See Steven D. Clymer, Unequal Justice: The Federalization of Criminal Law, 70 S. Cal. L. Rev. 643, 668-69 (1997) (“Notwithstanding some significant exceptions, defendants typically fare considerably worse when prosecuted in federal court. . . . [This] disparity . . . is a hallmark of federalization.”); Daniel C. Richman, The Changing Boundaries Between Federal and Local Law Enforcement, in 2 Office of Justice Programs, U.S. Dep’t of Justice, Criminal Justice 2000, Boundary Changes in Criminal Justice Organizations 81, 95 (2000), http://www.ncjrs.gov/criminal_justice2000/vol_2/02d2.pdf; infra note 149 and cross-references cited therein.

See, e.g.,Gerald G. Ashdown, Federalism, Federalization, and the Politics of Crime, 98 W. Va. L. Rev. 789, 789-90, 812-13 (1996); John S. Baker, Jr., State Police Powers and the Federalization of Local Crime, 72 Temp. L. Rev. 673, 673-74, 678 (1999); Sara Sun Beale, Too Many and Yet Too Few: New Principles to Define the Proper Limits for Federal Criminal Jurisdiction, 46 Hastings L.J. 979, 979-81 (1995); Clymer, supra note 1, at 645-46. The historical claim is, however, debatable. The federal government’s involvement in violent crime began around the time of the New Deal, precisely as the federal government began to intervene in many other areas previously considered the proper province of state and local officials but which, today, are readily accepted as proper subjects of federal control. See Daniel Richman, The Past, Present, and Future of Violent Crime Federalism, 34 Crime & Just. 377, 387 (2006) (noting that the notorious kidnapping of Charles Lindbergh’s son in 1932 began “a wave of congressional enactments targeting criminal behavior that had hitherto been the exclusive province of state and local enforcers”). Nevertheless, federal involvement in street crime did increase significantly beginning in the 1960s, see id. at 382, 390-400, and it is this relatively more recent policy shift that has generated criticism.

See, e.g., Beale, supra note 2, at 998-99; Clymer, supra note 1, at 669-75; Daniel C. Richman, “Project Exile” and the Allocation of Federal Law Enforcement Authority, 43 Ariz. L. Rev. 369, 398 (2001); Michael A. Simons, Prosecutorial Discretion and Prosecution Guidelines: A Case Study in Controlling Federalization, 75 N.Y.U. L. Rev. 893, 916-17 (2000); Tom Stacy & Kim Dayton, The Underfederalization of Crime, 6 Cornell J.L. & Pub. Pol’y 247, 286, 295 (1997); William J. Stuntz, Unequal Justice, 121 Harv. L. Rev. 1969, 2027-29 (2008).

See Paul H. Robinson, Distributive Principles of Criminal Law: Who Should Be Punished How Much? (2008); Paul H. Robinson & John M. Darley, Justice, Liability and Blame: Community Views and the Criminal Law (1995); Tom R. Tyler, Why People Obey the Law (Princeton Univ. Press 2006) (1990).

See Nathan Gorenstein, Man Who Avoided Conviction Despite 44 Arrests Now Faces 32-Year Term, Phila. Inquirer, Feb. 2, 2012, http://articles.philly.com/2012-02-02/news/31016967 _1_conviction-rate-robbery-cases-court-system; John Sullivan, Emilie Lounsberry & Dylan Purcell, Gun Arrests Galore, No Convictions at All, Phila. Inquirer, Dec. 16, 2009, http://articles.philly.com/2009-12-16/news/24988541_1_robbery-victim-mandatory-minimum -five-year-sentence-gun-crime (detailing the defendant’s criminal record).

See Brief of Appellee United States of America at 26, 37-39, United States v. Andrews, 270 F. App’x 5 (2d Cir. 2008) (No. 06-4240) (on file with author).

See United States v. Thomas, 447 F. App’x 568, 569 (5th Cir. 2011) (per curiam); Brief for Appellee, the United States of America at 7-11, Thomas,447 F. App’x 568(No. 11-30148), 2011 WL 2527772, at *7-11 (detailing the defendant’s criminal record).

Clymer, supra note 1; Stuntz, supra note 3.

See sources cited supra note 3.

See Rachel E. Barkow, Federalism and the Politics of Sentencing, 105 Colum. L. Rev. 1276, 1312-14 (2005); Beale, supra note 2, at 998-99; Clymer, supra note 1, at 674-75; Michael M. O’Hear, National Uniformity/Local Uniformity: Reconsidering the Use of Departures to Reduce Federal-State Sentencing Disparities, 87 Iowa L. Rev. 721, 730-32 (2002); Simons, supra note 3, at 916-17; Stacy & Dayton, supra note 3, at 286-87; Ronald Wright, Federal or State? Sorting as a Sentencing Choice, Crim. Just., Summer 2006, at 16, 17.

See Baker, supra note 2, at 685-86, 703-06; Clymer, supra note 1, at 669-73; John C. Jeffries, Jr. & John Gleeson, The Federalization of Organized Crime: Advantages of Federal Prosecution, 46 Hastings L.J. 1095, 1103-17 (1995).

See Richman, supra note 3, at 397 (observing that a lower ratio of cases per prosecutor is among federal prosecutors’ advantages relative to their local counterparts); Stacy & Dayton, supra note 3, at 294 (noting that as compared to the federal government, states “are especially inclined to underinvest” in street crime enforcement because it particularly afflicts poor urban neighborhoods); Ronald F. Wright, Trial Distortion and the End of Innocence in Federal Criminal Justice, 154 U. Pa. L. Rev. 79, 117-21, 134-37 (2005) (arguing that as case volume per federal prosecutor has fallen over the last five decades, acquittal rates have generally gone down, but acknowledging that the reasons for this correlation are unclear).

See Susan R. Klein & Ingrid B. Grobey, Debunking Claims of Over-Federalization of Criminal Law, 62 Emory L.J. 1, 10 (2012) (attributing the high federal guilty plea rate to the fact that federal prosecutors have the luxury of charging “only rock-solid cases[,] as they can decline most cases safely in the knowledge that state and local actors must pursue them”); Richman, supra note 1, at 95 (“Without the political obligations of State authorities to maintain order within a territorial jurisdiction and to prosecute every provable serious offense, Federal agencies are largely free to invest strategically in the cases they do take.”).

See infra notes 49-51 and accompanying text.

See infra notes 68-79 and accompanying text.

For a more thorough statement of legitimacy principles, see Tom R. Tyler & Jeffrey Fagan, Legitimacy and Cooperation: Why Do People Help the Police Fight Crime in Their Communities?, 6 Ohio St. J. Crim. L. 231, 234 (2008).

Much of the literature on legitimacy has focused on legitimacy’s impact on compliance, that is, willingness to obey the laws and legal authorities. In that vein, inverse correlations have been shown between citizens’ trust in the federal government and rates of street crime and homicide. See Gary LaFree, Losing Legitimacy: Street Crime and the Decline of Social Institutions in America 91-113 (1998); Randolph Roth, American Homicide 448-52 (2009). Because federal and state laws with respect to street crime are by now nearly co-extensive, see Richman, supra note 1, at 82, 90-91, there is no way to evaluate whether the federal criminal justice system in particular promotes heightened compliance with the criminal law (and thus lower crime rates). My aim instead is to explore legitimacy’s potential influence on deference to and cooperation with federal authorities in the prosecution of cases—primarily deference and cooperation of witnesses, jurors, and judges—and to theorize how such cooperation and deference might help, in part, to explain outcome disparities between the state and federal forums in criminal prosecutions.

See Josh Bowers & Paul H. Robinson, Perceptions of Fairness and Justice: The Shared Aims and Occasional Conflicts of Legitimacy and Moral Credibility, 47 Wake Forest L. Rev. 211, 217 (2012) (“To the extent that people see the system as in conflict with their judgments of justice, [] acquiescence and cooperation is likely to fade and be replaced with resistance and subversion. . . . Witnesses may lose an incentive to offer their information or testimony. Citizens may fail to report crimes in the first instance. Jurors may disregard their jury instructions. Police officers, prosecutors, and judges may make up their own rules. And offenders may resist adjudication processes and punishments rather than participate in them.”).

This is not to say that federal penalties for street crime are necessarily substantively unjust (although one could certainly make a strong argument that they are unjust in certain cases and even whole categories of cases). My point here, rather, is that penalties are most apt to be substantively just when supported by the communities most affected by the crime and its punishment.

See infra note 150.

Richman, supra note 1, at 88-91.

Pub. L. No. 90-618, 82 Stat. 1213.

Organized Crime Control Act of 1970, Pub. L. No. 91-452, Title IX, 84 Stat. 941.

Pub. L. No. 91-513, 84 Stat. 1236.

The Comprehensive Crime Control Act of 1984, Pub. L. No. 98-473, 98 Stat. 1976, among other things strengthened the penalties for certain drug offenses, established mandatory minimum sentences for certain gun crimes, and overhauled the federal sentencing system, abolishing parole and creating a sentencing commission directed to establish mandatory guidelines for all federal criminal defendants. In 1986 and 1988, Congress passed two Anti-Drug Abuse Acts, Pub. L. No. 99-570, 100 Stat. 3207, and Pub. L. No. 100-690, 102 Stat. 4181, each of which added additional mandatory minimum sentences for certain drug-trafficking offenses, particularly those related to crack cocaine. The Comprehensive Crime Control Act of 1990, Pub. L. No. 101-647, 104 Stat. 4789, further increased drug penalties.

E.g., 18 U.S.C. § 3559(c) (2012) (federal “three strikes” statute, described infra note 42); 18 U.S.C. § 924(e) (mandating a minimum of 15 years’ imprisonment for defendants convicted of any of nine offenses, including illegal gun possession, who have three prior convictions for certain drug-trafficking or violent crimes). Certain federal narcotics-trafficking penalties give prosecutors discretion to double penalties for defendants previously convicted of a felony drug trafficking crime. See 21 U.S.C. §§ 841(b)(1), 851.

By the term “street crime,” here and throughout this Article, I refer to gun offenses, narcotics offenses, and episodic violent crimes such as robbery and homicide. Street crimes may be perpetrated by individuals acting alone or as part of organized criminal activity (such as gang and drug-trafficking operations). I exclude from this term international (cross-border) narcotics trafficking.

See generally Richman, supra note 1, at 88-90 (citing cases). United States v. Lopez, 514 U.S. 549 (1995), has proven one of the few exceptions. Richman, supra note 1, at 90 (describing Lopez as “little more than a speed bump” in the Court’s acquiescence in the “complete federalization of criminal law”). Although United States v. Morrison, 529 U.S. 598 (2000), and Jones v. United States, 529 U.S. 848 (2000), are often cited as Lopez’s jurisprudential analogues,they are not. The Court in Morrison struck down a civil remedy under the Violence Against Women Act; several federal courts of appeals have upheld the constitutionality of that Act’s criminal provisions without disturbance. E.g., United States v. Page, 167 F.3d 325, 325 (6th Cir.) (en banc), cert. denied, 528 U.S. 1003 (1999); see United States v. Larsen, 615 F.3d 780, 786 (7th Cir. 2010) (joining “the Second, Fourth, Fifth, and Sixth Circuits in holding that the [criminal-penalty provision] is a valid exercise of Congress’s power under the Commerce Clause to regulate the channels or instrumentalities of, or persons in, interstate commerce”). And in Jones, the Court excluded private residences from the federal arson statute as a matter of statutory interpretation, not constitutional limitation.

See sources cited supra notes 2-3, 10-13.

See Rory K. Little, Myths and Principles of Federalization, 46 Hastings L. J. 1029, 1065-66 (1995).

See sources citedsupra note 1.

See, e.g., sources cited supra notes 2-3, 10-13.

See Jeffries & Gleeson, supra note 11; Stacy & Dayton, supra note 3, at 286-87.

E.g., Beale, supra note 2; Clymer, supra note1; O’Hear, supra note 10.

See supra notes 1-3, 10-13.

See Clymer, supra note 1; Wayne A. Logan, Erieand Federal Criminal Courts, 63 Vand. L. Rev. 1243, 1270 (2010); Wright, supra note 10, at 84-86, 129-34; see also Stacy & Dayton, supra note 3, at 286-87 (noting that relatively harsher federal sentences induce defendants to cooperate). Cooperation almost always entails pleading guilty to charged crimes (and sometimes also admitting to uncharged crimes).

Barkow, supra note 10, at 1312-14 n.169; Clymer, supra note 1; Stacy & Dayton, supra note 3, at 286-87.

See, e.g., O’Hear, supra note 10, at 731-32; Simons, supra note 3, at 916-19.

Even when commentators have acknowledged instances in which state penalties are harsher than federal counterparts, they have assumed that, in those instances, federal prosecutors defer to state prosecution. See, e.g., Clymer, supra note 1, at 669 n.128. In fact, though, federal prosecutors do prosecute even in instances where the applicable state statute would (if applied) result in a harsher sentence. See infra notes 50-51 and accompanying text.

For instance, prior to 2004, a first-time offender found in possession of four ounces (approximately 113 grams) of any controlled substance other than marijuana faced a mandatory minimum term of fifteen years. N.Y. Penal Law §§ 220.21, 60.05, 70.0(3) (McKinney 1995), amended by 2004 N.Y. Sess. Laws 1474 (McKinney). Under federal law, the highest mandatory minimum penalty for a first-time felony drug offense was (and still is) ten years. See 21 U.S.C. § 841(b)(1)(A) (2012). Then-mandatory federal sentencing guidelines for such an offense ranged from 27 to 71 months for cocaine and 63 to 150 months for heroin, depending on the offender’s criminal history. See U.S. Sentencing Guidelines Manual §§ 2D1.1(c)(11), 2D1.1(c)(7) & Sentencing Table (2003).

From 1973 to 1998, Michigan’s notorious “650-lifer law” (the subject of an unsuccessful Eighth Amendment challenge in Harmelin v. Michigan, 501 U.S. 957 (1991)) imposed mandatory lifetime imprisonment without parole for possession of at least 650 grams of cocaine; from 1998 to 2002, the statute imposed a mandatory term of twenty years, and it was not until 2003 that Michigan abolished mandatory terms for drug offenses. Mich. Comp. Laws § 333.7401 (1997), amended by 1998 Mich. Legis. Serv. 319 (West), amended by 2002 Mich. Legis. Serv. 665 (West). Florida’s drug trafficking statute, still in effect, mandates a sentence of seven years for possession of at least 200 grams of cocaine, fifteen years for at least 14 grams of heroin or 400 grams of cocaine, twenty-five years for at least 28 grams of heroin, and mandatory life for more than 30 kilograms of heroin or 150 kilograms of cocaine.See Fla. Stat. Ann. § 893.135(1)(b)-(c) (West 2012). First-time offenders and even some second-time offenders subject to these statutes would face lower mandatory penalties under federal law.See 21 U.S.C. § 841(b)(1) (mandating a sentence of ten years for trafficking in at least 1 kilogram of heroin or 5 kilograms of cocaine, five years for at least 100 grams of heroin or 500 grams of cocaine, and no mandatory minimum for lesser quantities);21 U.S.C. § 851 (giving prosecutors discretion to double applicable mandatory minimum terms for defendants previously convicted of a felony drug-trafficking crime).

The federal “three strikes” statute, 18 U.S.C. § 3559(c), imposes a mandatory sentence of life imprisonment upon conviction for a “serious violent felony” (defined as any offense punishable by at least ten years which requires the use, or involves significant risk, of force) if the defendant has been convicted on separate prior occasions of either two “serious violent felonies” or one “serious violent felony” and one “serious drug offense” (defined as conduct punishable under 21 U.S.C. § 841(b)(1)(A)). A provision of the Armed Career Criminal Act (ACCA), 18 U.S.C. § 924(e), mandates a minimum of fifteen years’ imprisonment for defendants convicted of illegal gun possession who have three prior convictions for certain drug-trafficking or violent crimes.

See Cal. Penal Code §§ 667(e)(2)(A)(ii), 1170.12(c)(2)(A)(ii) (West 1994), amended by Proposition 36, Cal. Secretary St. (2012), http://vig.cdn.sos.ca.gov/2012/general/pdf/text -proposed-laws-v2.pdf#nameddest=prop36. California’s statute required a minimum of at least 25 years’ imprisonment prior to parole-eligibility. California amended its three strikes statute in 2012 by making it applicable only if (i) the “third strike” conviction is for a “serious” or “violent” felony or for certain sex, drug, and firearm offenses, or (ii) the defendant has a prior conviction for rape, murder, or child molestation. See Cal. Penal Code §§ 667(e)(2)(A)(ii), 1170.12(c)(2)(A)(ii) (West 2012).

La. Rev. Stat. Ann. § 15:529.1(A)(3)(b) (West 2013). The “three strikes” provision renders convicted offenders ineligible for parole or any other form of early release. Louisiana also has a “four strikes” provision that mandates twenty years’ imprisonment upon a fourth felony conviction for any crime. See La. Rev. Stat. Ann. § 15:529.1(A)(4).

The law ultimately withstood the challenge, see Ewing v. California, 538 U.S. 11 (2003), as did a substantially similar three-strikes statute in Texas that was also challenged on Eighth Amendment grounds, see Rummell v. Estelle, 445 U.S. 263 (1980). Federal defendants routinely (and almost always unsuccessfully) challenge non-death penalty sentences as “cruel and unusual” in violation of the Eighth Amendment, see, e.g.,United States v. Vaughn, 527 F. App’x 826 (11th Cir. 2013) (upholding mandatory life sentence in federal drug case as non-violative of Eighth Amendment and noting that “outside the context of capital punishment, there have been few successful challenges to the proportionality of sentences” (internal citations omitted)). The Supreme Court has yet to hear such a challenge.

In Florida, since 1999, the possession or discharge of a firearm in connection with a drug-trafficking offense or certain violent felony offenses results in a consecutive mandatory minimum sentence of ten or twenty years, respectively—twice the terms applicable for single violations under the comparable federal law. Compare Fla. Stat. Ann. § 775.087(2)(a) (West 2012), with 18 U.S.C. § 924(c). Maryland, New York, and Virginia have all enacted mandatory minimum terms for possession of a firearm, whereas there is no mandatory minimum under federal law for the equivalent crime. See Md. Code Ann., Pub. Safety, § 5-133(c)(1)-(2) (West 2012) (five-year minimum for defendants with drug-trafficking felony and/or violent felony convictions); N.Y. Penal Law §§ 70.02(1)(b), 70.02(3)(d), 265.03(3) (2013) (3½-year mandatory minimum penalty); Va. Code Ann. §§ 18.2-308.2, 18.2-308.4(C) (2010) (five-year or two-year mandatory minimum penalty for violent and non-violent felons, respectively).

See 18 U.S.C. § 3624(b) (enabling federal prisoners to receive up to fifty-four days credit per year of imprisonment for “good time,” defined as exemplary compliance with prison disciplinary rules).

By 1999, fourteen states (including Florida, Illinois, Ohio, Arizona, and North Carolina) had abolished any early-release by a parole board, and thirty-four states (including New York, Pennsylvania, California, Michigan, Missouri, Louisiana, Virginia, and the District of Columbia) had required some or all prisoners to serve at least eighty-five percent of their imposed sentence notwithstanding any parole release or other earned adjustments. Bureau of Justice Statistics, NCJ 170032, Truth in Sentencing in State Prisons 2-3 (1999).

Indeed, that is what some scholars presume. See, e.g.,Clymer, supra note 1.

I use these laws as illustrations. An exhaustive state-by-state comparison of violent crime, firearms, and narcotics laws, and their penalties relative to comparable federal laws, is well beyond the scope of this Article, and, as noted, I do not quarrel with the basic premise that most federal penalties are more severe than their state counterparts. My point, rather, is that federal cases are brought even in instances where applicable state penalties are equally or more severe.

According to data from the Federal Judicial Center:

· Between 2001 and 2011, federal prosecutors in Florida charged at least 1,043 defendants (an average of 95 per year) with a violation of 18 U.S.C. § 924(c)—even though Florida law mandates double the imprisonment time for the same effective conduct. See supra note 46.

·Between 2001 and 2011, federal prosecutors in Virginia have prosecuted at least 1,572 defendants (an average of 143 per year) with a violation of 18 U.S.C. § 922(g), and 1,550 defendants (an average of 141 per year) with a violation of 18 U.S.C. § 924(c)—even though Virginia’s penalties for the same conduct are at least as, and in many instances more, severe. See supra note 46.

·In the five years preceding New York’s 2006 gun law amendment, which mandates a term of three and a half years for illegal firearm possession, there were at least 1,024 federal cases prosecuted in New York State for a violation of 18 U.S.C. § 922(g), while in the five years following the amendment, there were 846 defendants prosecuted for that crime—a decrease of only 17%, notwithstanding the absence of any mandatory penalty for the equivalent federal crime. See supra note 46.

According to data from the United States Sentencing Commission:

· Between 1991 and 2008 (the latest year for which offense-specific data is available), federal prosecutors have charged at least twenty-one California defendants and at least 386 Louisiana defendants under the ACCA (which mandates fifteen years’ imprisonment) even though these defendants would be eligible for mandatory life imprisonment under their respective states’ “three-strikes” statutes.

·Federal prosecutors have prosecuted at least 2,359 Florida defendants for cocaine offenses and 2,397 for heroin offenses that would have carried a mandatory minimum sentence of life imprisonment if prosecuted under state law.

·In 1997 (the earliest year for which drug weight data is available, and the year before Michigan amended its mandatory life term for offenses involving at least 650 grams of cocaine), federal prosecutors charged at least 126 defendants with cocaine offenses that would have carried a minimum sentence of life imprisonment if prosecuted under state law. Moreover, the changes in Michigan’s penalty scheme for cocaine offenses involving at least 650 grams, see supra note 41, appear to have had little effect on federal prosecutions of cocaine offenses: in the five years preceding Michigan’s 2002 repeal of a mandatory minimum for cocaine offenses involving at least 650 grams, federal prosecutors charged 373 defendants with such offenses, while in the five years following the repeal, they charged 322 defendants with such offenses—a decrease of just 14%.

(Datasets for FJC and USSC data are available at http://www.icpsr.umich.edu/icpsrweb /ICPSR/series/00072, and http://www.icpsr.umich.edu/icpsrweb/ICPSR/series/83, subject to restricted use agreement; data analysis is on file with author.)

A few notes on context and methodology are in order. First, context. Charging patterns vary widely across districts. The ACCA is a good example of such variation. Some districts in states less populous than California and Louisiana charge more ACCA cases, while others in states nearly as populous, or more populous (in the case of Louisiana), charge fewer. For instance, New York, nearly four times the population of Louisiana and more than half the population of California, charged just 109 ACCA cases between 1991 and 2008. See FJC and USSC data and analysis on file with author. For state population data, see http://quickfacts.census.gov/qfd/states/06000.html (California); http://quickfacts.census .gov/qfd/states/36000.html (New York); and http://quickfacts.census.gov/qfd/states /22000.html (Louisiana). (It should be noted that the Sentencing Commission data files indicate that there are thousands of federal defendants in California whose status under the ACCA went unreported, which may result in an understatement of the total number subject to the ACCA.)

Second, methodology. There are two variables that can change applicable mandatory minimums for federal gun and drug charges: (i) multiple counts of conviction; and (ii), for drug offenses, prior felony convictions for certain drug-trafficking offenses can, in some instances, double the applicable mandatory minimum sentence or even raise it to a mandatory life sentence. See supra note 41. To mitigate these confounding variables, I did several things. With respect to the offenses under 18 U.S.C. § 922(g) (which criminalizes firearm possession by prohibited persons such as convicted felons), I analyzed only defendants for whom this was the sole count of conviction. With respect to the offenses under 18 U.S.C. § 924(c) (which criminalizes the possession or use of a firearm in furtherance of a drug-trafficking crime or crime of violence, and so is typically charged with one or more of those other crimes), I analyzed only defendants for whom 18 U.S.C. § 924(c) was the count carrying the most severe penalty as calculated by the Federal Judicial Center, and for whom it was the sole count of conviction for that crime (because multiple convictions under § 924(c) significantly increase the applicable mandatory minimum under federal law). With respect to drug offenses, the sentencing commission’s data files do not provide specifics on defendants’ prior convictions, making it difficult to ascertain the mandatory minimum sentence the defendant faced prior to conviction. I therefore rely only upon charges that would have carried at least as severe a penalty under state law notwithstanding a defendant’s prior drug-trafficking convictions—that is, life imprisonment. See 21 U.S.C. §§ 841(b)(1)(A), 851 (providing a mandatory minimum sentence of life for a defendant with two prior qualifying drug-trafficking convictions, subject to prosecutorial discretion). This methodology also avoids the issue of multiple counts impacting applicable penalties, since the minimum penalty faced under state law in any event is life. My drug offense methodology led me to focus on these specific Michigan and Florida statutes, even though there are other state penalty schemes that carry (or carried) higher penalties for first-time and even some second-time drug-trafficking offenders. New York’s Rockefeller laws, which like the federal narcotics laws contain recidivist enhancements, are, of course, a notorious example. See supra note 40.

In short, the limitations of these datasets make it impossible to ascertain fully and accurately the number of federal criminal defendants who would have faced the same or harsher penalties under state law. And the methodology I employ with respect to the drug offenses carries its own pitfalls. Namely, it focuses on large-quantity drug offenses that may constitute interstate or even international drug-trafficking in addition to (or in lieu of) street-level offenses. Notwithstanding these limitations, though, this data at least illuminates an under-noticed feature of federal charging decisions in gun and drug offenses: they are not necessarily motivated by applicable state penalties.

See generally Scott H. Decker & Jack McDevitt,Nat’l Inst. Just., Project Safe Neighborhoods: Strategic Interventions, Gun Prosecution Case Screening 5-6 (2006) (discussing local prosecutors’ role in federal case screening); Lisa L. Miller & James Eisenstein, The Federal/State Criminal Prosecution Nexus: A Case Study in Cooperation and Discretion,30 Law & Soc. Inquiry 239, 244-47, 253-54 (2005) (discussing local law enforcement collaboration in federal charging decisions).

See generally Scott H. Decker et al., U.S. Dep’t Just., Project Safe Neighborhoods: Strategic Interventions: Gun Prosecution Case Screening: Case Study 1 (2006) (discussing intentional efforts to increase federal prosecution of gun crimes across all federal districts through partnerships with federal and local prosecutors and local and state law enforcement agencies).

Miller & Eisenstein, supra note 52, at 257-58 (“When asked why [the local ADAs] bother taking those cases [in which applicable state and federal penalties are similar] to federal court if the defendants are not likely to receive a longer sentence anyway, the ADA replied, ‘Because they are much more likely to get convicted in the federal system.’”). A prime example of this dynamic occurs in Baltimore, Maryland, where state law mandates a five-year penalty for gun possession by defendants previously convicted of certain crimes. See Md. Code Ann., Pub. Safety, § 5-133(c) (LexisNexis 2011). Notwithstanding the absence of anymandatory minimum term for the equivalent federal crime, federal prosecutors routinely threaten to prosecute defendants who do not plead guilty in state court to the mandatory five-year penalty—a tactic that has proven quite successful at garnering the desired state court outcomes. See Press Release, Baltimore EXILE Partners Announce 60% Increase in Violent Defendants Charged Federally Since 2005, Feb. 20, 2008, http://www .justice.gov/usao/md/news/archive/BaltimoreEXILEPartnersAnnounce60IncreaseinViolentDefendantsChargedFederallySince2005.html.

Had Andrews been prosecuted in state court for first degree armed robbery, he would, as a prior violent felony offender, have faced a sentence of 10 to 25 years. N.Y. Penal Law §§ 70.04, 70.02 (McKinney 2004 & Supp. 2007).

See 42 Pa. Cons. Stat. Ann. § 9714(a) (West 2007).

“Charge bargaining” is the practice in which prosecutors offer a defendant the chance to plead guilty to a charge with a lower penalty than might apply under initially filed or as-yet unfiled charges. Because charge bargaining goes unreported, its frequency is difficult to ascertain; one qualitative study of federal sentencing guidelines circumvention practices in ten districts found that circumvention occurred in approximately 20% to 35% of cases resolved by guilty plea, and that of the various methods of circumventing guidelines and mandatory penalties, charge bargaining was “the most important.” See Stephen J. Schulhofer & Ilene H. Nagel, Plea Negotiations Under the Federal Sentencing Guidelines: Guideline Circumvention and Its Dynamics in the Post-MistrettaPeriod, 91 Nw. U. L. Rev. 1284, 1293-94 (1997). Attempts to quantify charge-bargaining from available data kept by the Sentencing Commission can be misleading because those datasets do not contain information on counts initially filed and dismissed, or on counts promised to be filed absent a plea. The recent report by Human Rights Watch attempting to quantify a “trial penalty” for federal defendants charged with narcotics-related offenses suffers from this limitation. See Human Rights Watch, An Offer You Can’t Refuse: How US Federal Prosecutors Force Drug Defendants to Plead Guilty 102 n.326 (Dec. 2013). Defendants who go to trial receive on average higher sentences than those who plead guilty, id. at 102-04, but it is impossible to tell, based on sentencing data alone, whether this is the result of plea bargaining, reductions for cooperation, the effects of trials on judges’ sentencing decisions, or the fact that defendants facing higher penalties at the outset may more often chose to take their chances at trial—or a mix of some or all of these causes, and others. Research on these questions is sorely needed.

From 2002 to 2012, the guilty plea rate for homicide offenses has ranged from a low of 67.0% (in 2004) to a high of 79.0% (in 2012), with a mean of 73.8%. The overall guilty plea rate in the federal system has ranged from a low of 85.8% (in 2003) to a high of 89.0% (in 2012), with a mean of 87.3%. See Judicial Business Archive, U.S. Cts., http://www.uscourts .gov/Statistics/JudicialBusiness/archive.aspx (click on each of the desired years (2002-2012), then click on “Table D-4 Defendants Disposed of, by Type of Disposition and Major Offense”) (last visited Dec. 11, 2013) (providing copies of the annually published report Judicial Business of the U.S. Courts). “Homicide offenses” cover a number of different federal statutes, most of which carry mandatory minimum terms of imprisonment, and all of which carry the highest possible Guidelines ranges and maximum terms of life (or death—although prosecutors are not permitted to enter into a binding plea agreement with a death-eligible defendant before the Attorney General has decided whether to seek the death penalty, see United States Attorneys’ Manual (USAM) § 9-10.110, http://www.justice .gov/usao/eousa/foia_reading_room/usam/index.html (last updated 2011)). See, e.g., 18 U.S.C. § 924(c), (j) (2012) (mandatory minimum of ten years for discharge of firearm, and additional punishment if the discharge results in death); id. § 1111 (mandatory life imprisonment for murder in the first degree); 21 U.S.C. § 848(a)-(b) (mandatory minimum of twenty years’ imprisonment).

Over the last eight years, the guilty plea rate for immigration offenses excluding illegal re-entry has ranged from a low of 92.8% (in 2008) to a high of 95.4% (in 2010), with a mean of 94.0%. Judicial Business Archive, supra note 58 (click on each of the desired years (2005-2012), then click on “Table D-4 Defendants Disposed of, by Type of Disposition and Major Offense”). I exclude illegal reentry because the Sentencing Commission allows a four-level Guidelines reduction for a speedy guilty plea by those defendants, a unique penalty-based inducement. See U.S. Sentencing Guidelines Manual § 5K3.1 (2013); Memorandum from James M. Cole, Dep. Att’y Gen., U.S. Dep’t of Justice, Department Policy on Early Disposition or “Fast-Track” Programs 3 (Jan. 31, 2012), http://www.justice.gov/dag/fast -track-program.pdf.

Over the last decade, the guilty plea rate for narcotics offenses has ranged from a low of 86.5% (in 2005) to a high of 89.5% (in 2010), with a mean of 88.3%. Judicial Business Archive, supra note 58 (click on each of the desired years (2002-2012), then click on “Table D-4 Defendants Disposed of, by Type of Disposition and Major Offense”). For overall federal guilty plea rate data, see supra note 58.

The data show only terminated charges and do not indicate whether or how those differ from the charges initially filed. In the event of multiple counts, the data reflect the outcome of the most severe count upon termination. Non-guilty plea resolutions reflect defendants who did not enter a guilty plea to any count. See E-mail from Kristin Garry, Admin. Office of the U.S. Courts, to author (July 24, 2013) (on file with author).

See Stuntz, supra note 3, at 2021-25 (arguing that the relative ease of proving narcotics offenses as compared to violent crime has escalated the “war on drugs”).

See Stephanos Bibas, Plea Bargaining Outside the Shadow of Trial, 117 Harv. L. Rev. 2463, 2470 (2004) (arguing that of the variety of factors influencing plea-bargaining, “[t]he strength of the prosecution’s case is the most important factor”).

543 U.S. 220 (2005).

The Sentencing Guidelines permit up to a three-level reduction in offense level in cases resolved by guilty plea. See U.S. Sentencing Guidelines Manual § 3E1.1 (2013).

In 2003, before either Booker or its predecessor, Blakely v. Washington,542 U.S. 296 (2004), which invalidated a mandatory state guidelines regime, had been decided, the guilty plea rate in the federal system was 85.8% and the acquittal rate was 0.87%. Leonidas Ralph Mecham, 2003 Judicial Business tbl.D-4 (2003), in Judicial Business Archive, supra note 58, http://www.uscourts.gov/uscourts/Statistics/JudicialBusiness/2003/appendices/d4.pdf. Between 2004 and 2012, the federal guilty plea rate has climbed, from approximately 86% in 2004 to 89% in 2012, and the acquittal rate has steadily deceased, from 0.78% in 2004 to 0.40% in 2012. Judicial Business Archive, supra note 58 (click on each of the desired years (2004-2012), then click on “Table D-4 Defendants Disposed of, by Type of Disposition and Major Offense”). Meanwhile, the frequency of below-Guideline variances (i.e., Booker-based departures and/or below-Guidelines sentences) has steadily increased. In 2006, the year after Booker, sentencing courts granted non-government sponsored below-Guideline variances in 8% of cases; by 2012, the rate had jumped to 14.9%. Compare U.S. Sent’g Comm’n, Annual Sourcebook of Sentencing Statistics tbl.27 (11th ed. 2006) [hereinafter 2006 Sourcebook], http://www.ussc.gov/Research_and_Statistics/Annual _Reports_and_Sourcebooks/2006/sbtoc06.htm (tabulating both Booker-based departures and Booker based variances and reporting government-sponsored downward departures and variances in 27.6% of cases), with U.S. Sent’g Comm’n, Annual Sourcebook of Sentencing Statistics tbl.27 (17th ed. 2012) [hereinafter 2012 Sourcebook], http://www .ussc.gov/Research_and_Statistics/Annual_Reports_and_Sourcebooks/2012/sbtoc12.htm (demonstrating that the frequency of government-sponsored downward departures and variances had increased by just three percentage points from FY 2006 to FY 2012).

In 2003, the guilty plea and acquittal rates, respectively, for violent offenses were 82.3% and 1.5%; for drug offenses, 87.9% and 0.5%; for firearms offenses, 83.3% and 1.6%. Mecham, supra note 66, at app. 214 tbl.D-4. In 2012, the respective guilty plea and acquittal rates for violent offenses were 84.1% and 0.9%; for drug offenses, 89.5% and 0.3%; and for firearms offenses, 88.3% and 0.8%. Thomas F. Hogan, 2012 Annual Report of the Director: Judicial Business of the United States Courts tbl.D-4 (2012), in Judicial Business Archive, supra note 58, http://www.uscourts.gov/Statistics/JudicialBusiness/2012/statistical -tables-us-district-courts-criminal.aspx. Meanwhile, between 2006 and 2012, the percentage of cases in which courts granted Booker-based below-Guidelines variances or departures jumped markedly across the board in violent crimes and narcotics cases: in murder cases, from 2.6% to 14.3%; in robbery cases, from 11.1% to 20%; in narcotics trafficking cases, from 8.1% to 16.8%; and in firearms cases, from 8.9% to 17.2%. Compare 2006 Sourcebook, supra note 66, at tbl.27 (tabulating both Booker-based departures and Booker-based variances), with 2012 Sourcebook, supra note66, at tbl.27 (same).

See Richman, supra note 3, at 370.

Compare Michael Janofsky, Fighting Crime by Making Federal Case About Guns, N.Y. Times, Feb. 10, 1999, http://www.nytimes.com/1999/02/10/us/fighting-crime-by-making-federal -case-about-guns.html (citing those who credited Project Exile for the 36% decline in Richmond’s homicide rate between 1997 and 1998), with Steven Raphael & Jens Ludwig, Prison Sentence Enhancements: The Case of Project Exile, in Evaluating Gun Policy: Effects on Crime and Violence 251 (Jens Ludwig & Philip J. Cook eds., 2003) (arguing that the evidence is inadequate to show that Project Exile caused the decline).

Janofsky, supra note 69.

An exception is Richman, supra note 3, at 407-08 (noting that this failure tended to diminish prospects of a politically acceptable exit strategy for the federal program).

Sherri Johnson et al., Evaluation of the Virginia Exile Program: Final Report, Va. Dep’t Crim. Just. Services 6 (July 2003), http://www.dcjs.virginia.gov/research/documents/exileFinal .pdf [hereinafter Virginia Exile Report]; see U.S. Sentencing Guidelines Manual § 2K2.1 & Sentencing Table (2003).

Virginia Exile Report, supra note 72.

None of the six localities selected to implement Virginia Exile expended the full grant award amounts, indicating resources were not lacking. Id. at 16.

Id. at 57.

Charles D. Bonner, The Federalization of Crime: Too Much of a Good Thing?, 32 U. Rich. L. Rev. 905, 928-29 (1998); Richman, supra note 3, at 379. The federal jurisdictional bar is quite low, requiring, under 18 U.S.C. § 924(c), proof that the gun was used either in connection with a federal violent or drug trafficking crime, or, under 18 U.S.C. § 922(g), proof that the seized gun was manufactured outside the state where the defendant possessed it. Federal jurisdiction for illegal gun possession is readily had in Virginia, a state without a major firearms manufacturer. See Annual Firearms Manufacturing and Export Report,Bureau of Alcohol, Tobacco & Firearms (2011), http://www.atf.gov/files/statistics/download /afmer/2011-final-firearms-manufacturing-export-report.pdf (providing state-by-state listing of firearms manufacturers and firearms produced and showing that, of the over 6.3 million firearms manufactured in the United States in 2011, only 3,770—approximately 0.06%—were made in Virginia).

Of the 527 Exile defendants prosecuted in the program’s first year, 123 (23%) were prosecuted federally. Virginia Exile Report, supra note 72, at 57.

Indeed, this was one explanation given by local Virginia prosecutors. See id. at 2. I address this general theory in Section I.C.

See id. at 57. My calculation assumes 123 defendants would be added to the 404 locally prosecuted, and that each of these 123 would be convicted. The same 53% rate results if the analysis is performed on cases rather than defendants. See id.

See Wright, supra note 12, at 153 n.225, 154; supra note 57. This is certainly true in the context of narcotics defendants with prior felony drug-trafficking convictions, where federal law permits prosecutors to double applicable penalties at their discretion, offering an enormous plea-bargaining “chip.” See 21 U.S.C. §§ 841(b)(1), 851 (2012). Charge bargaining, though, raises more questions about forum disparities than it answers. What distinguishes federal prosecutors’ charge-bargaining leverage from that of local prosecutors when state law supplies similar penalties for a given crime? Resources play a role, to be sure, but do not quite account for the depth of the disparities we see. See supra note 74 and accompanying text;infra notes 106-113 and accompanying text.

E.g., Clymer, supra note 1, at 671-73.

See Michael J. Gorman, Survey: State Search and Seizure Analogs, 77 Miss. L.J. 417, 424, 427, 444, 446, 448, 453 (2007) (identifying in this category Connecticut, Georgia, New Jersey, New Mexico, North Carolina, and Pennsylvania).

See id. at 420, 435, 446-47, 457 (identifying Alaska, Massachusetts, New Mexico, New York, and Tennessee).

See Charles Steigler, Offering Monetary Rewards to Public Whistleblowers: A Proposal for Attacking Corruption at its Source, 9 Ohio St. J. Crim. L. 815, 821 n.21 (2012) (identifying seventeen states with such a ban).

Jeffries & Gleeson, supra note 11, at 1111 & n.63 (noting that the use of hearsay testimony before the grand jury is prohibited in New York, California, Alaska, Nevada, and South Dakota).

This has been the case in California (where, pursuant to a 1982 voter referendum, claims relating to the exclusion of evidence on grounds of unreasonable search and seizure arising under state law are measured by the same standard governing such claims under federal law), Florida, Kansas, Kentucky, Maine, Maryland, Michigan, Missouri, Nebraska, Texas, and West Virginia. See Gorman, supra note 82, at 422 & n.19, 425-26, 431-36, 439, 441, 457-58, 462.

See generally Gorman, supra note 82(providing an overview of the key differences between federal and state constitutional criminal procedure across all fifty states). Suppression motions in criminal cases invariably rest on the credibility of the law enforcement officers whose actions are the subject of constitutional challenge. In this respect, there may be differences in how such credibility questions are resolved between state and federal forums (and as between federal and local law enforcement agents, see infra Section III.A). But that is to be distinguished from forum-based differences in procedural rules and doctrine.

See Daniel C. Richman, Judging Untried Cases, 156 U. Pa. L. Rev. 219, 222 (2007).

This is not to say that local law enforcement always abides by state and federal constitutional requirements, but rather that, among those cases in which prosecutors have elected to file charges, the vast majority are not saddled with constitutional infirmities under either state or federal law. See infra Section I.C.

See Daniel C. Richman, Federal Criminal Law, Congressional Delegation, and Enforcement Discretion, 46 UCLA L. Rev. 757, 786 (1999) (“As federal interest moves down to more episodic criminal activity, like street crimes, agencies become more dependent on local police departments, the only entities whose tentacles reach every street corner.”).

For this reason, many such cases are investigated by task forces composed of federal, local, and sometimes state officials. Examples include the Organized Crime Drug Enforcement Task Forces; the High Intensity Drug Trafficking Areas Program; and REDRUM (murder spelled backwards). See Organized Crime Drug Enforcement Task Forces, U.S. Dep’t Just., http://www.justice.gov/criminal/taskforces/ocdetf.html (last visited Dec. 11, 2013); High Intensity Drug Trafficking Areas (HIDTA) Program, White House, http://www.whitehouse .gov/ondcp/high-intensity-drug-trafficking-areas-program (last visited Dec. 11, 2013); Report of the Drug Enforcement Administration: 1990-1994, U.S. Dep’t Just. 82, http://www .justice.gov/dea/about/history/1990-1994.pdf (last visited Dec. 11, 2013).

Although the expansion of federal criminal law has effectively made federal prosecution of most any narcotics or violent crime possible, in practice there is a divide between “federal” and “local” matters. See Richman, supra note 1, at 82-83; see also Principles of Federal Prosecution, United States Attorneys’ Manual § 9-27.240 (Aug. 2002), http://www .justice.gov/usao/eousa/foia_reading_room/usam/title9/27mcrm.htm (describing, in broad terms, those factors federal prosecutors should consider in deciding whether to charge a case or decline in favor of another prosecuting authority). Absent extenuating circumstances, federal prosecutors tend to decline drug-trafficking cases that involve lower quantities of drugs. See Klein & Grobey,supra note 13, at 49;Michael Edmund O’Neill, Understanding Federal Prosecutorial Declinations: An Empirical Analysis of Predictive Factors, 41 Am. Crim. L. Rev. 1439, 1456 (2004). As a result, a fair number of cases begun as federal investigations end up in state court.

See Exec. Office for U.S. Att’ys, U.S. Dep’t of Justice, United States Attorneys’ Annual Statistical Report, Fiscal Year 2012, at 36-37 tbl.3, 85 tbl.14 (2012), http://www.justice.gov/usao/reading_room/reports/asr2012/12statrpt.pdf (reporting that, in FY 2012, 715 narcotics cases and 1,641 violent crime cases were declined by federal authorities in favor of prosecution by another authority or on another charge or for lack of federal interest, and 13,942 narcotics cases and 11,890 violent crime cases were filed in federal court that year). While some of these declined cases may have been investigated by state or local rather than federal authorities, the increasing prevalence of joint task forces in violent crimes and narcotics-trafficking cases has made jurisdictional boundaries in street crime investigations more fluid. See supra note 91 and accompanying text.

To say that a state bars conviction based solely on accomplice testimony is, in any event, not as stringent an evidentiary limitation as it might seem. In New York, for instance, the courts have expressly rejected the notion that the accomplice testimony rule requires independent evidence of guilt. People v. Reome, 933 N.E.2d 186, 188 (N.Y. 2010) (“The corroborative evidence need not show the commission of the crime; it need not show that defendant was connected with the commission of the crime. It is enough if it tends to connect the defendant with the commission of the crime in such a way as may reasonably satisfy the jury that the accomplice is telling the truth.” (quoting People v. Dixon, 131 N.E. 752, 754 (N.Y. 1921))). In the Reome case, the court upheld a defendant’s rape conviction based on a single accomplice identification corroborated by the victim’s description of the events, notwithstanding the fact that the victim did not identify the defendant and DNA evidence matched the three accomplices but not the defendant. Id.

See U.S. Sent’g Comm’n, Report to the Congress: Mandatory Minimum Penalties in the Federal Criminal Justice System 117 (2011); Daniel C. Richman, Cooperating Defendants: The Costs and Benefits of Purchasing Information from Scoundrels, 8 Fed. Sent’g Rep. 292, 293 (1996).

Jeffries & Gleeson, supra note 11, at 1105-06 (“No matter how well accomplices have been quarantined [from the other witnesses and evidence], there is always a link among them—the prosecution team. Defense attorneys frequently are forced to argue, explicitly or implicitly, that the accomplice witnesses have told a conveniently consistent story because the government put them up to it.”). And on the battleground of integrity, the authors argue, federal prosecutors have the upper hand. Id.

See sources cited supra note 13.

See sources cited supra note 12.

For a detailed examination of how and why inadequate resources in many urban district attorneys’ offices severely damage prosecutorial effectiveness, see Adam M. Gershowitz & Laura R. Killinger, The State (Never) Rests: How Excessive Prosecutorial Caseloads Harm Criminal Defendants, 105 Nw. U. L. Rev. 261, 279-97 (2011).

See Barbara Boland & Elizabeth Brady, U.S. Dep’t of Justice, Bureau of Justice Statistics, The Prosecution of Felony Arrests, 1980, at 1 (1985), https://www.ncjrs.gov /pdffiles1/bjs/97684.pdf (finding in a study of 28 urban county courts that “[a]pproximately 50% of all felony arrests the police make do not lead to a conviction but are rejected by the prosecutor before court charges are filed or are later dismissed in court”); Kathleen B. Brosi, U.S. Dep’t of Justice, Law Enforcement Assistance Admin., A Cross-City Comparison of Felony Case Processing 7 fig.2 (1979) (finding pre-filing declination rates in five major urban jurisdictions ranging between 18% and 48%); Malcolm M. Feeley, The Process Is the Punishment: Handling Cases in a Lower Criminal Court 127-29 (1979); Milton Heumann, Plea Bargaining: The Experiences of Prosecutors, Judges, and Defense Attorneys 100 (1978). The sources cited here are all regrettably dated, highlighting the need for renewed attention to local police-prosecutor relations.

See Daniel C. Richman & William J. Stuntz, Al Capone’s Revenge: An Essay on the Political Economy of Pretextual Prosecution, 105 Colum. L. Rev. 583, 600 (2005) (noting the political pressures on local prosecutors to prosecute the crimes most visible to voters: violent felonies, major thefts, and hard drugs).

See, e.g., Gordon Van Kessel, Adversary Excesses in the American Criminal Trial, 67 Notre Dame L. Rev. 403, 442 n.164 (1992) (“Elected state prosecutors often face bitter contests in which their win-loss record becomes a campaign issue.”).

Federal prosecutors decline cases for a variety of reasons, which frequently include lack of evidence, absence of criminal intent, request by the investigating agency, prosecution by another authority or for another crime, insufficient federal interest, and lack of federal jurisdiction. See Exec. Office for U.S. Att’ys, supra note 93, at 85 tbl.14. Notably, prosecution by another authority is a distinct reason given for declination; weak cases are declined because they are weak, not because they will be prosecuted by local authorities.

Most federal robberies (apart from bank robberies) are prosecuted under the Hobbs Act, 18 U.S.C. § 1951 (2012), enacted pursuant to the Commerce Clause, which prohibits obstruction of interstate commerce by robbery. As a practical matter, this limits federal robbery prosecutions to those involving commercial establishments and illegal narcotics or narcotics proceeds.

Others have commented upon the variety of incentives influencing federal prosecutorial case selection, and the tendency of federal prosecutors to choose more challenging cases. See, e.g., Edward L. Glaeser et al., What Do Prosecutors Maximize? An Analysis of the Federalization of Drug Crimes, 2 Am. L. & Econ. Rev. 259, 273 (2000) (noting that, compared to drug defendants in state courts, federal drug defendants tend to be a “wealthier, more prestigious set of criminals, who are more likely to have private lawyers”); Daniel C. Richman, Old Chief v. United States: Stipulating Away Prosecutorial Accountability?, 83 Va. L. Rev. 939, 966-67 (1997) (“The motivations of those attracted to a job for which the remuneration can be comparatively low, in a legal market where the opportunities to shirk abound, will be far more complex [than just conviction-maximization]. Much work remains to be done in exploring the personal ideologies of prosecutors, the influence of long-term economic self-interest, and the psychological aspects of prosecutors’ self-selection.”); William J. Stuntz, Self-Defeating Crimes, 86 Va. L. Rev. 1871, 1884-85 (2000) (arguing that federal prosecutors tend to seek out marginal cases because a prosecutor’s value in the private market is enhanced by trial experience).

See Klein & Grobey, supra note 13, at 7; Richman, supra note 3, at 397.

See supra note 97 and accompanying text.

I explore these dynamics further infra Part III.

Charlie Savage, Another High-Profile Failure for a Justice Dept. Watchdog, N.Y. Times, May 31, 2012, http://www.nytimes.com/2012/06/01/us/edwards-case-a-blow-to-justice-dept -corruption-unit.html.

Juliet Macur, Clemens Found Not Guilty of Lying About Drug Use, N.Y. Times, June 18, 2012, http://www.nytimes.com/2012/06/19/sports/baseball/roger-clemens-is-found-not -guilty-in-perjury-trial.html.

Leslie Wayne, Bribery Case Falls Apart, and Tactics Are Doubted, N.Y. Times, Feb. 23, 2012, http://www.nytimes.com/2012/02/24/business/fbi-bribery-case-falls-apart-and-raises -questions.html.

For an expanded discussion of this dynamic, see Stuntz, supra note 105.

Commentators have attributed the African Sting fiasco in particular to an erosion of juror trust, compounded by embarrassing text messages sent by FBI agents to the government’s informant, which “savage[d] the credibility and professionalism of [the] agents.” Del Quentin Wilber, Racy, Vulgar Texts Hurt Justice Department’s Largest Sting Operation Targeting Foreign Bribery, Wash. Post, Feb. 13, 2012, http://articles.washingtonpost.com /2012-02-13/local/35445544_1_text-message-fbi-agents-informant.

See Tyler, supra note 4; Tom R. Tyler & Yuen J. Huo, Trust in the Law: Encouraging Public Cooperation with the Police and Courts (2002); Tom R. Tyler, Procedural Justice, Legitimacy, and the Effective Rule of Law, 30 Crime & Just. 283 (2003). The perceived obligation to obey the law reflects a belief that the authority is entitled to regulate behavior, and it is therefore a component of an authority’s legitimacy. Tyler & Huo, supra, at 101-03.

In his seminal study of 1,575 adults in Chicago, Tyler assessed the independent impact on compliance of people’s perceptions of eleven different factors: obligation to obey the law and allegiance to or support for the relevant authority (legitimacy); likelihood of getting caught (deterrence); judgments of their peer group (peer disapproval); personal belief in the morality of the law in question (personal morality); evaluation of the quality of performance by the authority (evaluation); and background factors: gender, race, age, income, education, and political leanings. The results revealed that personal morality had the strongest impact on compliance, followed by gender, age, income, and then legitimacy. Peer disapproval had a moderately significant influence, and least influential were political leanings, deterrence, race, and evaluation. Tyler, supra note 4, at 59.

Comparing legitimacy to deterrence, Tyler found legitimacy five times more likely than deterrence to influence compliance. Id.

I say “might” because the efficacy of deterrence depends in large part on the likelihood people ascribe to getting caught. If punishment is harsh but highly uncertain, it is unlikely to deter. Certain punishment, regardless of its severity, will probably have a much greater impact on behavior—albeit substantially less of an impact than non-instrumental factors such as legitimacy. See Tyler, supra note 4, at 59-60; Paul H. Robinson & John M. Darley, The Utility of Desert, 91 Nw. U. L. Rev. 453, 458-64 (1997).

For discussions of the two theories in tandem, see Bowers & Robinson, supra note 18; and Tom R. Tyler & John M. Darley, Building a Law-Abiding Society: Taking Public Views About Morality and the Legitimacy of Legal Authorities into Account When Formulating Substantive Law, 28 Hofstra L. Rev. 707 (2000).

See Tyler, supra note 114, at 292-97 (collecting studies).

See Bowers & Robinson, supra note 18; Tom R. Tyler & Jonathan Jackson, Popular Legitimacy and the Exercise of Legal Authority: Motivating Compliance, Cooperation and Engagement (Yale Law & Econ. Research Paper No. 477, 2013), http://ssrn.com/abstract =2292517.

John Thibaut & Laurens Walker, Procedural Justice: A Psychological Analysis 67-96, 102-16 (1975) (finding that in laboratory studies of simulated trials, litigants were more accepting of verdicts from adversary trials than inquisitorial trials, the former of which they viewed as a fairer procedure, independent of the favorableness of the verdict); Tyler, supra note 4, at 104-08 (finding, in a study of 1,575 diverse adults in Chicago, that respondents assessed the legitimacy of police and courts primarily on the basis of the fairness of the procedures used, rather than the outcome obtained); Tyler & Huo, supra note 114, at 30, 53-90 (finding, in a study of 1,656 diverse adults in Oakland and Los Angeles who had had recent personal experiences with the police and/or local courts, that respondents were more willing to defer to authorities they believed had treated them fairly and whom they trusted as having fair motives).

Social psychologists call the former a “relational” view of justice, because it concerns the relationship between the individual and the authority, and the latter an “instrumental” view of justice. Tyler, supra note 4, at 276. The substantial research in this area does not discount entirely instrumental influences; it comes as no surprise that people are more accepting of favorable decisions, and are more inclined to believe favorable decisions to be a result of fair process. Tyler & Huo, supra note 114, at 55-56, 84. Rather, the research demonstrates that when assessed independently, process is the dominant factor shaping people’s attitudes and behavior toward authorities. Id. at 53-56, 84.

Tyler & Huo, supra note 114, at 54-96.

Id. at 177-97 (finding, based on four separate studies, that people’s evaluations of the quality of local police and courts were influenced most by perceptions of whether those authorities treated community residents fairly and with dignity, and cared about residents’ concerns).

Jason Sunshine & Tom R. Tyler, The Role of Procedural Justice and Legitimacy in Shaping Public Support for Policing, 37 Law & Soc’y Rev. 513, 532-34 (2003); Tyler & Jackson, supra note120.

Sunshine & Tyler,supra note 124, at 532.

Tyler & Huo, supra note 114, at 162; Sunshine & Tyler,supra note 124, at 522-23.

See Tyler & Huo, supra note 114, at 115-22; Yuen J. Huo et al., Superordinate Identification, Subgroup Identification, and Justice Concerns: Is Separatism the Problem; Is Assimilation the Answer?, 7 Psych. Sci. 40 (1996). Thus, for instance, identification with one’s city or country has been shown to influence perceptions of legitimacy.

Tyler, supra note 4, at 26, 29.

Id. at 25-26, 279-80.

Tyler & Huo, supra note 114, at 126; Tyler, supra note 114, at 287.

Tyler & Huo, supra note 114, at 131.

Id. at 126-28.

Id. at 131.

Tyler, supra note 4, at 37 (citing five separate studies); Bowers & Robinson, supra note 18, at 258-63. But see Christopher Slobogin & Laura Brinkley-Rubinstein, Putting Desert in Its Place, 65 Stan. L. Rev. 77, 101-03 (2013) (presenting empirical data indicating that personal agreement with criminal laws does not substantially influence compliance).

See Bowers & Robinson, supra note 18, at 256-63 (collecting studies). Tyler’s work, however, indicates that while people’s individual compliance with the law is highly influenced by the extent to which they agree with the law on a moral level (sometimes referred to as “personal morality”), people’s perceptions of legitimacy—i.e., their support for authorities and sense of obligation to obey them—and their willingness to defer to authority are not significantly influenced by distributive fairness. See Tyler, supra note 4, at 97, 103; Tyler & Huo, supra note 114, at 54-56; see also Tyler, supra note 114, at 292 (noting that although “distributive justice judgments have a role in shaping people’s reactions to their encounters with legal authorities . . . procedural justice judgments consistently are found to have the major influence”).

See Bowers & Robinson, supra note 18, at 217 (“To the extent that people see the system as in conflict with their judgments of justice, that acquiescence and cooperation is likely to fade and be replaced with resistance and subversion. . . . Witnesses may lose an incentive to offer their information or testimony. Citizens may fail to report crimes in the first instance. Jurors may disregard their jury instructions. Police officers, prosecutors, and judges may make up their own rules. And offenders may resist adjudication processes and punishments rather than participate in them.”).

Paul Robinson and John Darley give as an example the breakdown wrought by dissonant moral beliefs on abortion. See Robinson & Darley, supra note 117, at 482-83 (describing a trajectory from peaceful protest to unlawful killing of doctors providing abortion services); see also Tracey L. Meares, Norms, Legitimacy and Law Enforcement, 79 Or. L. Rev. 391, 399-400 (2000) (“[L]egitimacy is a more stable basis for voluntary compliance than is personal morality,” because “[w]hile greater legitimacy translates into more compliance whether or not compliance is in the personal interest of an individual, one’s personal moral schedule may or may not be in line with authoritative dictates.”).

See Bowers & Robinson, supra note 18, at 216 (“[P]eople’s shared intuitions of justice are not justice, in a transcendent sense. People’s shared intuitions can be wrong.”).

Robinson & Darley, supra note 117, at 473.

See Tom R. Tyler & Gregory Mitchell, Legitimacy and the Empowerment of Discretionary Legal Authority: The United States Supreme Court and Abortion Rights, 43 Duke L.J. 703, 712 (1994) (discussing how, because of its legitimacy as an institutional authority, the Supreme Court has been able to legitimate abortion policies that lack moral support from a large portion of the American public).

See generally Bowers & Robinson, supra note 18, at 237-40 & nn.113-27 (summarizing studies of the public’s and defendants’ views of various adjudicatory procedures, including plea bargaining, as well as views on courts more generally).

See Amy Farrell, Liana Pennington & Shea Cronin, Juror Perceptions of the Legitimacy of Legal Authorities and Decision Making in Criminal Cases, 38 Law & Soc. Inquiry 773, 794 (2013) (noting that in studies of real juries, defendants’ rights prevent researchers from posing pre-trial questions to jurors, thus obscuring the extent to which jurors post-trial responses were influenced by the trial itself).

Paula Hannaford-Agor & Valerie P. Hans, Nullification at Work? A Glimpse from the National Center for States Courts Study of Juries, 78 Chi-Kent L. Rev. 1249, 1264 (2003) (“Indeed, some psychologists would argue that it is unlikely that the majority of jurors would be able to say with any certainty which specific factors led them to their decision.” (citing R.E. Nisbett & L. Ross, Human Inference: Strategies and Shortcomings of Human Judgment (1980))).

See, e.g.,John R. Hepburn, The Objective Reality of Evidence and the Utility of Systematic Jury Selection, 4 Law & Human Behav. 89, 90 (1980) (discussing the difficulties of accounting for and isolating variables in jury research).

See Farrell, Pennington & Cronin, supra note 142, at 779, 783 (2013) (analyzing post-trial questionnaire responses from nearly 2,000 jurors who had collectively served on 210 separate non-capital felony cases in four different urban jurisdictions—Washington, D.C.; the Bronx, New York; Los Angeles County, California; and Maricopa County, Arizona—and finding that, holding certain other juror characteristics and case-specific variables constant, the probability of a juror favoring the prosecution prior to deliberations is 75% for jurors reporting high trust and confidence in police, and 47% for jurors reporting no trust and confidence in the police); Stephen P. Garvey, Paula Hannaford-Agor, Valerie P. Hans, Nicole L. Mott & G. Thomas Munsterman, Juror First Votes in Criminal Trials, 1 J. Empirical Legal Studies 371, 396-97 (2004) (analyzing the same dataset used by Farrell, Pennington, and Cronin and finding that in addition to evidentiary strength, the credibility of police testimony and juror beliefs about fairness of applicable laws and penalties all significantly affected jurors’ pre-disposition to the prosecution, as measured by jurors’ first pre-deliberation votes); Hepburn, supra note 144, at 97, 98 (analyzing the effect of juror attitudes and demographics on verdicts in simulated trials and concluding that “the strength of the evidence is relative, influenced by case-relevant juror attitudes” towards police and punishment).

Parenthetical summaries of jury studies necessarily gloss over complexities of the findings, which, as noted, must be approached with caution. The Farrell, Pennington & Cronin study in particular presented mixed findings: it measured the effects of juror trust and confidence in police on jurors’ predisposition to the prosecution prior to deliberations and on jurors’ first, pre-deliberation votes, and found a stronger correlation with respect to the former than the latter. Farrell, Pennington & Cronin, supra note 142,at 784, 787. The study further found a stronger correlation between trust in police and predisposition to the prosecution for black jurors than for white jurors, with the racial gap narrowing as black juror trust and confidence in the police increased. Id. at 788-90. But see Garvey et. al, supra, at 397-98 (finding, in a study of the same dataset, that race had a limited influence on juror decision-making except with respect to drug offenses tried in the District of Columbia). The Farrell, Pennington & Cronin study also measured the effect on juror predisposition and first juror votes of juror trust and confidence in the courts and found a reverse correlation: jurors reporting high trust in the courts were less predisposed to the prosecution than jurors reporting low trust in the courts. Id. at 784. (Trust in courts had no effect on juror first votes. Id. at 787.) The authors postulate that the inverse correlation between trust in courts and predisposition to the prosecution could reflect the former as evincing a strong belief in defendants’ due process rights, which might translate into a more robust application of reasonable doubt. Id. at 793-94. Likewise, jurors with low trust in courts may perceive courts as too lenient towards criminal defendants. Id. My focus in this Article is perceptions of the legitimacy of law enforcement (police, federal agents, and federal and local prosecutors) as opposed to courts.

Farrell, Pennington & Cronin, supra note 142, at 785; Garvey et al., supra note 145, at 396-97; Hannaford-Agor & Hans, supra note 143, at 1266-68 (analyzing the same dataset as Garvey et al. and finding that jurors’ attitudes about the fairness of applicable laws influenced juror decision-making).

See Dan M. Kahan, Social Influence, Social Meaning, and Deterrence, 83 Va. L. Rev. 349 (1997) (discussing the expressive function of criminal prosecution); Richman & Stuntz, supra note 101, at 586 (“[C]riminal litigation is not just a means of rationing criminal punishment. It is also a source of productive signals and valuable information.”); Robinson & Darley,supra note 117, at 472 (“[C]riminal law enforcement and adjudication activities send daily messages to all who read or hear about them. Every time criminal liability is imposed, it reminds us of the norm prohibiting the offender’s conduct and confirms its condemnable nature. . . . Further, every adjudication offers an opportunity to confirm the exact nature of the norm or to signal a shift or refinement of it.”).

See Jeffrey Fagan, Legitimacy and Criminal Justice: Introduction, 6 Ohio St. J. Crim. L. 123, 130 (2008) (“[S]trong enforcement of laws that the community greets with ambivalence can erode legitimacy. Under-enforcement for acts that are widely viewed as deserving punishment damages legitimacy, as well. Distributive justice concerns also damage legitimacy when there is (perceived) asymmetry between the harm of the act and the severity of the punishment.”).

These characterizations of the federal criminal justice system are, of course, broad generalizations. The ninety-four federal judicial districts are notoriously impervious to collective description, as are the hundreds of local urban county districts throughout the United States. Although an exacting empirical comparison (were it even possible) is beyond the scope of this Article, examples of these forum disparities appear throughout. See supra notes 5-7, 54, 68-79 and accompanying text; infra notes 211-220, 240 and accompanying text. In addition, the existing national data paint, in broad strokes, an overall picture worth noting.

In 2006, the most recent year for which data is available on felony case outcomes in the nation’s 75 largest counties, only 50% of defendants charged with a violent felony offense were convicted of a violent felony offense (an additional 11% were convicted of a misdemeanor offense) and only 59% of defendants charged with a felony drug offense were convicted of a felony drug offense (an additional 10% were convicted of a misdemeanor offense). The guilty plea rate was 45% for violent felony offenses (an additional 10% of felony defendants pled guilty to a misdemeanor offense), and 56% for felony drug offenses (an additional 9% pled guilty to a misdemeanor drug offense). Thomas H. Cohen & Tracey Kyckelhahn, Felony Defendants in Large Urban Counties, 2006, Bureau Just. Stat. 11 tbl.11 (2010), http://www.bjs.gov/content/pub/pdf/fdluc06.pdf. Now consider federal case disposition data for the same year: an 88.9% conviction rate (and an 81% plea rate) for violent felony offenses, and a 91.7% conviction rate (and 87.8% plea rate) for felony drug offenses. James C. Duff, Administrative Office of the United States Courts, 2006 Judicial Business of the United States Courts 249-50 tbl.D-4 (2007), in Judicial Business Archive, supra note 58, http://www.uscourts.gov/uscourts/Statistics/JudicialBusiness/2006 /appendices/d4.pdf.

The above comparison should be taken as a rough indication, not a measure. The urban county dataset measures both initial charges and outcomes in a statistical sample of felony cases filed in 2006 and resolved within one year, while the federal dataset measures outcomes of all federal cases terminated in 2006, regardless of when charged and without regard to the initial charge. (The county data thus does not account for 11% of cases within the statistical sample that were not resolved within one year—a factor that likely substantially under-represents trial outcomes, and thus likely over-represents conviction rates, because tried cases generally take longer to resolve than untried cases, and some tried cases end in acquittals whereas all pled cases end in convictions.) In addition, it bears noting that some of the disparity reflected in this comparison likely comes from differences in case processing: federal prosecutors are able to screen and decline cases almost entirely before initiating legal process, while in some counties, local prosecutors do not assess cases until after charging. In such counties, post-charging screening will invariably result in a higher number of dismissals.

Most of the empirical work on procedural legitimacy has focused on local criminal justice systems. There are inherent limitations of using studies of legitimacy in one context (local policing and local criminal cases) to advance theories in another (the federal criminal justice system). Cf. Tyler & Mitchell, supra note 140, at 747 (cautioning that empirical studies of local justice systems might not apply in the context of national-level legal authorities such as the Supreme Court). In the particular context of street crime, though, empirical research within local justice systems can inform our understanding of the federal criminal justice system. Because federal street crime prosecutions supplant local cases, they are a parallel alternative that can be contrasted with local prosecution at every stage. I do not apply research on local criminal justice systems to national-level political institutions that perform entirely different roles. Rather, I undertake a comparison of local and federal prosecution of substantially similar cases.

In some instances the state attorney general might investigate and/or prosecute, but rarely for episodic street crime.

For instance, federal prosecutions of felon-in-possession cases—the street crime cases most often adopted from state courts—will usually involve an ATF agent who acts as the “case agent.” Even though the ATF agent will not have had any involvement in the apprehension of the defendant, the agent will usually either testify himself or arrange for testimony by another ATF agent as to the interstate nexus element of the firearm.

Tom Tyler cites the Vietnam War as an example of a crisis that taxed citizens’ willingness to defer to institutions of government authority. See Tyler, supra note 4, at 271; see also Margaret Levi, Consent, Dissent, and Patriotism 177 (1997) (discussing how public “dissatisfaction with federal government policy increased markedly during the Vietnam era”). Other examples include Watergate; the FBI abuses uncovered by the Church Committee hearings; and, more recently, the atrocities committed by U.S. armed forces in the Abu Ghraib prison in Iraq, the federal government’s inadequate and ineffectual response to the Hurricane Katrina disaster, and the perceived abuses of the government’s “warrantless wiretapping” and NSA phone data collection programs.

The “federal authority” exercises power through a variety of different institutions, and public opinion surveys consistently show that Americans accord them very different levels of trust and respect (for instance, the Supreme Court tends to garner relatively high levels of respect and trust, whereas Congress does not). See, e.g., Jeffrey M. Jones, Confidence in U.S. Public Schools at New Low,Gallup Pol. (June 20, 2012), http://www.gallup.com/poll/155258 /Confidence-Public-Schools-New-Low.aspx. And context matters. Citizens’ trust in and respect for federal authorities, and willingness to obey them, will undoubtedly vary depending on whether the matter concerns social policy, fiscal policy, national security, the criminal law, or any other of the broad areas subject to federal authority. For instance, in one study of Supreme Court legitimacy and abortion rulings, respondents indicated a weak general feeling of obligation to federal authority: 60% disagreed with the statement, “I feel that I should accept the decisions made by government leaders in Washington even when I disagree with them,” 48% agreed that “[t]here are times when it is all right for people to disobey the government,” and 66% agreed with the statement, “I can think of situations in which I would stop supporting the policies of our government.” Tyler & Mitchell, supra note 140, at 807. The authors postulated that these weak feelings of an obligation to obey could be explained by the fact that some federal laws implicate moral or ideological differences among citizens. Id. at 761-62. This is, of course, the case for abortion, the subject of the study. It would not be the case for federal laws related to, for instance, robbery or murder—conduct that is equally prohibited by local, state, and federal law and for which there is no significant room for moral disagreement. For this reason, any consideration of the “federal authority” must be narrowly tailored to the precise federal entity and federal laws at issue.

There are a number of ways this happens. The government might introduce evidence of a defendant’s prior bad acts under Federal Rule of Evidence 404(b). In a conspiracy case, the government might introduce prior arrests of the co-conspirators to demonstrate the background of the conspiracy and the relationship of the co-conspirators. In a RICO prosecution, the government must introduce evidence of a “pattern of racketeering activity” (so-called “predicate acts”), which, in a violent crime or drug case, will almost always involve prior local arrests and/or convictions. See 18 U.S.C. §§ 1962(a), 1961(1)(A) (2012).

For instance, felon-in-possession cases typically establish the element of possession solely based on the testimony of the police who caught the defendant carrying a gun. The remaining elements—possession “in interstate commerce,” satisfied by proof that the gun was manufactured in another state, and the defendant’s prior conviction—are almost never in dispute.

In a study of 260 victims with cases pending in Criminal Court in Bronx County, New York, 57% of victims who had not been threatened feared reprisals, and 71% of all respondents said that they would feel threatened if the defendant were released on bail. Robert C. Davis, Barbara E. Smith & Madeline Henley, Victim/Witness Intimidation in the Bronx Courts: How Common Is It, and What Are Its Consequences?, Victim Services Agency 13 (1990), http://www.popcenter.org/problems/witness_intimidation/PDFs/Davis_etal_1990.pdf.

See Gary Gately, Baltimore Struggles to Battle Witness Intimidation: Prosecutors Say Violence, Threats Hinder Testimony, Bos. Globe, Feb. 12, 2005, http://boston.com/news/nation /articles/2005/02/12/baltimore_struggles_to_battle_witness_intimidation (reporting one drug gang’s murder of a woman and her family after she confronted gang members about selling drugs on her street and another gang’s hurling Molotov cocktails at the house of a woman who had called police to report drug dealing); Nancy Phillips, Craig R. McCoy & Dylan Purcell, Witnesses Fear Reprisals, and Cases Crumble, Phila. Inquirer, Dec. 14, 2009, http://articles.philly.com/2009-12-14/news/24988519_1_witnesses-murder-case-police-and -prosecutors (reporting that an alleged Philadelphia drug kingpin was accused of ordering at least seven witness-related murders in order to protect his drug business).

E.g., United States v. Wallace, 532 F.3d 126, 131 (2d Cir. 2008) (holding that a jury could properly infer intent to distribute narcotics from, among other things, defendant’s possession of a firearm); United States v. White, 969 F.2d 681, 684 (8th Cir. 1992) (“Because a gun is ‘generally considered a tool of the trade for drug dealers, [it] is also evidence of intent to distribute.’” (quoting United States v. Schubel, 912 F.2d 952, 956 (8th Cir. 1990))); United States v. Garrett, 903 F.2d 1105, 1113 (7th Cir. 1990) (“Intent to distribute has been inferred in cases where small amounts of drugs . . . have been possessed in conjunction with other indicia of drug distribution, such as a weapon.” (footnote omitted)).

See Gately, supra note 158 (describing Baltimore prosecutors’ estimate that witness intimidation occurs in 95% of all homicide cases in Baltimore, with the result that cases are dropped or defendants are acquitted, and Boston police’s estimate that two-thirds of Boston’s 64 murders in 2004 remained unsolved due to witness intimidation); David Kocieniewski, With Witnesses at Risk, Murder Suspects Go Free, N.Y. Times, Mar. 1, 2007, http://www.nytimes.com/2007/03/01/nyregion/01witness.html (describing how rampant witness intimidation in New Jersey and elsewhere has led to prosecutors’ refusals to even bring charges in violent crime cases that depend on a single witness’s testimony); Phillips, McCoy & Purcell, supra note 158 (describing a continuing epidemic of witness intimidation in Philadelphia, in which 13 witnesses were killed over 10 years, countless others are threatened into silence, and case after case collapses as a result).

See supra notes 121-123 and accompanying text.

See Fagan, supra note 148, at 127-28.

See Alexandra Natapoff, Underenforcement, 75 Fordham L. Rev. 1715, 1725-27 (2006).

Fagan, supra note 148, at 127-28.

See Randall Kennedy, Race, Crime, and the Law 71 (1997) (“Large numbers of blacks are convinced that, in general, law enforcement authorities value the safety and well-being of whites more than that of blacks.”); Fagan, supra note 148, at 127-28; Jeffrey Fagan & Garth Davies, Street Stops and Broken Windows: Terry, Race, and Disorder in New York City, 28 Fordham Urb. L.J. 457, 458-63, 482, 499 (2000); Ronald Weitzer, Racialized Policing: Residents’ Perceptions in Three Neighborhoods, 34 Law & Soc’y Rev. 129, 151-152 (2000) (demonstrating that blacks in low-income, high-crime neighborhoods perceived themselves as receiving inferior treatment by police as compared to both blacks and whites in higher-income, lower-crime neighborhoods).

See Fagan & Davies, supra note 165, at 462, 470-71 (describing collateral effects of New York’s order-maintenance policing on minorities’ trust in police); Bernard E. Harcourt & Jens Ludwig, Broken Windows: New Evidence from New York City and a Five-City Social Experiment, 73 U. Chi. L. Rev. 271, 272 (2006) (noting that New York, Chicago, and Los Angeles have all adopted policing strategies predicated on “more aggressive enforcement of minor misdemeanor laws, also known as ‘order maintenance’ policing”); Philip B. Heymann, The New Policing, 28 Fordham Urb. L.J. 407, 422-40 (2000) (discussing order maintenance policing efforts in New York, Boston, and Chicago); Weitzer, supra note 165, at 151 (describing similar dynamic in Washington, D.C.).

See Julian V. Roberts & Loretta J. Stalans, Public Opinion, Crime, and Criminal Justice 144-45 (1997) (collecting studies showing that “how police handle investigatory stops and requests for assistance from citizens shapes evaluations of police performance and perceptions of the fairness of police in general”).

Tyler & Huo, supra note 114, at 28-30, 132-33, 178 (finding, in a study of 1,656 adults in Oakland and Los Angeles, that 30% of the variance in people’s overall assessment of legitimacy was linked to their judgments about their own recent interactions with the police, and, applying same method to Tyler’s 1990 study of 1,575 adults in Chicago, finding that 24% of the variance in people’s overall assessment of legitimacy in that study was linked to their judgments about their own recent interactions with the police).

Three months after a local jury in Simi County, California acquitted four police officers charged in a videotaped beating of King, sparking days of rioting and nationwide protests, the U.S. Attorney for the Central District of California filed federal criminal civil rights charges against those officers. See Robert Reinhold, U.S. Jury Indicts 4 Police Officers in King Beating, N.Y. Times, Aug. 6, 1992, http://www.nytimes.com/books/98/02/08/home/rodney -indict.html. The federal trial resulted in the convictions of two officers and the acquittal of two officers. Seth Mydans, Verdict in Los Angeles; 2 of 4 Officers Found Guilty in Los Angeles Beating, N.Y. Times, Apr. 18, 1993, http://www.nytimes.com/books/98/02/08/home/rodney -guilty.html.

The U.S. Attorney for the Eastern District of New York prosecuted the police officers involved in a brutal beating and sexual assault of Louima, a Haitian immigrant, as well as those involved in a subsequent cover-up attempt. Of seven officers tried, six officers were convicted—the ringleader of the attack was convicted of assault and sentenced to 30 years’ imprisonment, while three others were convicted of conspiracy to obstruct justice (one of whom was also convicted of joining in the attack), and two others for lying to investigators. The conspiracy convictions and the conviction of the officer who joined in the attack were reversed on appeal. The Louima Ruling; Chronology of the Case, N.Y. Times, Mar. 1, 2002, http://www.nytimes.com/2002/03/01/nyregion/the-louima-ruling-chronology-of-the -case.html.

This is why the upended federal convictions in United States v. Bowen, the highly-publicized “Danziger Bridge” case (the federal prosecution of five New Orleans police officers for the September 4, 2005, shooting of six unarmed civilians, causing the death of two), has ramifications for federal authority that extend well beyond that case. See Order and Reasons at 124, United States v. Bowen, No. 10-204, (E.D. La. Sept. 17, 2013), 2013 WL 5233325, at *63 (ordering a retrial based on prosecutors’ online postings regarding the defendants, described as misconduct “committed by those with significant authority who act in the name of the ‘United States of America’ when they enter court and at all other times, and who have now left a fractured public trust”). The incident is all the more disheartening to New Orleans residents because the federal prosecution, like some other federal civil rights prosecutions, arose following a failed prosecution in state court. See Juliet Linderman, Reversal of Danziger Bridge Convictions a “Bitter Pill” for Hurricane Katrina Survivors, New Orleans Times-Picayune, Sept. 17, 2013, http://www.nola.com/crime/index.ssf/2013/09 /reversal_of_danziger_bridge_co.html.

See, e.g., John Marzulli, Guilty Brooklyn Cop: “I Don’t Know Why I Didn’t Commit Suicide Yet,” N.Y. Daily News, Oct. 15, 2012, http://www.nydailynews.com/new-york/brooklyn /guilty-brooklyn-don-didn-commit-suicide-article-1.1184296 (describing a guilty plea by police officer charged with federal civil rights violations for unlawfully handcuffing and pepper-spraying a man at a bar without justification); Don Terry, Philadelphia Shaken by Criminal Police Officers, N.Y. Times, Aug. 28, 1995, http://www.nytimes.com/1995/08/28/us /philadelphia-shaken-by-criminal-police-officers.html (detailing the federal prosecutions and resulting convictions of a string of Philadelphia police officers who unlawfully framed innocent citizens for drug crimes, a scandal that substantially undermined the community’s trust in the police); Mark Fazlollah & Aubrey Whelan, Grand Jury Probing Corruption in Phila. Police Narcotics Unit, Phila. Inquirer, Jan. 19, 2004, http://www.philly.com/philly /news/20140118_Federal_grand_jury_probing_corruption_in_Phila__police_narcotics_unit .html (describing federal grand jury investigation and prosecution into corruption offenses among certain officers in the Philadelphia Police Department’s narcotics unit).

See 42 U.S.C. § 14141 (2006) (authorizing the Department of Justice to investigate law enforcement agencies for engaging in a “pattern or practice” of civil rights violations and to impose systemic reforms upon those agencies, by court order if necessary). Since the law’s passage in 1994, the U.S. Department of Justice has exercised its oversight authority aggressively in actions against dozens of state and municipal police departments, including those in New Orleans, Los Angeles, Detroit, Pittsburgh, and the District of Columbia. See Civil Rights Div., Special Litigation Section Cases and Matters: Law Enforcement Agencies, U.S. Dep’t Just., http://www.justice.gov/crt/about/spl/findsettle.php#police (last visited Dec. 20, 2013); Police Exec. Research Forum, Civil Rights Investigations of Local Police: Lessons Learned 1 (2013) (noting that “more than 25 police departments have experienced some form of DOJ involvement in the last two decades,” which has primarily involved the investigation of improper use of force, unlawful stops and searches, and biased policing).

The federal “Witsec” program has generated television shows, e.g., In Plain Sight (USA Network); books, e.g., Pete Earley & Gerald Shur, Witsec: Inside the Federal Witness Protection Program (2002); and movies, e.g., Federal Protection (Chariot Communications 2002); Eraser (Warner Bros. 1996); Bird on a Wire (Universal Pictures 1990).

Gately, supra note 158 (quoting Nicole Krivda, a Baltimore prosecutor).

The late William Stuntz postulated that, because “metropolitan counties typically include both cities and close-in suburbs,” the rise in the white suburban population in the latter half of the twentieth century resulted in a political power-shift that has led to negative consequences for local urban justice systems. See Stuntz, supra note 3, at 2002. While Stuntz’s description of county scope is accurate with respect to certain counties (for instance, his examples of Chicago, Atlanta, Cleveland, and Detroit), it is not for a number of the nation’s most notoriously crime-ridden inner cities. Bronx County, in New York, is comprised entirely of the borough of the Bronx and includes no surrounding suburbs. The same is true for Philadelphia, Baltimore (which operates as an independent city with no county control), New Orleans, and Washington, D.C., as well as some smaller urban areas (Richmond, Virginia, the site of Project Exile, see supra notes 68-79 and accompanying text, among them). These so-called “consolidated city-counties” thus present real-world applications of some of the reform theories that Stuntz and other scholars have advanced, a point I take up in further detail in Section IV.A.

The United States Attorney and judges for the district are appointed by the President of the United States (usually selected for the President by the senators of the district’s state). Federal judges serve lifetime appointments, while U.S. Attorneys typically serve until the conclusion of the appointing President’s time in office.

The sole exception is the District of Columbia, in which the local and federal court districts are geographically coterminous.

Particularly populous districts, or those spanning large geographic areas, may have more than one courthouse location in the district, in which case each courthouse location may draw jurors from a narrower pool. For instance, the Southern District of New York is comprised of eight counties: two counties in New York City (New York County and Bronx County), and six in the suburbs. The two courthouses in Manhattan draw jurors from these two New York City counties and three surrounding counties (Westchester, Rockland, and Putnam counties), while the courthouse in White Plains draws jurors from six suburban counties (Westchester, Rockland, Putnam, Orange, Dutchess, and Sullivan). See Amended Plan for the Random Selection of Grand and Petit Jurors in the United States District Court for the Southern District Of New York, U.S. Dist. Ct., S. Dist. N.Y. (Feb. 2009), http://www.nysd .uscourts.gov/jury_handbook/juryplan_feb_2009.pdf.

The geographic breadth of federal districts results in less minority representation in federal jury venires as compared to local venires within the same district. See, e.g., Laura G. Dooley, The Dilution Effect: Federalization, Fair Cross-Sections, and the Concept of Community, 54 DePaul L. Rev. 79 (2004); Nancy Gertner, 12 Angry Men (and Women) in Federal Court, 82 Chi.-Kent L. Rev. 613 (2007).

Richman & Stuntz, supra note 101; Stuntz, supra note 3.

See Huo et al., supra note 127.

See id. at 42-43 (finding that persons proud to be part of their employer’s organization were more likely to value relational concerns in their interactions with supervisors than those less proud of the organization); see also Heather J. Smith & Tom R. Tyler, Justice and Power: When Will Justice Concerns Encourage the Advantaged to Support Policies Which Redistribute Economic Resources and the Disadvantaged to Willingly Obey the Law?, 26 Eur. J. Soc. Psychol. 171, 183, 191-92 (1996) (finding, in two studies, one of 352 white adults and one of 150 African-American adults, a significant correlation between strong self-identification as “American” and relational views of justice, as well as a significant correlation between weak self-identification as “American” and instrumental views of justice).

Tyler & Huo, supra note 114, at 116-22.

Yuen J. Huo, Procedural Justice and Social Regulation Across Group Boundaries: Does Subgroup Identity Undermine Relationship-Based Governance?, 29 Personality & Soc. Psychol. Bull. 336, 346 (2003) (“[E]ven in a highly volatile context fraught with ethnic tension—minorities’ encounters with the police and courts—identification with a diffuse social category such as the nation can facilitate cooperation with the directives of authorities whose responsibility it is to maintain social cohesion.”); Huo et al., supra note 127, at 45.

See Huo, supra note 185, at 336; Smith & Tyler, supra note 183, at 182-83.

See Tyler & Huo, supra note 114, at 30, 120 tbl.7.4.

See id. at 122 tbl.7.6 (identifying a beta weight of 0.31 in a regression of legitimacy against pride in the United States and identifying a beta weight of only 0.12 in a regression of legitimacy against pride in city).

The study tested perceptions of legitimacy of the police and courts. See id. at 28, 32-33. Although the study did not specifically limit itself to local police and courts, it appears to have been so limited in practice, simply as a result of the nature of most citizen encounters with law enforcement and the judicial system. See id. at 32-33; see also supra notes 167-168 and accompanying text (describing the importance of recent interactions in driving perceptions). The study thus indicates, interestingly, that self-identification on a national level translates into perceptions of legitimacy of local institutions.

In 2007-2013, the most recent years for which conviction rate data is available from the D.C. Superior Courts, the overall felony conviction rate has averaged 67% (and the conviction rate in jury trials, 62.2%) while the overall conviction rate average from 2007-2012 (2013 is not yet available) in federal court in the District of Columbia was 84.4% (and the conviction rate in jury trials was 92.8%). Compare Annual Reports and Documents, D.C. Cts., http://www.dccourts.gov/internet/about/orgperf/annualreports.jsf (click on the “Annual Report – Statistical Summary” for each of years 2007-2013 and scroll down to the “Criminal Division Case Activity” table) (last visited Feb. 18, 2014), with Judicial Business Archive, supra note 58 (click on each of the desired years (2007-2012), then click on “Table D-7 Defendants Disposed of, by Type of Disposition and District”). (For the D.C. Superior Court conviction rate, I calculated the total number of felony guilty pleas and jury and bench trial guilty verdicts as a percentage of total disposed felony cases; for the jury trial conviction rate, I calculated guilty verdicts as a percentage of total jury trials.) It should be noted that the federal court dataset tracks outcomes by individual defendants, while the D.C. Superior Court dataset tracks outcomes by cases (which may or may not include multiple defendants). Because of the potential for mixed verdicts or judgments in multi-defendant cases, disposition data tends to reflect a higher conviction rate if calculated by case outcome as opposed to defendant outcome. In other words, were these two court systems to use the same methodology, the disparity might be even more pronounced.

To be clear, there are a number of causes contributing to the conviction rate differential in the District of Columbia, including, most significantly, differences in the size and nature of the docket in each forum; in the total mix of factors, superordinate identity surely plays a relatively minor role. Nevertheless, the comparison is worth noting in light of other constants (or “controls”) unique to that district, namely, the jury pool and the prosecuting office are the same in federal and local court. This also complicates the influence of superordinate identity in D.C. Superior Court prosecutions, because federal and local authority are blended: federal prosecutors charge violations of the local city code that arise almost entirely from local arrests. See Exec. Office for U.S. Att’ys, U.S. Dep’t of Justice, United States Attorneys’ Annual Statistical Report: Fiscal Year 2011, at 14 (2011), http://www.justice.gov/usao/reading_room/reports/asr2011/11statrpt.pdf.

See Sam Dimeo, Political Status Referendum Falls Short in U.S. Virgin Islands, Associated Press, Oct. 11, 1993, http://www.apnewsarchive.com/1993/Political-Status-Referendum -Falls-Short-In-U-S-Virgin-Islands/id-a35f49515fb08b46bcd53142ad7fbf0a (observing that “[a]pathy appeared to be the strongest sentiment” in a 1993 political status referendum, where voter turnout was so low that the referendum was deemed invalid).

Judicial Business Archive, supra note 58, (click on each of the desired years (2008-2012), then click on “Table D-7 Defendants Disposed of, by Type of Disposition and District”) (showing that over the last five years, the conviction rate in the District of the Virgin Islands averaged 44.9%, as compared to an average conviction rate of 90.9% for the federal judicial system as a whole). While the District of the Virgin Islands has a much higher jury conviction rate, see id. (jury conviction rate of 69.8%), the total conviction rate reflects not just outcomes in tried cases, but importantly, prosecutors’ assessments of how cases will fare at trial. Generally speaking, a dismissal reflects the prosecutor’s belief that she will not secure a conviction at trial.

It is interesting to compare the Virgin Islands with Puerto Rico (86% conviction rate and 90.3% jury conviction rate over the last five years, id.), where in a 2012 political status plebiscite, sixty-one percent of voters who indicated a preference for one of three non-territory political status options wanted Puerto Rico to become a U.S. state. See Associated Press, Puerto Ricans Opt for Statehood in Referendum, USA Today, Nov. 7, 2012, http://www.usatoday.com/story/news/world/2012/11/07/puerto-rico-referendum/1689097. (Other unincorporated territories are Guam and the Northern Marianas Islands, which also have federal conviction rates well below average, but which have significantly smaller caseloads.) Again, conviction rate differences are the product of multiple contributing factors. The District of Puerto Rico has a high volume of cases, most of which involve narcotics charges, whereas the District of the Virgin Islands has a lower volume of cases, many of which involve violent crimes including firearm offenses. Judicial Business Archive, supra note 58 (click on each of the desired years (2008-2012), then click on “Table D-7 Defendants Terminated, by Major Offense and District”). As noted, the former are relatively easier to prove than the latter. See supra note 62 and accompanying text. For a discussion of the tangled and complex influences on Puerto Ricans’ and U.S. Virgin Islanders’ identification as Americans, including in particular issues of race and ethnicity, see Robert E. Statham, Jr., Colonial Constitutionalism: The Tyranny of United States’ Offshore Territorial Policy and Relations 25-64 (2002).

See Campbell Robertson, Last Chapter for a Court with a Place in History, N.Y. Times, Sept. 17, 2012, http://www.nytimes.com/2012/09/18/us/mississippi-courthouse-with-rich-civil -rights-past-set-to-close.html (recounting the 1967 federal conviction, by an all-white federal jury in Mississippi, of seven defendants for the brutal murders of three civil rights activists—the first of subsequent civil rights convictions against white defendants by all-white federal juries in the South).

The complexities of race, crime and the jury have generated a deep and thoughtful body of literature. See Kennedy, supra note 165, at 242-44; Paul Butler, Racially Based Jury Nullification: Black Power in The Criminal Justice System, 105 Yale L.J. 677 (1995); William T. Pizzi, Batson v. Kentucky: Curing the Disease but Killing the Patient, 1987 Sup. Ct. Rev. 97, 129-30 (1987) (explaining social psychological phenomenon of “ingroup-outgroup bias,” and its application to race-based peremptory challenges); Samuel R. Sommers and Phoebe C. Ellsworth, How Much Do We Really Know About Race and Juries? A Review of Social Science Theory and Research, 78 Chi.-Kent L. Rev. 997 (2003) (surveying the empirical literature); see also Fagan, supra note 148 (discussing tensions within minority communities regarding crime, law enforcement and imprisonment); Jeffrey Fagan & Tracey L. Meares, Punishment, Deterrence and Social Control: The Paradox of Punishment in Minority Communities, 6 Ohio St. J. Crim. L. 173 (2008) (same); Tracey L. Meares, Charting Race and Class Differences in Attitudes Toward Drug Legalization and Law Enforcement: Lessons for Federal Criminal Law, 1 Buff. Crim. L. Rev. 137 (1997) (providing empirical data suggesting African-American views on drug law enforcement are complex and sometimes in tension).

See Debra Livingston, Police Discretion and the Quality of Life in Public Places: Courts, Communities, and the New Policing, 97 Colum. L. Rev. 551, 553, 558-59 (1997).

The statute, enacted in 1994, mandated a sentence of at least 25 years to life imprisonment for any defendant convicted of a third felony offense, regardless of severity. See Cal. Penal Code §§ 667, 1170 (West 1995). A 2012 referendum modified the law to permit this enhanced sentence only for certain enumerated felonies. See Cal. Penal Code §§ 667, 1170 (West 2012); Statement of Vote November 6, 2012, General Election, Cal. Sec’y of State Debra Bowen 13 (2012), http://www.sos.ca.gov/elections/sov/2012-general/sov-complete.pdf.

See supra note 176.

See Tony Perry & Maura Dolan, Two Counties at Opposite Poles of “Three Strikes” Crime Debate: San Francisco Is Restrictive in Applying Law, San Diego Takes Hard Line, L.A. Times, June 24, 1996, http://articles.latimes.com/1996-06-24/news/mn-18004_1_san-diego -county.People v. Superior Court, 917 P.2d 628 (Cal. 1996) found that judges had authority to dismiss the allegation that a defendant had an otherwise qualifying prior felony conviction “in furtherance of justice.” Id. at 630.

Perry & Dolan, supra note 198.

See supra notes 68-70 and accompanying text.

See Virginia Exile Report, supra note 72, at 73.

Id. at 30-31. Almost all defendants charged with gun crimes in Richmond County elected to proceed before a judge rather than a jury. Id. at 30. Virginia is one of the few states in the nation to permit jury sentencing.

See Sam Roberts, Prison Isn’t as Mandatory as State’s Gun Laws Say,N.Y. Times, Jan. 20, 2013, http://www.nytimes.com/2013/01/21/nyregion/prison-not-as-mandatory-as-ny-state -gun-laws-say.html. Like California’s three strikes statute, see supra note 198, the New York gun statute permits judges to sentence a defendant to less than three-and-a-half years if the interests of justice so require. And as in California, that provision has been interpreted unevenly across different localities. See Roberts, supra; supra note 198.

Compare Urbanized Areas and Urban Clusters: 2010, U.S. Census Bureau, http://www .census.gov/geo/maps-data/maps/pdfs/thematic/2010ua/UA2010_UAs_and_UCs_Map.pdf (last visited Feb. 1, 2012), with Geographic Boundaries of the United States Courts of Appeals and the United States District Courts, U.S. Cts., http://www.uscourts.gov/uscourts/images /CircuitMap.pdf (last visited Dec. 11, 2013). Note that the U.S. Census Bureau defines an “urbanized area” as one containing more than 50,000 inhabitants and an “urbanized cluster” as one containing between 2,500 and 50,000 inhabitants. See Urban and Rural Classification, U.S. Census Bureau, http://www.census.gov/geo/reference/urban-rural.html (last visited Dec. 11, 2013). For a breakdown of the percentage of urban and rural areas in each county of the United States, both by population and land area, see Percent Urban and Rural in 2010 by State and County, U.S. Census Bureau,http://www2.census.gov/geo/ua/PctUrbanRural _County.xls (last visited Feb. 10, 2014). For a state-by-state breakdown of the specific counties within each federal judicial district, see Federal Judicial Districts, FedStats http://www.fedstats.gov/mapstats/fjd (last updated Apr. 20, 2007).

These maps and datasets illustrate the tremendous range of population densities within federal judicial districts. But even these fail to fully capture the reality of lived experience. Densely-populated suburban areas, such as Nassau County in New York State, can be classified by the U.S. Census as “urbanized areas.” See Urbanized Areas and Urban Clusters: 2010, supra; Percent Urban and Rural in 2010 by State and County, supra. Accordingly, it is worth noting the U.S. Attorneys’ Offices’ own descriptions of their districts. For example, the Eastern District of New York’s website describes Nassau and Suffolk counties as “suburban.” See U.S. Att’y’s Off. for E. District N.Y., http://www.justice.gov/usao/nye (last visited Dec. 11, 2013).

Compare Urbanized Areas and Urban Clusters, supra note 204, with Geographic Boundaries of the United States Courts of Appeals and the United States District Courts, supra note 204.

In this vein, it is also worth noting that where they exist, differences in jury selection procedures between federal and local districts may exacerbate this distinction in legislative-adjudicative alignment. All but seventeen federal districts compile juror venires from lists of registered voters. See Jody George, Deirdre Golash & Russell Wheeler, Fed. Jud. Ctr., Handbook on Jury Use in the Federal District Courts 11-12 (1989) (noting that a small number of districts with large minority populations have supplemented voter registration lists with other sources, including driver’s license records and state jury selection lists); John P. Bueker, Note,Jury Source Lists: Does Supplementation Really Work?, 82 Cornell L. Rev. 390, 391 (1997). While some states similarly select jurors solely from voter registration lists, many supplement from additional sources such as driver’s license records or income or property tax records. See Gregory E. Mize, Paula Hannaford-Agor & Nicole L. Waters, The State-of-the-States Survey of Jury Improvement Efforts: A Compendium Report, Nat’l Center for St. Cts. 13-14 (Apr. 2007). One effect of these selection differences may be minority representation on juries. But see Bueker, supra,at 392 (arguing that source supplementation has been shown not to enhance minority representation on juries). But another important effect is perceptions of legitimacy: to the extent that federal districts cull only from the ranks of registered voters, they are selecting from among those citizens who participate in the larger political process, and who therefore may be more predisposed to believing in the system’s legitimacy. (I am grateful to Hosea Harvey for pointing me to the issue of jury selection procedures.)

See supra notes 134-140 and accompanying text.

See supra notes 128-129 and accompanying text.

Research indicates that anywhere from forty to ninety percent of youths involved in the criminal justice system carry firearms. See Deanna Wilkinson & Jeffrey Fagan, Situational Contexts of Gun Use by Young Males in Inner Cities 6 (2000), https://www.ncjrs.gov/pdffiles1/nij/grants/194120.pdf; see also id. at 1-2 (“[G]uns have become an important part of the discourse of social interactions in modern urban life, with both symbolic meaning (power and control), social meaning (status and identity), and strategic importance. Getting and using a gun against another person has become a rite of passage into manhood, or at least into a respectable social identity within this context.”).

See generally Bureau of Alcohol, Tobacco, Firearms & Explosives, State Laws and Published Ordinances – Firearms (31st ed. 2010-2011), https://www.atf .gov/files/publications/download/p/atf-p-5300-5-31st-editiion/2010-2011-atf-book-final.pdf (collecting firearms laws and ordinances of all fifty states, effective through January 2011). In a number of states convicted felons may have their firearms rights restored under certain circumstances; such state reinstatements are recognized and permitted under federal law. See Michael Luo, Felons Finding it Easy to Regain Gun Rights, N.Y. Times, Nov. 13, 2011, http://www.nytimes.com/2011/11/14/us/felons-finding-it-easy-to-regain-gun-rights .html (reporting on reinstatement rules in 11 states). But see Jay Buckey, Firearms for Felons? A Proposal to Prohibit Felons from Possessing Firearms in Vermont, 35 Vt. L. Rev. 957, 957 (2011) (noting that “Vermont has no state law that bans firearms for felons”).

See New York State Violent Felony Offense Processing 2006-2010, Rep. N.Y. St. Div. Crim. Just. Services 6, 13 (2012), http://www.criminaljustice.ny.gov/crimnet/ojsa/nys-violent-felony -offense-processing-2006-2010.pdf (noting that as a result of the law, which went into effect in November 2006, prison sentences for gun possession crimes in New York City increased by 126.8% between 2006 and 2010, even as the total number of weapons possession cases in the city declined by 18%); N.Y. Penal Law §§ 70.02.3(c), 70.02.4(b) (McKinney 2013).

See supra note 203 and accompanying text.

See Virginia Exile Report, supra note 72, at 37-38, 53, 58-59.

See 18 Pa. Cons. Stat. Ann. §§ 6105(a.1)(1), 1103(2) (West 2000 & Supp. 2013) (providing statutory penalties for the offense); Summary of Sentences by Offense Category and County, Pa. Sent’g Comm’n 2010 (on file with author and available from Pennsylvania Commission on Sentencing) (for 18 P.S.A. § 6105); Pennsylvania sentencing Guidelines Implementation Manual 36, 53 (6th ed. 2008); U.S. Sentencing Guidelines Manual § 2K2.1(a)(6) & Sentencing Table (2010) (guidelines range from 15 months to ten years depending on the circumstances of the offense and the defendant’s criminal history). I cite to the manuals in effect as of 2010, the latest year for which sentencing data is available from the Pennsylvania Commission on Sentencing. With respect to the felon-in-possession offenses, there have been no subsequent amendments to the sentencing guidelines in either Pennsylvania or the federal system. See Pennsylvania sentencing Guidelines Implementation Manual 47, 71 (7th ed. 2012); U.S. Sentencing Guidelines Manual § 2K2.1(a)(6) & Sentencing Table (2013).

Hogan, supra note 67, at tbl.D-5, (showing a median sentence of 46 months).

See id.

See 2012 Sourcebook, supra note 66, at tbl.12 (reporting that just 2.9% of defendants charged with firearms-related crimes received only probation). This statistic, moreover, includes those defendants who received a sentence reduction for cooperation with the government.

I chose these states because (i) they have publicly-available, offense-specific sentencing data for felon-in-possession cases in a mode that is readily comparable to the equivalent federal crime (on the dearth of offense-specific state sentencing data, seeMarc 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, 1354-55 (2005)); and (ii) these states, or cities within the states, have made visible efforts to combat gun crime through the use of these statutes. The dynamic in illegal gun possession cases in these states, though, has been observed by researchers more broadly. See Edmund F. McGarrell et al., Nat’l Insts. Justice, Project Safe Neighborhoods - A National Program to Reduce Gun Crime: Final Project Report 16-17 & n.15 (2009) (“PSN officials in many jurisdictions report that for years illegal possession of a firearm by a felon or concealed carrying offenses, and even crimes committed with a firearm present but no shooting, were routinely treated as non-violent offenses with high rates of dropped charges, dismissed cases, and suspended sentences.”).

Bureau of Justice Statistics, Felony Sentences in State Courts, 2004, U.S. Dep’t Just. 3 (2007), http://www.bjs.gov/content/pub/pdf/fssc04.pdf.

2012 Sourcebook, supra note 66, at tbl.12 (showing that for most violent crimes no defendants received probation, for robbery 1.9% of defendants received probation, and for assault 10.6% of defendants received probation; the table does not indicate whether these defendants were cooperators).

See sources citedsupra notes 10, 13.

It is a common feature of many urban justice systems to assign judges to either a criminal or civil docket. This is the practice, for instance, in New York, Chicago, Philadelphia, Baltimore, Washington, D.C., and San Francisco.

There are certainly other factors at work, too. Federal prosecutors tend to prosecute defendants with more egregious criminal records, which naturally garner more severe sanction. But to some degree this explanation is circular: federal prosecutors choose such defendants precisely because they have not, despite their records, received sufficient (in the prosecutor’s view) sentences in local court. The examples offered in the Introduction illustrate this dynamic.

See McGarrell et al., supra note 218, at 16-17 & n.15 (discussing the dynamic in some jurisdictions when gun cases become routine). For studies of how courtroom norms develop, particularly with respect to the disposition of frequently-prosecuted crimes, seeRoy B. Flemming, Peter F. Nardulli & James Eisenstein, The Craft of Justice: Politics and Work in Criminal Court Communities (1992); and Peter F. Nardulli, James Eisenstein & Roy B. Flemming, The Tenor of Justice: Criminal Courts and the Guilty Plea Process (1988).

For a theoretical account of punishment’s expressive power to influence social norms, see Kahan, supra note 147.

See Decker & McDevitt, supra note 52, at iii, 8; Press Release, supra note54 (discussing the Maryland U.S. Attorney’s Office’s use of “Federal Letters of Intent to Prosecute”).

See Meares, supra note 137; Robinson & Darley, supra note 117. But see Kahan, supra note 147.

See, e.g., Americans Decry War on Drugs, Support Legalizing Marijuana,Angus Reid Pub. Opinion 1-2 (June 6, 2012), http://www.angus-reid.com/wp-content/uploads/2012/06/2012 .06.06_Drugs_USA.pdf (finding, in a nationwide survey of 1,017 adults, that less than 10% of respondents favored legalizing powder cocaine, crack, heroin, or methamphetamine, while over 50% favored legalizing marijuana); 11% Favor Legalizing, Regulating Cocaine,Rasmussen Rep. (May 21, 2012), http://www.rasmussenreports.com/public_content /lifestyle/general_lifestyle/may_2012/11_favor_legalizing_regulating_cocaine (finding, in a survey of 1,000 likely voters, that only 11% of respondents favored legalizing cocaine). I discuss the federal response to marijuana-legalization trends infra notes 265-266 and accompanying text.

Peter H. Rossi & Richard A. Berk, Just Punishments: Federal Guidelines and Public Views Compared 4-6, 13-15 (1997).

Id. at 4, 33. Respondents were given vignettes on a series of crimes, each setting forth specific details about the nature of the crime and its impact, as well as background information about the offender.Id. at 34-50, 55. For each vignette, respondents were asked to specify the type and amount of punishment they would impose. Id. at 40. By varying specific features in each vignette (for example, the use of a gun in a robbery, or whether the defendant had children), the researchers ascertained the weight respondents accorded to particular features, thus providing an account of what factors the public considers important in apportioning punishment. Id. at 36-37. The researchers then assessed the extent to which the Guidelines reflected public views of sentences and sentencing rationales.Id. at 75-79.

Id. at 79-82.

Id. at 98. An example from the study illustrates the point. In a vignette in which the defendant is convicted of robbing a convenience store, the median sentence of respondents was 5 years; the mean was 8.5 years.Id. at 64 tbl.4.5. These appear to approximate the Guidelines median and mean, which are 6.5 years and 7.6 years, respectively. Id. at 92 tbl.5.5. Yet individual responses varied considerably. Of the 596 respondents to this vignette, 1.6% chose probation, 0.3%, chose death, 4% chose life imprisonment, and the remainder varied within those extremes. Id. at 64 tbl.4.5.

Id. at 149 (finding “remarkable agreement between average respondent sentences and guidelines sentences, not just overall but in sentencing determinants. Time and again, the patterns [of guidelines and respondent sentences for given crimes] were nearly identical”). Conformity between respondents’ views and the Guidelines’ penalties, though, varied for each vignette and across different crimes. With respect to a street robbery, for instance, the respondent median was 4.1 years lower than the Guidelines median, but there was almost no mean difference. Id. at 88 tbl.5.4. For firearms offenses, the respondent median was 0.5 years lower than the Guidelines median, but the respondent mean was 2.5 years higher than the Guidelines mean. Id.

Crack penalties “represent[ed] the greatest disparity between the guidelines and respondents” in the Rossi and Berk study.Id. at 95. The median respondent sentence for a crack cocaine offense vignette was 10 years, while the median Guidelines sentence for the same vignette was 22 years. Id. at 92 tbl.5.5.

Results of Survey of United States District Judges: January 2010 Through March 2010, U.S. Sent’g Comm’n (June 2010) [hereinafter U.S.S.C. Survey], http://www.ussc.gov/Research _and_Statistics/Research_Projects/Surveys/20100608_Judge_Survey.pdf.

Whereas the Rossi and Berk study used a factorial survey to gather data on what penalties respondents believed were appropriate for a given set of facts and then compared the results to the sentence that would be calculated under the Federal Sentencing Guidelines on those facts, see Rossi & Berk, supra note 229, the Sentencing Commission’s survey instead asked judges whether, as a general matter, they believed the Guidelines and certain statutorily mandated penalties were “appropriate,” “too high,” or “too low,” U.S.S.C. Survey, supra note 235, at tbls.1, 8.

U.S.S.C. Survey, supra note 235, at tbls.1, 8.

Seventy-six percent thought the mandatory minimum penalties for crack cocaine were too high, and 70% thought the Guidelines penalties were too high. Id. at tbls.1, 8.The survey was conducted from January through March 2010, while the Fair Sentencing Act of 2010, which greatly reduced the mandatory minimum penalties for crack offenses, was pending in Congress but had not yet been passed. See Bill Summary and Status,111th Congress, S. 1789, Libr. of Congress: Thomas, http://thomas.loc.gov/cgi-bin/bdquery/z?d111:SN01789 :@@@D&summ1& (last visited Mar. 11, 2014).

71% of judges believed the mandatory minimum penalty for receipt of child pornography was too high, and 69% and 70% of judges believed the Guidelines penalties were too high for the offenses of, respectively, receipt and possession of child pornography. U.S.S.C. Survey, supra note 235, at tbls.1, 8.

See 2012 Annual Report, U.S. Sent’g Comm’n 44 (2012), http://www.ussc.gov/Research _and_Statistics/Annual_Reports_and_Sourcebooks/2012/ar12toc.htm. In only 13.9% of cases did judges sentence defendants to a term below the applicable Guidelines range, where the below-range sentence was pursuant to Booker rather than a guidelines’ departure provision and not at the prosecution’s request; of those cases, the median decrease from the minimum guidelines term was 14 months, a median decrease of 35.1%. Id. For street crimes (comprising the offense categories of murder, manslaughter, robbery, burglary, drugs and firearms), the average rate of Booker-based below guidelines sentences was 13.5%, and the median percent decrease from the Guidelines minimum ranged between 28% (for firearms) and 100% (for simple drug possession). See 2012 Sourcebook, supra note 66, at tbls.27, 31C. Local prosecutors and judges do not necessarily accord state penalty-drafters the same level of deference. See supra notes 196-203 and accompanying text.

Dorsey v. United States, 132 S. Ct. 2321, 2328-29 (2012) (reciting the history of crack laws, subsequent public opposition, and the Fair Sentencing Act of 2010); William Spade, Jr., Beyond the 100:1 Ratio: Towards a Rational Cocaine Sentencing Policy, 38 Ariz. L. Rev. 1233, 1282 (1996) (describing the critical response to the disparity among African-American commentators and academics, as well as federal judges, prosecutors, and juries). Despite Reno’s criticism, her Justice Department did not ultimately support proposed amendments to the laws. It is worth noting as well that criticism, though widespread, was not uniform. See Randall Kennedy, The State, Criminal Law, and Racial Discrimination: A Comment, 107 Harv. L. Rev. 1255, 1256 (1994) (arguing that “the state was justified in penalizing possession of crack cocaine more harshly than possession of powdered cocaine”); Meares, supra note 194, at 138 n.2 (citing sources supporting the federal crack penalty scheme).

Spade, supra note 241, at 1281-82; William J. Stuntz, Race, Class, and Drugs, 98 Colum. L. Rev. 1795, 1827 (1998) (reporting anecdotal evidence of jury nullification).

See Fair Sentencing Act of 2010, Pub. L. 111-220, 124 Stat. 2372 (codified as amended at 21 U.S.C. § 841(b)(1) (2012)) (reducing the powder/crack ratio of the weight required to generate certain mandatory sentences from 100:1 to 18:1).

In the year preceding Booker, 29.5% of defendants convicted of crack cocaine offenses earned a government-sponsored downward departure based on substantial assistance to the government, compared to 26.9% of powder cocaine defendants, 21.5% of heroin defendants, 31% of methamphetamine defendants, and 17% of marijuana defendants. See U.S. Sent’g Comm’n, Annual Sourcebook of Sentencing Statistics tbl.45 (9th ed. 2004), http://www.ussc.gov/Research_and_Statistics/Annual_Reports_and_Sourcebooks /2004/sbtoc04.htm. That distribution of cooperation rates among drug offenders has not substantially changed following either Booker or the Fair Sentencing Act of 2010 (FSA), which reduced the crack/powder penalty disparity. See 2012 Sourcebook, supra note 66, at tbl.45 (reporting substantial assistance rates of 27.1% for crack defendants, 28.2% for powder cocaine defendants, 31.5% for methamphetamine defendants, 24.3% for heroin defendants, and 16.3% for marijuana defendants).

See U.S. Sent’g Comm’n, Report to the Congress: Cocaine and Federal Sentencing Policy 51 (2007).

In the ten years preceding the FSA, crack cocaine cases as a percentage of all narcotics cases remained consistently above 20%, with a low of 20.2% in 2002, and a high of 24.3% in 2008. See Annual Reports and Sourcebook Archives, U.S. Sent’g Comm’n, http://www.ussc.gov /Research_and_Statistics/Annual_Reports_and_Sourcebooks/Archives.cfm (click on the link in the “Sourcebook” column for each of the desired years (2000-2010), then click the link for “Primary Drug Type of Offenders Sentenced Under Each Drug Guideline,” which corresponds to Table 33) (last visited Dec. 11, 2013). It is only since the passage of the FSA that federal crack prosecutions have declined. See id. at year 2011 (17.3%) and 2012 (13.8%) (2012 link is at http://www.ussc.gov/Research_and_Statistics/Annual_Reports_and _Sourcebooks/2012/Table33.pdf). During the pre-FSA years, moreover, the government agreed to very few below-guidelines sentences for non-cooperating defendants prosecuted for crack offenses: between 2006, when the Sentencing Commission began reporting government-sponsored below-range sentences for non-cooperating defendants, and 2010, the year the FSA was enacted, the government only agreed to below-guidelines sentences in, on average, 3.2% of crack cases. See Sourcebook Archives, supra, (click on the link in the “Sourcebook” column for each of the desired years (2006-2010), then click the link for “Sentences Relative to the Guideline Range for Drug Offenders by Each Drug Type,” which corresponds to Table 45) (last visited Dec. 11, 2013). Unfortunately, there is no data analysis available on national conviction rates for crack offenses.

U.S. Sent’g Comm’n, Report on the Continuing Impact of United States v. Booker on Federal Sentencing, Part C 6, 7 (2012) [hereinafter Booker Crack Cocaine Report], http://www.ussc.gov/Legislative_and_Public_Affairs/Congressional_Testimony _and_Reports/Booker_Reports/2012_Booker/Part_C4_Crack_Cocaine_Offenses.pdf; U.S. Sent’g Comm’n, Report on the Continuing Impact of United States v. Booker on Federal Sentencing, Part C (Analysis of Powder Cocaine Trafficking Offenses) 6, 7 (2012), http://www.ussc.gov/Legislative_and_Public_Affairs/Congressional_Testimony _and_Reports/Booker_Reports/2012_Booker/Part_C3_Powder_Cocaine_Offenses.pdf.

See Booker Crack Cocaine Report, supra note 247, at 7 (showing a decrease of non-Government sponsored below-Guidelines cases from approximately 30% to approximately 20%).

Although child pornography is not a street crime, and for this reason falls outside this Article’s focus, I consider it here because any consideration of the crack penalties suggests a comparison to the other documented moral credibility gap in the federal system. As discussed infra notes 260-262 and accompanying text, a number of differences in the nature of these crimes makes comparison difficult. Despite these limitations, though, it is nevertheless worth considering other, non-offense-specific differences that might also explain the somewhat different paths of these two moral credibility gaps.

See Carol S. Steiker, Lessons from Two Failures: Sentencing for Cocaine and Child Pornography Under the Federal Sentencing Guidelines in the United States, 76 Law & Contemp. Probs. 27, 37 (2013). As noted, Rossi and Berk did not canvass public views on pornography penalties, which were at the time substantially less severe than they were 15 years hence, at the time of the Sentencing Commission’s survey of federal judges. See supra note 239 and accompanying text.

U.S. Sent’g Comm’n, Report to Congress: The History of the Child Pornography Guidelines 6-8 (2009), http://www.ussc.gov/Research_and_Statistics/Research_Projects /Sex_Offenses/20091030_History_Child_Pornography_Guidelines.pdf.

Steiker,supra note 250, at 37-44.

Id. at 38.

Thus far, there have been no bills proposed in Congress to reduce child pornography penalties. In comparison, bills to amend the crack/powder disparity were introduced in Congress almost yearly beginning in 1993. See Steiker, supra note 250, at 30.

Prosecutorial Remedies and Other Tools to End the Exploitation of Children Today Act of 2003, Pub. L. No. 108-21, 117 Stat. 650 (2003) (codified in scattered sections of 18, 28, 42, and 47 U.S.C.).

2012 Sourcebook, supra note 66, at tbl.27.

2012 Sourcebook, supra note 66, at tbl.31C.

See U.S. Sent’g Comm’n, Federal Child Pornography Offenses 134-35 (2012), http://www.ussc.gov/Legislative_and_Public_Affairs/Congressional_Testimony_and _Reports/Sex_Offense_Topics/201212_Federal_Child_Pornography_Offenses/Full_Report _to_Congress.pdf (observing that government-sponsored below range sentences have steadily increased from 2.1% in 2005 to 11.2% in 2010).

See A.G. Sulzberger, Defiant Judge Takes on Child Pornography Law, N.Y. Times, May 21, 2010, http://www.nytimes.com/2010/05/22/nyregion/22judge.html (reporting how, following a guilty verdict in a federal jury trial for possession and receipt of child pornography, Judge Jack Weinstein informed the jurors of the penalties that would result; learned that a number of jurors would have voted to acquit had they been aware of the penalties; and subsequently dismissed the verdict and ordered a new trial); Benjamin Weiser, A Judge’s Struggle to Avoid Imposing a Penalty He Hated, N.Y. Times, Jan. 13, 2004, http://www.nytimes.com/2004/01/13/nyregion/a-judge-s-struggle-to-avoid-imposing -a-penalty-he-hated.html (reporting how, in a federal trial for advertising child pornography, then-district Judge Gerry Lynch pledged to advise jurors, pre-deliberations, of the applicable ten-year mandatory penalty—a practice ultimately disallowed by the Second Circuit).

Compare U.S. Sent’g Comm’n, supra note 258, at 143 (reporting that, in 2010, over 80% of child pornography offenders fell within the Guidelines’ lowest criminal history category, Category I), with U.S. Sent’g Comm’n, Annual Sourcebook of Sentencing Statistics tbl.37 (15th ed. 2010) [hereinafter 2010 Sourcebook], http://www.ussc.gov /Research_and_Statistics/Annual_Reports_and_Sourcebooks/2010/sbtoc10.htm (reporting that 28.3% of crack offenders fell within the Guidelines’ highest criminal history category, Category VI, while 22.2% fell within Category I and the rest were distributed fairly evenly across the remaining categories). The differences in criminal history, though, don’t entirely explain the greater adherence to the Guidelines in crack cases given that the applicable Guidelines range takes into account an offender’s criminal history.

Compare U.S. Sent’g Comm’n, supra note 258, at 143 (reporting that, in 2010, approximately 90% of child pornography defendants were white), with 2010 Sourcebook, supra note 260, at tbl.34 (reporting that, in 2010, 78.5% of crack defendants were black); see also Michelle Alexander, The New Jim Crow 102-05 (2010) (describing the rise of racial bias in crack cocaine prosecutions). The primary reason a large majority of judges objected to federal crack penalties, though, is because of the penalties’ racially disparate impact—a circumstance that would seem to undercut race-based theories of relatively greater Guidelines-adherence in crack cases. Moreover, post-Booker sentencing practices in crack cases on the whole mitigated racial disparities wrought by the federal crack laws. See Amy Baron-Evans & Kate Stith, BookerRules, 160 U. Pa. L. Rev. 1631, 1688 (2012) (“By imposing below-guideline sentences that they could not have imposed under the mandatory guidelines, in fiscal year 2010 alone, judges spared more than 860 black defendants sentenced under the crack or career offender guidelines over 3300 years of unnecessary incarceration.”).

See Steiker, supra note 250, at 28-33, 37-41 (discussing the different trajectories of congressional and Commission penalty-drafting with respect to crack and child pornography).

The FSA was indeed an effort to engender greater credibility for federal law and for the law enforcement agencies (both federal and local) that enforce it. See Restoring Fairness to Federal Sentencing: Addressing the Crack-Powder Disparity: Hearing Before the Subcomm. on Crime & Drugs of the S. Comm. on the Judiciary, 111th Cong. 26 (2009)(statement of John F. Timoney, Chief of Police, Miami Police Dep’t) (advocating for the elimination of the crack/powder penalty disparity on the ground that “police departments across America face a much more difficult challenge gaining the trust of their communities if there are glaring inequities in the justice system that are allowed to persist”); id. at 4-6 (statement of Lanny A. Breuer, Assistant Att’y Gen., Criminal Div., U.S. Dep’t Justice) (“Public trust and confidence are essential elements of an effective criminal justice system. Our laws and their enforcement must not only be fair, but they must also be perceived as fair. . . . We also believe that the [sentencing] structure [for crack offenses] is especially problematic because a growing number of citizens view it as fundamentally unfair.”).

I explore the significance of this observation to local justice systems infra Subsection IV.A.2.

The Justice Department’s “Smart on Crime” initiative directs federal prosecutors to consider alternatives to incarceration for low-level, non-violent offenders.See U.S. Dep’t of Justice, Smart on Crime: Reforming the Criminal Justice System for the 21st Century (2013), http://www.justice.gov/ag/smart-on-crime.pdf. The Department’s new marijuana policy, implemented in the wake of a series of state ballot referendums that legalized marijuana possession for medicinal, and then recreational, purposes, advises federal prosecutors to refrain from enforcing federal marijuana laws that conflict with the laws in those states, except where federal enforcement is necessary to address certain enumerated federal interests (such as preventing distribution for the benefit of gangs and cartels, preventing violence and drugged driving, and preventing distribution to minors or to states where marijuana remains illegal). See Memorandum on Guidance Regarding Marijuana Enforcement from Deputy Att’y Gen. James M. Cole to All U.S. Att’ys (Aug. 29, 2013), http://www.justice.gov/iso/opa/resources/3052013829132756857467.pdf.

See Att’y Gen. Eric Holder, Remarks at the Annual Meeting of the American Bar Association’s House of Delegates (Aug. 12, 2013), http://www.justice.gov/iso/opa/ag /speeches/2013/ag-speech-130812.html (“Because [mandatory minimums] oftentimes generate unfairly long sentences, they breed disrespect for the system. When applied indiscriminately, they do not serve public safety. They—and some of the enforcement priorities we have set—have had a destabilizing effect on particular communities, largely poor and of color. And, applied inappropriately, they are ultimately counterproductive. . . . Ultimately, this is about much more than fairness for those who are released from prison. It’s a matter of public safety and public good.”); see also Conflicts Between State and Federal Marijuana Laws: Hearing Before the S. Comm. on the Judiciary, 113th Cong. 3 (2013) (statement of James M. Cole, Deputy Att’y Gen.),http://www.judiciary.senate.gov/pdf/9 -10-13ColeTestimony.pdf (describing the new federal marijuana enforcement policy, which largely defers to state laws and enforcement practices in states that have legalized or partly legalized marijuana, as “consistent with our efforts to maximize our investigative and prosecutorial resources during this time of budget challenges, and with the more general message the Attorney General delivered last month to all federal prosecutors, emphasizing the importance of quality priorities for all cases we bring, with an eye toward promoting public safety, deterrence, and fairness”).

See supra notes 49-51 and accompanying text.

See supra notes 37-38 and accompanying text.

Rossi & Berk, supra note 229, at 167-206. Among demographic factors, they considered age, race/ethnic group, gender, income, education, and residential locus (both geographic region and community size). Among experiential factors, they considered the respondent’s level of contacts with the justice system, whether the respondent had ever been a crime victim, and the respondent’s perception of the crime problem in his or her community. Among attitudinal factors, they considered the respondent’s views on the level of rights that should be accorded an accused, the level of civil rights that should be accorded minorities, the amount of welfare benefits that should be accorded the poor, the amount of investigative freedom that should be given to police, the level of attention that should be given to environmental pollution, and how the respondent self-identified politically. Id.

Rossi & Berk, supra note 229, at 200-01 tbl.8.17.

Thus, Rossi and Berk break down what I define as street crimes into separate categories of “drug trafficking,” “drug possession,” and “street crimes.” As a result, some of the “drug trafficking” crimes they study might fall under the categories of international or interstate drug trafficking, as opposed to street-level dealing. My use of the term “street crimes” when referring to the Rossi and Berk data reflects their definition, not mine.

That is, the factors with p-values less than 0.05.

Respondents from New England tended to give the least severe sentences; respondents in the West South Central (Texas, Oklahoma, Arkansas, Louisiana) and East South Central (Kentucky, Tennessee, Alabama, Mississippi) regions gave the most severe sentences. Id. at 200-01 tbl.8.17 (respondents in both these regions imposed over 6.5 years more than New Englanders in drug trafficking cases and over 3 years more than New Englanders in street crime cases). Other regions ranged between the two extremes. Id. For a description of geographic regions in table 8.17, seeid. at 175 tbl.8.1.

Respondents aged 50-64 sentenced on average 2.1 years more for drug trafficking offenses and 0.6 years more for drug possession offenses than respondents aged less than 35. Id. at 200-01 tbl.8.17. Respondents aged 35-49 on average sentenced 0.91 years more in street crimes cases than respondents aged less than 35. Id.

Those who believed police are given too much investigative freedom sentenced less severely than those who believed police investigative freedom to be appropriate or insufficient. Id. (on average 1.67 years less for drug trafficking, 0.31 years less for drug possession, and 0.76 years less for street crimes).

Those who believed an accused should have fewer rights sentenced more severely than those who believed an accused should have more rights. Id. (on average 1.64 years more for drug trafficking offenses and 1.09 years more for street crimes; views on defendants’ rights were not statistically significant for drug possession sentencing).

Those who believed minorities should be accorded fewer civil rights sentenced more severely in drug trafficking crimes than those who believed minorities should be accorded more civil rights. Id. (on average 1.15 years more in drug trafficking cases; civil rights views were not statistically significant for drug possession or street crimes sentencing).

Those who had more contacts with the justice system, including the civil justice system, on average sentenced drug trafficking offenses and street crimes more severely than those who had fewer justice system contacts. Id. (respectively, 0.98 years and 0.59 years more; justice system contacts were not a statistically significant factor for drug possession sentencing). The study counted as “contact” any of the following interactions: jury service, reporting crimes to the police, being sued in court, testifying in court as a witness, being arrested, or serving time in jail or prison. Id. at 192 tbl.8.11. The study did not differentiate among types of contact (i.e., whether the person had been a defendant, had simply called the police to report a crime, or had some other type of contact with the justice system). Id. at 192.

Those who self-identified as “liberal” in social and political views sentenced less severely for drug trafficking, drug possession, and street crimes than those who did not so identify. Id. at 200-01 tbl.8.12 (0.74 years, 0.12 years, and 0.26 years less, respectively).

Those who were more educated sentenced less severely for drug trafficking, drug possession, and street crimes as compared to those who were less educated. Id. (0.26 years, 0.10 years and 0.20 years less, respectively).

As the authors noted, their measure of community size was “not very sensitive to the considerable differences in the actual living conditions that respondents face within each size class,” an unfortunate limitation of their dataset. Id. at 178.

This will be true at the district court level, but it may not be true at the federal appellate level, at least in circuits that encompass significant areas in more than one geographic region. These include the Second, Sixth, and Ninth Circuits.

See Lindsay M. Howden & Julie A. Meyer, 2010 Census Briefs: Age and Sex Composition: 2010, U.S. Census Bureau 11 fig.7 (May 2011), http://www.census.gov/prod/cen2010/briefs /c2010br-03.pdf.

Using the results of the latest presidential election as a proxy for political views, with few exceptions urban counties strongly favored the Democratic candidate (Barack Obama) over the Republican candidate (Mitt Romney), even in states carried overwhelmingly by Romney. See Election 2012: President Map, N.Y. Times, http://elections.nytimes.com/2012 /results/president (last updated Nov. 29, 2012).

See supra notes 162-166 and accompanying text.

See supra text accompanying notes 222-223.

Paul Robinson and John Darley call this phenomenon the “generalization of disrespect”: the more laws citizens perceive as immoral, the more apt they are to discredit the entire system as opposed to the select laws with which they disagree. More generalized systemic aversion is also more likely to occur where unjust results are not readily attributable to a particular legal rule. Robinson & Darley, supra note 117, at 483-85.

See supra notes 128-129 and accompanying text.

See supra notes 196-203, 209, and accompanying text. In this sense, it is worth noting that even with the greater number of below-range sentences in child pornography cases, the median sentence in those cases has risen along with the Guidelines range, albeit at a lower rate than the Guidelines’ increase. And even post-Booker, the median sentence is still far above the statutory mandatory minimum. U.S. Sent’g Comm’n, supra note 258, at 135.

The Post Office Act of 1872 and, subsequently, the Offenses Against the Postal Service Act of 1909 were the precursors of the modern-day federal mail and wire fraud statutes, 18 U.S.C. §§ 1341, 1343 (2012), but despite these acts’ breadth, it was not until the New Deal era that federal prosecution of large-scale frauds became widespread. See Richman, supra note 1, at 83-87. The federal securities fraud statutes were first passed in 1933 and 1934. Securities Act of 1933, ch. 38, Pub. L. No. 73-22, 48 Stat. 74, (codified as amended at 15 U.S.C. §§ 77a-77aa (2012)); Securities Exchange Act of 1934, ch. 404, Pub. L. No 73-291, 48 Stat. 881 (codified as amended at 15 U.S.C. §§ 78a-78pp). The federal RICO statute, the primary tool used to combat organized crime, was first passed in 1970. Organized Crime Control Act of 1970 § 901(a), Pub. L. No. 91-452, 84 Stat. 941 (codified as amended at 18 U.S.C. §§ 1961-1968).

Congress criminalized fraud upon health care benefits in 1996, Health Insurance Portability and Accountability Act § 242(a)(1), Pub. L. No. 104-191, 110 Stat. 2016 (1996) (codified as amended at 18 U.S.C. § 1347); fraud in connection with food stamps in 1964, The Food Stamp Act of 1964, Pub. L. No. 88-525, 78 Stat. 708 (codified at 7 U.S.C. § 2024); and fraud in connection with an entity receiving federal housing funds in 1948, Crimes and Criminal Procedure § 1012, Pub. L. No. 80-772, 62 Stat. 752 (1948) (codified at 18 U.S.C. § 1012).

For a cogent argument as to why class-based enforcement undermined both Prohibition and the federal crack laws, seeStuntz, supra note 242. See generally Daniel Okrent, Last Call: The Rise and Fall of Prohibition (2011).

This is not to say the federal government would have better luck enforcing the Volstead Act today. Just as the federal government holds an entirely different place in modern American life, so too does liquor. For a richly detailed history of alcohol, alcoholism, and public drunkenness at the dawn of the twentieth century, and its impact on Prohibition’s rise, see Okrent, supra note 292.

See Matthew C. Waxman, National Security Federalism in the Age of Terror, 64 Stan. L. Rev. 289 (2012).

Anti-terrorism presents a particularly tangled set of reputational consequences for federal and local law enforcement. Some federal anti-terrorism efforts, most notably the enforcement of immigration laws among select minority communities, have detracted from federal law enforcement’s standing in those communities. As a result, local police forces, which depend on the support and cooperation of minority communities to fight street crime, have in many instances declined to cooperate with federal immigration enforcement efforts. See Daniel C. Richman, The Right Fight, Bos. Rev., Dec. 1, 2004, http://www .bostonreview.net/forum/right-fight(describing this dynamic). In other circumstances, joint anti-terrorism enforcement efforts by federal, state and local law enforcement authorities has burnished the reputational standing of each; a recent example of this phenomenon is the successful search for the Boston Marathon bombing suspects. See Katharine Q. Seele, William K. Rashbaum & Michael Cooper, 2nd Bombing Suspect Caught After Frenzied Hunt Paralyzes Boston, N.Y. Times, Apr. 19, 2013, http://www.nytimes.com /2013/04/20/us/boston-marathon-bombings.html.

See supra notes 72-73 and accompanying text.

A number of states have enacted laws similar to 18 U.S.C. § 924(c), the federal prohibition that carries a minimum five-year sentence of imprisonment for possession of a firearm in connection with a drug trafficking or violent crime. (The five-year minimum penalty was enacted in 1971 pursuant to Public Law Number 91-644.) Such states include California, see Cal. Penal Code §§ 12022, 12022.53 (West 2010) (3, 4, or 10 year minimum, depending on coordinate offense); Florida, see Fla. Stat. Ann. § 775.087(2)(a) (West 1999) (minimum of 10 or 20 years, depending on coordinate offense); New York, see N.Y. Penal Law § 265.09(2) (McKinney 1980) (five year minimum); and Pennsylvania, see 42 Pa. Cons. Stat. Ann. § 9712.1(a) (West 1994) (five year minimum).

See sources cited supra notes 10-11.

Some may argue that the federal system’s circumvention of localism makes it illegitimate. My use of the term “legitimacy” is not in the normative sense, but rather in the descriptive theoretical sense: federal prosecutors achieve success relative to their local counterparts because interactions of authority, community, and personal morality play out to prosecutors’ advantage in the federal system.

See Rachel A. Harmon, The Problem of Policing, 110 Mich. L. Rev. 761 (2012); Kami Chavis Simmons, Cooperative Federalism and Police Reform: Using Congressional Spending Power to Promote Police Accountability, 62 Ala. L. Rev. 351, 383-85 (2011).

See sources citedsupra note 91; see also Protecting America from Terrorist Attack, Fed. Bureau Investigation, http://www.fbi.gov/about-us/investigate/terrorism/terrorism_jttfs (last visited Jan. 28, 2014) (describing the Joint Terrorism Task Force); Violent Gang Task Forces, Fed. Bureau Investigation, http://www.fbi.gov/about-us/investigate/vc_majorthefts /gangs/violent-gangs-task-forces (last visited Jan. 28, 2014) (describing violent gang task forces); Human Trafficking – FBI Initiatives, Fed. Bureau Investigation, http://www .fbi.gov/about-us/investigate/civilrights/human_trafficking/initiatives (last visited Jan. 28, 2014) (describing human trafficking task forces).

See 2012 Annual Report, Off. of the Special Narcotics Prosecutor of the City of New York, http://www.nyc.gov/html/snp/downloads/pdf/ar_2012.pdf.

For discussion of the risks of increased federal policing, see William J. Stuntz, Terrorism, Federalism, and Police Misconduct, 25 Harv. J.L. & Pub. Pol’y, 665, 670-74 (2002).

J. Edgar Hoover’s FBI is a historical example. See generally Athan G. Theoharis & John Stuart Cox, The Boss: J. Edgar Hoover and the Great American Inquisition (2012).

See Akhil Reed Amar, Of Sovereignty and Federalism, 96 Yale L.J. 1425, 1495 (1987) (“[I]n separating and dividing power, whether horizontally or vertically, the Federalists pursued the same strategy: Vest power in different sets of agents who will have personal incentives to monitor and enforce limitations on each other’s powers.”).

See Harmon, supra note 300 (discussing the inability of constitutional criminal procedure to constrain unlawful policing practices and arguing in favor of new legal paradigms for regulating police conduct); Debra Livingston, The Unfulfilled Promise of Citizen Review, 1 Ohio St. J. Crim. L. 653, 661-69 (2004) (discussing the importance of remedying minor police misconduct and the challenges of doing so through standard rule enforcement methods); William J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 Yale L.J. 1 (1997) (discussing how the Warren Court’s constitutional criminal procedure doctrines have incentivized many of the police abuses we see today).

State Witness Protection Act of 2012, S. 2127, 112th Cong. (2012).

The proposed bill, for example, would have reached only witness tampering with a specified connection to interstate commerce. See id. § (c). Practically speaking, this means it could be used only in cases involving threats made over the phone or Internet, through the mail, or by persons who, in the course or in furtherance of the witness tampering offense, crossed state lines or used a firearm transported over state lines.

For an overview of the membership criteria for each state’s sentencing commission, see Neil B. Kauder & Brian J. Ostrom, State Sentencing Guidelines: Profiles and Continuum, Nat’l Center for St. Cts. (July 2008), http://www.pewtrusts.org/up loadedFiles/wwwpewtrustsorg/Reports/sentencing_and_corrections/NCSC%20Sentencing %20Guidelines%20profiles%20July%202008.pdf.

See 2012 Annual Report, DC Sentencing and Criminal Code Revision Commission, at i, iii (Apr. 26, 2013), http://acs.dc.gov/acs/frames.asp?doc=/acs/lib/acs/annual_report_2012.pdf (listing membership and noting continual modifications to the guidelines based on assessments of judicial compliance with the guidelines, to insure that the guidelines’ sentencing recommendations are appropriate).

Id. at iii, 53.

Dan M. Kahan & Tracey L. Meares, Foreword: The Coming Crisis of Criminal Procedure, 86 Geo. L.J. 1153, 1153-54 (1998); Livingston, supra note 195, at 554-56.

E.g., City of Chicago v. Morales, 527 U.S. 41 (1999); see Kim Strosnider, Anti-Gang Ordinances After City of Chicago v. Morales: The Intersection of Race, Vagueness Doctrine, and Equal Protection in the Criminal Law, 39 Am. Crim. L. Rev. 101 (2002).

E.g., McDonald v. City of Chicago, 130 S. Ct. 3020 (2010).

John Byrne & Hal Dardick, City Council Passes Daley Gun Restrictions 45-0, Chi. Trib.: Clout Street (July 2, 2010, 11:26 AM), http://newsblogs.chicagotribune.com/clout_st /2010/07/city-council-passes-daley-gun-restrictions-450.html. A number of lawsuits challenging Chicago’s gun ordinances are winding their way through the courts. For a list of cases, see Gun Rights Lawsuits: Court Status of 9 Challenges to City, State and County Rules, Chi. Trib., Mar. 8, 2012, http://articles.chicagotribune.com/2012-03-08/news/ct-met-gun -lawsuit-box -20120308_1_second-amendment-arms-firearms-retailers-gun-rights.

See 18 Pa. Cons. Stat. Ann. § 6108 (West 2000). In Pennsylvania, unlike in Illinois, state firearms laws preempt conflicting local laws. As a result, city-specific firearms offenses must be enacted by the state legislature.

In Pennsylvania, for instance, carrying a concealed weapon in Philadelphia is a misdemeanor, see id. §§ 6108, 6119, and carrying a firearm at all in any public park, playground or recreation center in Philadelphia constitutes a violation that results in a maximum $2,000 fine. See Phila., Pa Code § 16-306 (2011) (criminalizing carrying a firearm in government-owned buildings). In Chicago, the penalties for gun ordinance violations are similarly lenient. See Monica Davey, Strict Gun Laws in Chicago Can’t Stem Fatal Shots, N.Y. Times, Jan. 29, 2013, http://www.nytimes.com/2013/01/30/us/strict -chicago-gun-laws-cant-stem-fatal-shots.html (discussing how lax penalties result in anemic enforcement of Chicago’s gun ordinances).

For instance, currently under consideration by the Pennsylvania legislature—a result of sustained lobbying by Philadelphia’s mayor, police commissioner, and city council—is a bill that would mandate a two-year prison term for anyone carrying a gun illegally in Philadelphia. See Allison Steele & Amy Worden, Mandatory 2-Year Prison Term Proposed for Illegal Firearm Users in Phila., Phila. Inquirer, Apr. 6, 2013, http://articles.philly.com/2013 -04-06/news/38309721_1_gun-purchase-gun-violence-gun-control. The genesis of this proposal lies more in the complicated politics of gun regulation than in ideals of local empowerment. Nevertheless, the concept is worth serious consideration.

See supra notes 262-264 and accompanying text.

See id.

See Stuntz, supra note 3, at 2034-36 (advocating for more narrowly-drawn urban jury districts separated from suburban communities); cf. Heather K. Gerken, The Supreme Court 2009 Term—Foreword: Federalism All the Way Down, 124 Harv. L. Rev. 4, 30-33 (2010) (more broadly advancing a theory of localism as a means of dissent from sovereign power, using the local jury—either criminal or civil—as an example).

See supra notes 196-203, 211-214, and accompanying text.

See Clyde Haberman, Fix a Ticket, Alienate a Bronx Jury, N.Y. Times: City Room (June 6, 2011, 8:35 AM), http://cityroom.blogs.nytimes.com/2011/06/06/fix-a-ticket-alienate-a -bronx-jury (“[T]he Bronx has long been rough going for [prosecutors], especially when police testimony is pivotal.”); Chris Herring, Bronx Acquittals Set Record, Wall St. J., May 4, 2010, http://online.wsj.com/news/articles/SB10001424052748704608104575221271762806324 (reporting a jury conviction rate of only 43% in 2009, which many attribute to “tensions between the police and the community”); cf. William Glaberson, Faltering Courts, Mired in Delays, N.Y. Times, Apr. 13, 2013, http://www.nytimes.com/2013/04/14/nyregion/justice -denied-bronx-court-system-mired-in-delays.html (reporting a jury conviction rate in Bronx criminal court that has declined from 67% in 1989 to 46% in 2011, and discussing the deleterious effects of trial delays on conviction rates).

The case, Floyd v. City of New York, No. 08 Civ. 1034, 2013 WL 4046209 (S.D.N.Y. Aug. 12, 2013), raised public awareness of the New York City Police Department’s stop-and-frisk practices’ impact on minorities, catapulting the issue onto the political stage. The case precipitated legislation by the New York City Council, reforms by the NYPD, and a landslide victory by a mayoral candidate whose promises to end the practices became a centerpiece of his campaign. See Michael Barbaro & David W. Chen, De Blasio Is ElectedNew York City Mayor in Landslide, N.Y. Times, Nov. 5, 2013, http:// http://www.nytimes.com/2013/11/06/nyregion/de-blasio-is-elected-new-york-city-mayor.html; Jim Dwyer, Vowing to Slay the (Already Subdued) Stop-and-Frisk Dragon, N.Y. Times, Dec. 5, 2013, http://www.nytimes.com/2013/12/06/nyregion/de-blasio-and-bratton-promise-to -deliver-on-goals-but-give-no-credit.html; J. David Goodman, City Council Votes to Increase Oversight of New York Police, N.Y. Times, June 27, 2013, http://www.nytimes.com /2013/06/27/nyregion/new-york-city-council-votes-to-increase-oversight-of-police-dept.html (reporting passage of the Community Safety Act, which would, among other things, create an independent inspector general to monitor NYPD policies and practices and make it easier for plaintiffs to sue the NYPD for bias-based profiling); Wendy Ruderman, Number of Frisks Fell in ‘12, Police Data Show, N.Y. Times, Feb. 8, 2013, http://www.nytimes.com /2013/02/09/nyregion/number-of-frisks-fell-in-12-police-data-show.html (reporting that “[a]fter criticism from a federal judge last spring and widening political protests,” new police officer stop-and-frisk training was instituted, resulting in a 34% decrease in stops in the last three quarters of 2012 compared to the same period in 2011).

Although the district judge’s findings and remedies were initially stayed pending the City’s appeal, Mayor De Blasio has reversed the New York City Law Department’s course in the case, vowing to reform the NYPD’s stop-and-frisk practices largely in accordance with the district court’s ruling, to settle the case, and to dismiss the City’s appeal. See Benjamin Weiser & Joseph Goldstein, Mayor Says New York City Will Settle Suits on Stop and Frisk Practices, N.Y. Times, Jan. 30, 2014, http://www.nytimes.com/2014/01/31/nyregion/de -blasio-stop-and-frisk.html. In that vein, it is no accident that De Blasio picked Zachary Carter to lead the City’s legal department; as the United States Attorney in Brooklyn in the 1990s, Carter had overseen the federal civil rights prosecution in the Abner Louima matter. See supra notes 169-173 and accompanying text; Annie Correal, De Blasio Names City’s Top Lawyer, Appearing to Signal a Further Shift in Policy, N.Y. Times, Dec. 29, 2013, http:// http://www.nytimes.com/2013/12/30/nyregion/de-blasio-announces-pick-for-new-york-citys-top -lawyer.html.

Jeremy W. Peters, Albany Reaches Deal to Repeal ’70s Drug Laws, N.Y. Times, Mar. 25, 2009, http://www.nytimes.com/2009/03/26/nyregion/26rockefeller.html (describing how budgetary pressure to reduce imprisonment rates and shifting political winds combined to permit legislation to reduce drug penalties).

Consent Decree, Bailey v. City of Philadelphia (E.D. Pa. June 21, 2011) (No. 10-5952), http:// http://www.aclupa.org/download_file/view_inline/744/198; see also Erica Goode, Philadelphia Defends Policy on Frisking, with Limits, N.Y. Times, July 11, 2012, http://www.nytimes.com /2012/07/12/us/stop-and-frisk-controls-praised-in-philadelphia.html (describing the class action lawsuit’s role in altering the Philadelphia Police Department’s approach to citizen stops); Phillips, McCoy & Purcell, supra note 158 (describing city’s low conviction rates).

Press Release, U.S. Dep’t of Justice, Justice Department Announces Consent Decree with City of New Orleans to Resolve Allegations of Unlawful Misconduct by New Orleans Police Department (July 24, 2012), http://www.justice.gov/opa/pr/2012/July/12-ag-917.html (detailing DOJ’s suit against NOPD under 42 U.S.C. § 14141). Jury conviction rates in New Orleans are notoriously low, a circumstance that many observers attribute to jurors’ mistrust of police testimony.John Simerman, Orleans Parish Conviction Rates in Jury Trials Hold Steady, and Relatively Low, Times-Picayune, Jan. 16, 2012, http://www.nola.com/crime /index.ssf/2012/01/orleans_parish_conviction_rate.html (reporting a conviction rate in tried cases that has “perennially hovered around 55 percent to 60 percent,” which many observers have attributed to jurors’ mistrust of police testimony).

Brown v. Plata, 131 S. Ct. 1910 (2011). See generally Barkow, supra note 10, at 1285 (arguing that cost constraints have a salutary effect on states’ sentencing policies).

See Adrienne Austin, Criminal Justice Trends: Key Legislative Changes in Sentencing Policy, 2001-2010, Vera Inst. Just. 4 (Sept. 2010), http://www.courts.ca.gov/documents/ccp -sentencingtrends.pdf (attributing states’ recent sentencing reforms to a combination of lower crime rates, fiscal pressures, concerns about over-burgeoning prison populations, and new, evidence-based research on behavioral modification).

See Jack McDevitt et al., Offender Notification Meetings: Case Study 2,in Project Safe Neighborhoods: Strategic Interventions, U.S. Dep’t Just., https://www.bja.gov /Publications/Offender_Notification_Meetings.pdf. In some offender meetings, reformed ex-gang members are brought in to speak to offenders, as well. Id.

See Andrew V. Papachristos, Tracey L. Meares & Jeffrey Fagan, Attention Felons: Evaluating Project Safe Neighborhoods in Chicago, 4 J. Empirical Legal Stud. 223 (2007) (presenting empirical evidence demonstrating that homicide levels have decreased in Chicago neighborhoods that have instituted PSN relative to those that have not).

One of the brilliant successes of the offender notification meetings tactic is that it seeks to pre-empt criminal conduct, thus reducing the need for prosecution at all. My proposals, in contrast, are aimed at making local criminal prosecution more effective in instances where it is needed. But as local criminal prosecution becomes more effective (in the sense that outcomes are more certain and transparent, and align more closely with written law), it, too, might be used as a tool for crime-reduction in the way that PSN uses the threat of federal prosecution.

See Alex Wigglesworth, CeaseFirePA: Advocates Take to Courtrooms to Tip Scales of Justice Against Gun Felons, Phila. Metro, Sept. 30, 2012, http://www.metro.us/philadelphia/news /local/2012/09/30/ceasefirepa-advocates-take-to-courtrooms-to-tip-scales-of-justice-against -gun-felons.

Id.; see also Courtwatch, CeaseFirePA, http://www.ceasefirepa.dreamhosters.com/programs /court-watch (last visited Feb. 1, 2013) (describing the Courtwatch program and listing recent Courtwatch reports).

SeeWigglesworth, supra note 333. Among the most notable achievements was the resentencing of Kevin Pickard, a defendant who, in the process of shooting his intended 19-year-old target, also shot two brothers, aged 2 and 8, who were playing outside in the street. The brothers survived after extended hospitalization. At Pickard’s initial sentencing, the local judge assigned to his case sentenced him to three concurrent terms of five to ten years’ imprisonment. Craig R. McCoy, Minimum Sentence for Gunman in Shooting of 2 Children, Phila. Inquirer, Aug. 11, 2012, http://articles.philly.com/2012-08-11/news /33138250_1_kevin-pickard-three-concurrent-sentences-minimum-sentence. The District Attorney’s Office moved for reconsideration and resentencing, which the judge granted. At the resentencing hearing one week later, the courtroom was filled with community members expressing outrage over the initial sentence. At his resentencing, Pickard was sentenced to 17 to 34 years’ imprisonment. See Mensah M. Dean, Community Anger Leads to Stricter Sentence for Man Who Shot Kids, Phila. Daily News, Aug. 19, 2012, http://articles.philly.com/2012 -08-19/news/33273288_1_community-work-shot-kids-court-sessions.

Cf. Dan M. Kahan, Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms Problem, 67 U. Chi. L. Rev. 607 (2000) (discussing the tension between criminal law and social norms in the context of date rape, drunk driving, and domestic violence, and arguing that the law can better shift social norms through a more graduated approach).

Omnibus Public Safety Amendment Act, 2006 D.C. Legis. Serv. 16-306 (Act 16-482) (West) (codified at D.C. Code § 22-4503 (2013)).

According to data from the Federal Judicial Center, in the five years preceding the 2006 law, federal prosecutors in the District of Columbia prosecuted 353 cases in which 18 U.S.C. § 922(g) (2012) was the most severe count, an average of approximately 71 cases per year; in the five years following the law, federal prosecutors prosecuted only 53 such cases, an average of just 10 per year. (Datasets for FJC and USSC data are available at http://www .icpsr.umich.edu/icpsrweb/ICPSR/series/00072 and http://www.icpsr.umich.edu/icpsrweb /ICPSR/series/83 subject to restricted use agreement; data analysis is on file with author.)

See, e.g., Rachel E. Barkow, Recharging the Jury: The Criminal Jury’s Constitutional Role in an Era of Mandatory Sentencing, 152 U. Pa. L. Rev. 33, 85-86 (2003); Rachel E. Barkow, Sentencing Guidelines at the Crossroads of Politics and Expertise, 160 U. Pa. L. Rev. 1599, 1620-21 (2012) [hereinafter Barkow, Sentencing Guidelines].

See Barkow, Sentencing Guidelines, supra note 339, at 1620 & n.103; Kate Stith, The Arc of the Pendulum: Judges, Prosecutors, and the Exercise of Discretion, 117 Yale L.J. 1420, 1430 (2008) (discussing the transfer of discretionary power from judges to prosecutors under a mandatory guidelines regime, which is effectively similar in this respect to legislatively-mandated penalties).

See Rachel E. Barkow, Prosecutorial Administration: Prosecutor Bias and the Department of Justice, 99 Va. L. Rev. 271, 272 (2013) (noting that over the last decade, the federal prison population has increased 400%, a rate nearly three times that of the states, and that federal prisons currently house more inmates than the prisons of any single state).