 730 (reporting that only 3% of sentences could not be matched using the docket sheet alone, mostly in immigration cases). Like Schanzenbach and Tiller, however, I encountered a surprising number of sentences, about 8.5% of those in the initial search, that did not look similar to any of the Commission's records. I echo their concern that this is a significant amount of missing data. See id.
244. Fiscal years 2004 and 2005 include fewer sentences because they exclude sentences imposed between Blakely and Booker. Boston cases were identified using the Commission's parole office code, except that cases without any parole office code were included.
245. Because chi-square analysis depends on a minimum number of cases per cell, the race variable (the Commission's NEWRACE) was limited to white, black, and Hispanic offenders, omitting the "other" category. Similarly, the education variable (NEWEDUC) omitted the "college graduate" category, which applied to too few defendants. The Commission's age variable was coded into three categories: age 18-29, age 30-39, and age 40 and over.
246. Chi-square tests on age uncovered no significant relationship in any period. Tests on education uncovered no significant relationship in any period except Kimbrough/Gall, and that result likely was affected by the smaller population of cases. Tests on gender uncovered no significant relationship in any period except Post-Booker II. Given the results for gender in adjacent periods and for the dataset as a whole, that result does not call into question the premise that the distribution of cases was random.
247. Hofer et al., supra note 47, technical app. at 320.
248. Cf. Waldfogel, Empirically Based Sentencing, supra note 55, at 295 (relying exclusively on a test of randomness using gender).
249. See U.S. SENTENCING GUIDELINES MANUAL § 5G1.1(c)(2) (2009). If the statutory minimum exceeds both the guideline minimum and the guideline maximum, then the statutory minimum becomes the guideline sentence. Id. § 5G1.1(b).
250. See id. § 5G1.1 cmt.
251. Missing data prevented the coding of constraints for 1.2% of cases in the dataset. Percentages reported for each constraint are based on the remaining cases.
252. See 18 U.S.C. § 3585(b) (2006).
253. Although not a legal constraint, the federal judges with whom I have spoken cannot imagine circumstances in which a judge would impose a sentence of less than time served, which would imply that the prior detention was unlawful. See Telephone Interview with Paul Cassell, Professor and Former District Court Judge for the District of Utah (Oct. 10, 2008); cf. U.S. SENTENCING GUIDELINES MANUAL § 1B1.10 cmt. n.3 (2009) (prohibiting the reduction of a sentence "below time served" following a downward amendment to the Guidelines).
254. To the extent that the "time served" constraint overlapped with other constraints, the case was coded as "time served."
255. U.S. SENTENCING GUIDELINES MANUAL § 5B1.1(a)(1) (2009).
256. U.S. SENTENCING GUIDELINES MANUAL § 5B1.1(a)(1) (2009) § 5B1.1(a)(2).
257. See Hofer et al., supra note 47, at 275 & nn.101-03 (recognizing the distorting effect of mandatory minimums).
258. See Waldfogel, Empirically Based Sentencing, supra note 55, at 295.
259. In linear regression, the R-squared statistic is a value between 0 and 1 that describes the percentage of variance in the dependent variable that is explained by the independent variable. See generally MICHAEL O. FINKELSTEIN & BRUCE LEVIN, STATISTICS FOR LAWYERS 345 (1990). For a discussion of some uses and limitations of R-squared, see David R. Stras & Ryan W. Scott, An Empirical Analysis of Life Tenure: A Response to Professors Calabresi & Lindgren, 30 HARV. J.L. & PUB. POL'Y 791, 817 (2007). The R-squared values here are shown as percentages.
260. Hofer et al., supra note 47, at 287. Actual months of variance explained were determined by "(1) multiplying the total variance by the portion of the variance accounted for by judges, and (2) finding the square root of the result, thus translating the numbers back into absolute terms." Id. n.127.
261. Distance from the guideline range should always be either zero or a positive number, and a handful of cases were omitted due to logic problems, likely because the total sentence reflected consecutive sentences but the judge or the Commission recorded the guideline minimum and maximum for only one offense. The Commission codes a sentence of life imprisonment as 470 months. See U.S. SENTENCING COMM'N, VARIABLE CODEBOOK FOR INDIVIDUAL OFFENDERS 64 (2009). For consistency, in calculating distance from the guideline range, I treated a guideline minimum or maximum of life imprisonment as 470 months as well.
262. Hofer et al., supra note 47, at 312.
263. See JEFFREY M. WOOLDRIDGE, ECONOMETRIC ANALYSIS OF CROSS SECTION AND PANEL DATA 518-20 (2002).
264. See supra Table 3 and accompanying text.
265. See supra notes 235-236 and accompanying text.
266. See supra Table 1 and accompanying text.
267. See supra Table 2 and accompanying text.
268. See Anderson et al., supra note 23, at 290-91 (two years before the Guidelines, six years after); Hofer et al., supra note 47, at 284 (two years); Waldfogel, Empirically Based Sentencing, supra note 55, at 295 (four years).