Skip navigation

The Role of Judicial Political Affiliation in Criminal Sentencing Outcomes, 2020

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Draft – Do not cite without authors’ permission
Wendy R. Calaway,1 Jennifer M. Kinsley,2
and Taylor Wadian3
Legislative efforts to bring consistency to criminal sentencing outcomes has been much
discussed in academic literature and Congressional hearings alike. Despite these efforts disparate
sentencing outcomes persist. Researchers have studied many variables seeking to understand
these disparities but have been unable to form a consensus around the cause. Perhaps because of
the lack of a firm understanding of the issue among researchers, legislative intervention at both
the state and federal level has largely failed to address the issue of judicial characteristics that
may drive sentencing disparities. As a result, absent from the conversation on criminal sentencing
reform is empirical and anecdotal evidence about how judges make determinations within the
range of outcomes specified by the legislature. New data on federal sentencing outcomes collected
by Harvard researchers, however, finds a direct connection between the political party of the
President who appointed the federal judge and the length of a defendant’s sentence. As the
Harvard study reports, federal judges appointed by Republican presidents sentence defendants on
average to three more months in prison than federal judges appointed by Democratic presidents.
Republican-appointed judges in the federal system also sentence black defendants more harshly
than Democratic-appointed judges.
As will be discussed in this Article, the central premise of the Harvard political sentencing
study – that judicial political affiliation influences sentencing outcomes, even those that are highly
guided by legislative criteria – also holds true on the state level with respect to elected, rather
than appointed, judges. As we report, empirical evidence from the state of Ohio demonstrates that
elected Republican judges sentence defendants to lengthier terms of incarceration than elected
Democratic judges by a statistically significant margin. This evidence suggests that, rather than
being entirely guided by specified statutory criteria, judges bring preexisting sentencing ideologies
to the bench and make decisions with a range of sentencing outcomes based at least in part on
their individual philosophies and beliefs. Based on these findings, we argue that in order to
address the issue of sentencing disparities, reform efforts should take action to specifically address
the behavior and motivation of individual judges.

Associate Professor of Criminal Justice, University of Cincinnati Blue Ash, J.D., University of
Cincinnati College of Law.
Professor of Law, Northern Kentucky University Salmon P. Chase College of Law, J.D., Duke
University Law School. The author wishes to thank Emily Robbins and Raegan Freeman for their
excellent research assistance and extensive efforts in collecting and organizing data for this paper,
as well as Jacob Erdman for his assistance in researching peremptory strikes.
Assistant Professor of Psychology, University of Cincinnati Blue Ash, PhD, Kansas State

Electronic copy available at:

Like no other aspect of the American legal system, criminal sentencing outcomes lie at the
intersection of the executive, legislative, and judicial branches.4 To be sure, no criminal defendant
is sentenced without the involvement of all three branches of government. At the outset, the
executive branch plays a both gatekeeping role, by determining which individual defendants and
which legal matters are brought before the courts, and an advisory role, by making sentencing
recommendations to the court in individual cases. In both regards, the executive branch retains a
high degree of discretion and wide-ranging powers.5 The legislative branch also plays an advisory
role, by enacting statutes that establish sentencing outcomes for particular crimes and people and
guide judicial decision-making in the realm of criminal sentencing.6 Lastly, the judicial branch
directly imposes criminal sentences in the cases brought before it by the executive branch and
within the constraints imposed upon it by the legislative branch.7


Julia A. Black, The Constitutionality of Federal Sentences Imposed under the Sentencing Reform
Act of 1984 after Mistretta v. United States, 75 Iowa L. Rev. 767, 778-85 (March 1990) (discussing
“criminal sentencing historically has been considered within the scope of all three branches”).
Much has been said about the role of prosecutors in driving sentencing outcomes. Because
prosecutors in many jurisdictions choose the charges to render in a particular case, and the level
of those charges within tiered felony and misdemeanor statutes, in many respects a defendant’s
sentence will be driven by the charges the prosecutor elects to pursue. Prosecutors also exercise
broad discretion in the plea bargaining phase, where they may elect to dismiss or reduce certain
charges to drive a particular sentencing outcome. See, e.g., Shima Baradaran Baughan,
Subconstitutional Checks, 92 Notre Dame L. Rev. 1071 (January 2017) (discussing the heightened
role of prosecutorial discretion in plea bargaining and proposing additional checks and balances
by the other branches of government). In some instances, prosecutors even dictate a specific
sentence or sentencing range as part of a defendant’s plea agreement, and they may also use a
defendant’s perceived degree of cooperation or information-sharing as a factor in reaching a given
sentencing outcome. See, e.g., U.S.S.G. § 5K1.1 (permitting federal prosecutors to recommend
downward sentencing departure where defendant provides substantial assistance to the
See, e.g., Ohio Senate Bill 2.
See, e.g., Ohio Rev. Code § 2929.12 et seq.

Electronic copy available at:

Despite the triumvirate nature of criminal sentencing, the conversation around sentencing
reform – particularly as it relates to documented disparate outcomes across populations and regions
- has focused almost exclusively on legislative action. 8 To this end, the statutory schemes
underlying sentencing have migrated from indeterminate ranges loosely specified by the
legislature to determinate sentencing schemes in which judicial discretion is highly constrained –
all with the stated goal of ensuring that criminal sentences are proportionate, fair, and adequately
reflect the defendant’s conduct and history.9 As a result of prolonged and sustained legislative
reform, modern sentencing statutes typically impose a range of sentencing outcomes and
alternatives from which the sentencing judge choose and then offer a detailed list of factors or
criteria to guide the judge’s selection within that range.10
At times absent from the conversation on criminal sentencing reform, however, is empirical
and anecdotal evidence about how judges make determinations within the range of outcomes
specified by the legislature. New data on federal sentencing outcomes collected by Harvard
economist Alma Cohen and Harvard Law Professor Crystal Yang and reported in the American
Economic Journal finds a direct connection between the political party of the President who
appointed the federal judge and the length of a defendant’s sentence.1112 As the Harvard political
sentencing study reports, federal judges appointed by Republican presidents sentence defendants
on average to three more months in prison than federal judges appointed by Democratic


Black, supra note ___, at 785-7 (discussing federal legislative sentencing reform).
See, e.g., Steven L. Chanenson, The Next Era of Sentencing Reform, 54 Emory L. J. 377, 382-86
(Winter 2005) (discussing relative attributes of determinate and indeterminate sentencing
schemes); 11 Tenn. Prac. Crim. Prac. & Procedure § 32.2 (discussing history of sentencing reform
in Tennessee and the progression from indeterminate to determinate sentencing practices).
See, e.g., Ohio Rev. Code § 2929.12 et seq.
Cohen and Yang, “Judicial Politics and Sentencing Decisions,” American Economic Journal:
Economic Policy 2019, 11(1): 160–191.
This study will be referenced throughout the Article as “the Harvard political sentencing study.”

Electronic copy available at:

presidents.13 Republican-appointed judges in the federal system also sentence black defendants
more harshly than Democratic-appointed judges.14
As will be discussed in this Article, the central premise of the Harvard political sentencing
study – that judicial political affiliation influences sentencing outcomes, even those that are highly
guided by legislative criteria – also holds true on the state level with respect to elected, rather than
appointed, judges.15 As we report, empirical evidence from the state of Ohio demonstrates that
elected Republican judges sentence defendants to lengthier terms of incarceration than elected
Democratic judges by a statistically significant margin.16 This evidence suggests that, rather than
being entirely guided by specified statutory criteria, judges bring preexisting sentencing ideologies
to the bench and make decisions with a range of sentencing outcomes based at least in part on their
individual philosophies and beliefs.17
Drawing from this new data, this Article will discuss the role of political affiliation in
judicial sentencing. Part One of the paper will discuss the various models by which judges are
selected, including appointment, retention, and election, and the role of political affiliation in each
model. Part One will also summarize legislative sentencing reforms over the past four decades
and how the role of the judiciary and its discretion in criminal sentencing have shifted over time.
Part Two of the paper will summarize the emerging schools of thought around sentencing reform
and the existing data on the role politics plays in sentencing outcomes, including the Harvard
political sentencing study. Part Three will describe the results of the current Ohio empirical
sentencing study, including a discussion of the study methodology and conclusions that can be


See Section ___, infra.

Electronic copy available at:

drawn from the available data around the role of judicial political affiliation in criminal sentencing
outcomes. Lastly, the Article concludes in Part Four that legislative reforms are an insufficient
mechanism for obtaining proportional sentencing outcomes, given that judges sentence within a
range of alternatives at least in part based upon preexisting political ideology. In Part Four, the
Article proposes additional remedies to achieve sentencing fairness, including improved data
collection and public education around individual judicial sentencing patterns and the use of
judicial peremptory strikes by prosecutors and defendants. As this Article posits, purely legislative
attempts to reform criminal sentencing will be ineffective absent recognition of the role judicial
political affiliation plays in driving sentencing outcomes.
A. Judicial Selection Models
There is no consistent method by which individuals come to serve as judges, although there
do exist a number of common models for judicial selection, most notably 1) appointment, where a
particular branch or branches of government have the sole authority to select and seat judges; 2)
retention, where an initial judicial appointment is made by the government subject to a later
retention election by the people, and 3) popular elections, which can either be partisan or nonpartisan. Even within a single jurisdiction, multiple models may be used to select judges at
different levels of the judiciary.18 New York, for example, selects its appellate court judges by
appointment, but selects its trial court judges by election.19 The role of judicial partisan affiliation


See, e.g., “Methods of Judicial Selection: New York,” National Center for State Courts,
judges.cfm?state=NY (last viewed July 16, 2020).

Electronic copy available at:

varies within each model, although all methods of selection at least implicitly if not explicitly
account for a judge’s chosen political party.
1. Appointment
One model for the selection of judges vests one or both of the non-judicial branches of
government with judicial appointment authority. The most recognizable version of this model lies
with the federal judiciary, where federal judges are nominated by the President of the United States
and confirmed by the Senate.20 Once appointed, federal judges serve for life.21
A number of states also employ an appointment model for the selection of judges. In these
states, appointment of judges is made by the executive branch after confirmation by the legislative
branch.22 On the state level, gubernatorial appointment systems are also used to fill short-term
judicial vacancies, such as when an elected judge resigns or otherwise leaves the bench before her
term has expired.23
Politics plays a role in both scenarios. As with the federal system, judges are selected by
bureaucrats with existing political party affiliations and tend to appoint judges who are aligned
with their particular political party.24 In addition, in some appointment systems, governors must


U.S. Const. Art. II, § 2.
U.S. Const. Art. III, § 1.
In Maine, for example, district court judges are appointed to seven-year terms by the Governor
and must be confirmed by the State Senate. Maine Const. Art. V, §8; Art. VI, §4. New Jersey
follows a virtually identical appointment process.
See, e.g., Ohio Const. Art. IV, § 13 (requiring judicial vacancies to be filled by gubernatorial
See, e.g., Lee Epstein and Eric A. Posner, Supreme Court Justices’ Loyalty to the President, 45
J. Legal Stud. 401 (June 2016) (studying the loyalty effect between appointed Supreme Court
justices and the President who appointed them and concluding that Democratic justices exhibit
stronger loyalty to Presidents who appointed them compared to future Democratic administrations
than do Republican justices). Interestingly, Epstein and Cohen attribute the results of their study
in part due to their hypothesis that Republican justices remain more ideologically committed over
time than Democratic justices, leaving less room for Republicans to demonstrate loyalty to the
particular President who appointed them. Id. at 428-9.

Electronic copy available at:

select appointed judges solely from a list a candidates provided by a nominating committee, which
may itself be subject to partisan influence. Increasingly, legislative confirmation of judicial
nominees is highly politicized, resulting in votes that are almost perfectly divided along party lines.
The legislative confirmation process often typically involves public hearings, at which a nominee’s
political, social, moral, and ethical views are debated and vetted.25 The appointment process is
therefore largely political, and a judicial candidate’s political affiliations are likely to be a strong
consideration in her appointment.26
2. Retention
In judicial retention schemes, an initial judicial appointment is made by the government,
and the decision whether to retain or remove a judge following an initial term of appointment is
made by the electorate. Retention elections are uncontested, meaning that voters cast either a yes
or no ballot to retain the judge in her appointed position.27 Initial appointments are typically made
after vetting by a nominating commission, which is often itself appointed by the governor.28
While in theory retention elections are non-partisan and therefore less subject to political
influence than other judicial selection methods, recent high-profile campaigns to unseat appointed
judges in retention elections cast doubt on the apolitical nature of these regimes. In Florida, for


See, e.g., Sabrina Hersi Issa, “Brett Kavanaugh’s Confirmation Reveals the Moral Rot that
Allows America to Ignore so Many Assault Survivors,” NBC News (Oct. 6, 2018), available at (last viewed July 19, 2020).
See Epstein and Posner, supra note __, at 427-30.
See, e.g., Mont. Code Ann. § 13-14-212(2) (specifying form of ballot in judicial retention
election as a “yes” or “no” vote).
Iowa, for example, employs a State Judicial Nominating Commission, staffed with nine
members serving six-year terms. Iowa Stat. § 46.1(1), (2). Commissioners are appointed by the
governor and confirmed by the state senate. Id. at § 46.1(1). Other than a disclaimer that
commissioners are appointed without regard to political affiliation, there are no regulations to
ensure ideological diversity on the Commission, although no more than a simple majority of the
Commission may be staffed by members of the same gender. Id. at § 46.1(3), (4).

Electronic copy available at:

example, an outside super PAC supported by prominent Republicans waged a large-scale
campaign to oust so-called sitting activist judges.29 Similar efforts have taken place in Iowa,
Illinois, Tennessee, Michigan, and North Carolina in recent years as well.30
3. Election
While a slight majority of states employ appointment or retention systems for selecting
judges, a sizeable minority of states allow the electorate to choose who sits on the bench.31 Judicial
elections in these states are either partisan or non-partisan, with candidates declaring a party
affiliation in the former but not in the latter.32
Even in states with non-partisan judicial elections, political party affiliation still plays a
major role in election outcomes. For one thing, although candidates do not run under party
designations, judicial races are often funded by political parties and political action committees
with partisan leanings.33 In addition, even non-partisan judicial candidates have often previously
aligned themselves with one party or the other or have expressed positions on perceived political


See Chris McGreal, “Florida Republicans Wage Campaign to Oust Judges from State Supreme
at (last
viewed July 15, 2020).
See Norm Ornstein, “Courting Corruption: The Auctioning of the Judicial System,” The Atlantic
(Oct. 15, 2014), available at (last viewed July 15, 2020) (describing
largely Republican-led efforts to unseat appointed judges in retention elections).
“Judicial Selection in the States,”, available at https:// (last viewed July 15, 2020).
“Clearly, Nonpartisan Judicial Elections Remain Partisan,” Associated Press (Nov. 11, 2018),
available at (last viewed July 15,

Electronic copy available at:

issues. 34 In these cases, voters often form impressions about a judicial candidate’s political
leanings and implicitly identify them with one political party or the other.35
B. The Changing Role of Judicial Discretion in Sentencing
Criminal sentencing is generally guided by the rhetorical commitment to consistency,
fairness, and non-discrimination.36 Legislative efforts both at the state and federal level purport to
operationalize these values by enacting sentencing schemes to direct judicial sentencing
outcomes.37 However, regardless of the method of judicial selection, trial court judges come to
play a pivotal role in criminal sentencing outcomes as a result of inherently political processes.
Theoretically, then, judges as human beings bring with them predisposed ideologies about crime,
public safety, policing tactics, and other political issues when they take the bench.38 But the
current and historical statutory sentencing schemes fail to take these individual ideological
differences into account, focusing instead in recent years on guiding judicial sentencing discretion
through detailed lists of sentencing factors39 and more historically on holistic characteristics about
the defendant and societal norms in the judge’s geographic area. A review of modern and historical


See, e.g., Anthony Champagne and Kyle Cheek, The Cycle of Judicial Elections: Texas as a
Case Study, 29 Fordham Urb. L. J. 907, 916-18 (Feb. 2002) (discussing ways in which political
parties contribute to and support non-partisan judicial elections); John Futty, “Ohio’s Judicial
Races are Nonpartisan in Name Only, Expert Says,” Columbus Dispatch (Oct. 20, 2017), available
at (last viewed July 16, 2020).
See Laurence Baum, Judicial Elections and Judicial Independence: A Voter’s Perspective, 64
Ohio St. L. J. 16, 21-23 (2003) (discussing ways in which voters form perceptions about judicial
candidates in elections).
36 See e.g. The Protect Act of 2003 (Pub.L. 108–21, 117 Stat. 650, S. 151, enacted April 30,
2003)(identifying principles of consistency and fairness advanced by federal sentencing policy).
37 See United States Sentencing Commission; Brian J. Ostrom, Charles W. Ostrom, Roger A.
Hanson and Matthew Kleiman, Assessing Consistency and Fairness in Sentencing: A
Comparative Study in Three States, NIJ Grant Award 2003-IJ-CX-1015 (August 2008).
For a comprehensive discussion about how judicial ideology informs decision-making, see
Matthew Tokson, Judicial Resistance and Legal Change, 82 U. Chi. L. Rev. 901 (Spring 2015).
See, e.g., Ohio Rev. Code 2929.12 et seq.

Electronic copy available at:

sentencing models is instructive at demonstrating the ways in which the role of judges in criminal
sentencing has shifted over time.
1. Historical Indeterminate Sentencing Models
Indeterminate sentencing models predominated state and federal criminal justice systems
prior to the 1970s and 1980s. 40 Under indeterminate sentencing regimes, defendants were
sentenced to a range of imprisonment and that range set a minimum and maximum term of
incarceration.41 A defendant’s actual release from prison within the range specified by the court
was typically determined by a parole board or other administrative agency. 42 The focus in
indeterminate sentencing schemes was on rehabilitation, premised on the notion that a defendant
would improve himself and be compliant in prison to avoid serving the maximum limit of his
indeterminate sentence.43 Releasing defendants on parole following an indeterminate sentence
was also thought to diminish recidivism and enhance effective reentry into society, although later
research called that assumption into question.44
The role of judges in indeterminate sentencing was marginal at best. While judges were
responsible for issuing an indeterminate sentencing range, the ultimate decision as to a defendant’s
release from incarceration was made by extrajudicial governmental actors, typically a parole


Joshua Logan Pennel, The End of Indeterminate Sentencing in New York: The Death and Rebirth
of Rehabilitation, 58 Buff. L. Rev. 507, 507-8 (April 2010).
As an example of an indeterminate sentence, a defendant convicted of felony robbery might
receive 27 months to 10 years in prison.
Michelle Pifferi, Individualization of Punishment and the Rule of Law: Reshaping Legality in
the United States and Europe between the 19th and the 20th Century, 52 Am. J. Legal Hist. 325,
338-9 (July 2015).
Alan M. Dershowitz, Indeterminate Confinement: Letting the Therapy Fit the Harm, 123 U.
Penn. L. Rev. 297, 301-03 (Dec. 1974) (identifying institutional incentives for rehabilitation postsentencing in indeterminate sentencing schemes).
Gary L. Mason, Indeterminate Sentencing: Cruel and Unusual Punishment, or Just Plain
Cruel?, 16 New Eng. J. on Crim. & Civ. Confinement 89 (Winter 1990).

Electronic copy available at:

board.45 Indeterminate sentencing schemes fell out of favor in the mid-20th Century and by the
late 1990s were nearly all but abolished.46
2. The Determinate “Truth in Sentencing” Movement
In place of indeterminate sentencing, American legal systems shifted towards definite
terms of imprisonment imposed by trial court judges.47 Underlying the change was criticism from
the right that defendants sentenced to indefinite terms of imprisonment wound up serving too little
time and that early release was contributing to rising crime rates. 48 In contrast, left-leaning
advocates argued that indeterminate sentences were too harsh and that broad judicial discretion to
impose wide sentencing ranges produced unfair disparity in sentencing outcomes.49 The result
was a push towards definite terms of imprisonment imposed by the trial court after considering a
range of factors. This came to be known as the “truth in sentencing” movement.
In the federal system, sentencing reform resulted in the adoption of the federal sentencing
guidelines. 50 A bipartisan effort of Congress, the guidelines in essence reduced criminal


Michael M. O’Hear, Beyond Rehabilitation: A New Theory of Indeterminate Sentencing, 48 Am.
Crim. L. Rev. 1247, 1259-60 (Summer 2011) (describing respective roles of the sentencing court
and parole board in indeterminate sentencing schemes).
Pennel, supra note ___.
47 Williams J. Sabol, Katherine Rosich, Kamala Mallik Kane, David P. Kirk and Glenn Dubin,
The Influences of Truth-in-Sentencing Reforms on Changes in State’s Sentencing Practices and
Prison Populations, Report to the National Institute of Justice, Grant #NIJ 98-CE-VX-0006, Urban
Institute Justice Policy Center (2002)(examining the federal government’s role in attaching grant
money to a state’s legislative commitment to, among other things, replace indeterminate
sentencing schemes with determinate sentencing).
Joseph A. Colquitt, Can Alabama Handle the Truth (in Sentencing)?, 60 Ala. L. Rev. 425, 430
Id. (“Liberal-minded individuals complained that terms of imprisonment were too long and
harsh and that judges had too much discretion, which resulted in widely disparate sentences even
for similar offenses. More conservative opponents of the system objected to lenient sentences and
early parole releases, and they blamed the existing practices for the perceived skyrocketing crime
18 U.S.C. § 3553; 28 U.S.C. § 994.

Electronic copy available at:

sentencing outcomes to a mathematical calculation that considered both the severity of the crime
and the defendant’s criminal history.51 By looking at a chart, which contains the offense score on
one axis and the defendant’s criminal history score on another, federal judges were given a narrow
range of months of imprisonment from which they were all but required to select.52
The “truth in sentencing” movement coincided with renewed emphasis in criminal law on
retribution as opposed to rehabilitation. 53 At the same time, state and federal criminal law
expanded more generally to either increase the scope of what conduct was considered a crime or
to lengthen the terms of incarceration associated with particular crimes - or both.54 Incarceration
rates skyrocketed, and along with them the cost of housing millions of inmates across the United
3. Modern Guided Discretion Sentencing Models
Shortly after the introduction of determinate sentencing schemes focused on sentencing
factors, the United States Supreme Court decided United States v. Booker 56 and Blakely v.
Washington.57 In Blakely, the Court held that state sentencing schemes which enable a judge to
impose sentence based upon factual findings not determined by a jury violate the Sixth


David Krajicek, “Birth of a Prison State: The Bipartisan Disaster that put America Behind Bars,”
ner/ (last viewed July 17, 2020); U.S.S.G. Sentencing Table.
U.S.S.G. Sentencing Table; see Frank O. Bowman, The Failure of the Federal Sentencing
Guidelines: A Structural Analysis, 105 Colum. L. Rev. 100, 110 (May 2005).
Colquitt, 60 Ala. L. Rev. at 428-9.
54 See Sabol, et al. supra note 45.
Susan Turner et al., The Impact of Truth-In-Sentencing and Three Strikes Legislation: Prison
Populations, State Budgets, and Crime Rates, 11 Stan. L. & Pol’y R. 75 (Winter 1999).
543 U.S. 220 (2005).
542 U.S. 296 (2004).

Electronic copy available at:

Amendment.58 The Court extended that holding to the federal sentencing guidelines in Booker.59
Numerous state supreme courts also reached similar holdings with respect to the constitutional
validity of judicially-determined sentencing factors.60 As a result of the Booker/Blakely line of
cases, the sentencing factors used by courts to guide determinate sentences are now deemed to be
advisory rather than mandatory.61
4. A Case Study in Sentencing: Ohio
Prior to 1995, most criminal sentences in Ohio, at least in serious felony cases, were
indeterminate.62 Defendants were sentenced to a statutory range by the sentencing court, and the
defendant automatically served the minimum amount of the indeterminate range, less any credit
for “good time” in prison.63 After the minimum term expired, defendants were eligible for parole
and were automatically released at the expiration of the indeterminate sentencing range if they had
not yet been granted release by the parole board.64


543 U.S. 220.
See, e.g., State v. Foster, 109 Ohio St.3d 1 (Ohio 2006) (applying Booker and Blakely to Ohio’s
determinate sentencing scheme).
Booker, 543 U.S. at 233; Kimbrough v. United States, 552 U.S. 85 (2007). The Court later
clarified that state court judges can utilize factors to guide its decision whether to run sentences
consecutively or concurrently without running afoul of the Sixth Amendment. Oregon v. Ice, 555
U.S. 160 (2009). The impact of Booker on federal sentencing decisions is still under debate;
however, disparities in federal sentencing decisions continue to be widespread, and the United
States continues to lead the world in incarceration rates. See, e.g., Joshua B. Fischman and Max
M. Schanzenbach, Racial Disparities Under the Federal Sentencing Guidelines: The Role of
Judicial Discretion and Mandatory Minimums, Journal of Empirical Legal Studies 9 (4): 729–64
(2012); David S. Abrams., Marianne Bertrand, and Sendhil Mullainathan, Do Judges Vary in Their
Treatment of Race? Journal of Legal Studies 41 (2): 347–83 (2012); The Sentencing Project,
Trends in U.S. Corrections, June 2019, available at: (last viewed July 21, 2020).
Baldwin’s Ohio Prac. Crim. L. § 118:3 (Definite Sentences).

Electronic copy available at:

In 1996, the Ohio Legislature passed comprehensive sentencing reform legislation that all
but mandated definite sentencing for all but a very small minority of cases involving life
imprisonment or serious sex offenses.65 The purpose of the law was to shift primary responsibility
in determining sentencing outcomes away from the parole board, an appointed not elected body,
and back to trial court judges elected66 by the people.67 Known as Senate Bill 2, this “truth in
sentencing” bill “established a type of determinate sentencing structure called a presumptive
system that required minimum sentences with judicial discretion from a range of possible
Within a decade following the 1996 sentencing overhaul, Ohio’s prison population soared
to numbers never seen before and larger than predicted.69 These higher numbers were driven by
longer periods of incarceration rather than an increase in the number of offenders sentenced to
prison.70 At the same time, the determinate sentencing scheme enacted in Senate Bill 2 became
the subject of widespread criticism.71 Denounced as too complex and expensive, Ohio’s “truth in
sentencing” scheme itself became the subject of reform efforts.72
Adopted in 2011 and amended in 2012, House Bill 86 made a number of modifications to
Senate Bill 2, the net result of which was to reduce the skyrocketing terms of incarceration
characteristic of “truth in sentencing” reforms.73 Notably, the bill elevated felony theft thresholds,


Baldwin’s Ohio Prac. Crim. L. § 118:3.
Ohio selects judges through non-partisan elections. See Ohio Rev. Code. 3505.04.
Id.; Ohio Criminal Sentencing Commission, “Criminal Justice Reform in Ohio” (April 12,
general/CJReformOhioCupp2019.pdf (last viewed July 17, 2020).
“Criminal Justice Reform in Ohio,” supra note ___, at p. 2.
Id. at p. 4.
Id. at 4-5.

Electronic copy available at:

in effect reducing the prison terms that apply to low-level theft offenses.74 In addition, House Bill
86 eliminated the sentencing disparity between crack and powder cocaine, eliminated certain
sentencing enhancements for drug offenders, and capped sentence lengths for mid-level felony
property and drug offenses.75 After its implementation, Ohio’s prison population dropped to its
lowest point since 2008.76 However, despite these efforts, Ohio’s prison population continues to
outpace projections.77
Criminal sentencing in Ohio today remains a complex endeavor. Trial court judges are
guided by statutory presumptions of either prison, in the case of serious felonies, or probation, in
the case of low-level felonies.78 To depart from a presumption of probation, a judge must make
detailed factual findings supporting an enhanced criminal penalty.79 If a judge elects to sentence
a defendant to prison, the sentencing statutes provide ranges of months or years in prison by level
of felony.80 In deciding the term of imprisonment within the range judges must consider a lengthy
list of factors related to the offense and the offender.81 Despite these legislative directives there is


Id. at 5.
Mary Schladen, Ohio’s Prison Population Grows Despite Justice Reform Study Shows, The
at: (last viewed July 21, 2020); citing Building on Ohio’s Sentencing Changes to
Keep Prison Populations in Check, Alliance for Safety and Justice, Americans for Prosperity Ohio
and the Buckeye Institute.(July 2019), available at:, (last viewed July
21, 2020).
Ohio Rev. Code § 2929.13(B)(1)(a), (D)(1).
Id. at § 2929.13(B)(1).
Id. at § 2929.14.
Id. at §§ 2929.12, 2929.14.

Electronic copy available at:

very little oversight of judicial sentencing decisions.82 Although highly guided by the Legislature,
the elected judiciary in Ohio therefore retains broad discretion in imposing criminal sentences.
A. Research on Party Affiliation and Sentencing Outcomes
Existing scholarly and social science research demonstrates that sentencing disparities exist
both at the state and federal level.83 Efforts to understand these varied sentencing outcomes have
been studied from a variety of perspectives, both within and outside of the legal academy.
Characteristics of individual sentencing judges have been examined and have provided
inconsistent results. While early research in this area focus on differences in judicial sentencing
philosophy, it now appears that these differences can be more narrowly explained by affiliation as
a primary driver of judicial decision making in criminal sentencing.84

For example, an Ohio appellate court’s review of criminal sentences is quite limited. Pursuant
to Ohio Rev. Code § 2953.08 (G)(2), ”[t]he appellate court may [increase, reduce, otherwise
modify, or vacate a sentence] if it clearly and convincingly finds either of the following: (a) That
the record does not support the sentencing court's findings under division (B) or (D) of section
2929.13, division (B)(2)(e) or (C)(4) of section 2929.14, or division (I) of section 2929.20 of the
Revised Code, whichever, if any, is relevant; (b) That the sentence is otherwise contrary to law.
Courts have interpreted this statutory provision to mean that reviewing courts must “clearly and
convincingly find” that either (1) the record does not support the mandatory sentencing findings,
or (2) that the sentence is otherwise contrary to law to overrule a sentence on appeal. See State v.
Marcum, 59 N.E.3d 1231(Ohio 2016); see also State v. White, 997 N.E.2d 629 (Ohio 2013).
83 See, e.g., Ryan D. King and Michael T. Light, Have Racial and Ethnic Sentencing Disparities
Declined?, 48 Crime & Just. 365 (2019) (summarizing historical and existing sentencing
disparities based on race); Christine DeMaso, Advisory Sentencing and the Federalization of
Crime: Should Federal Sentencing Judges Consider the Disparity between State and Federal
Sentences under Booker?, 106 Colum. L. Rev. 2095 (Dec. 2006) (discussing disparities between
state and federal sentences for similar crimes).
Lee Epstein, William M. Landes and Richard A. Posner, The Behavior of Federal Judges.
Cambridge, MA: Harvard Univ. Press (2013); Lee Epstein and Jack Knight, The Choices Justices
Make, Washington, D.C.: CQ Press (1997); Lydia Tiede, Robert Carp and Kenneth Manning,
Judicial Attributes and Sentencing Deviation Cases: Do Sex Race and Politics Matter?, The
Justice System Journal, Vol. 31 Number 3 (2010); c.f. Orley Ashenfelter, Theodore Eisenberg and
Stewart J. Schwab, Politics and the Judiciary: The Influence of Judicial Background on Case

Electronic copy available at:

The primary focus of prior academic research has been aimed explaining decision making
at the federal level, especially at the Supreme Court.85 There have been a few efforts to understand
criminal sentencing decisions in federal district courts. For example, in a study published by the
University of Chicago Law Review, Professors Schanzenbach and Tiller looked at serious drug
offenses by randomly selecting dates in the federal database.86 Using a sample of 2,265 cases,
they found that there were statistically significant differences in how judges make decisions based
on the party affiliation of the president who appointed the judge. Their research found that that
judges appointed by Democratic presidents reduced offense level and lowered sentences more
often than their counterparts appointed by Republican presidents. This study confirmed prior
research by the same authors which examined sentencing decisions aggregated at court-level
variation in the percent of Democratic or Republican-appointed judges within a district court to
study the impact of political affiliation on sentencing.87 In this original study, researchers found
that political affiliation of both the circuit and the district courts affects sentencing decisions. The
authors replicated the study to correct for a reliance on aggregate court level data. As the authors
and others have noted, the use of aggregate court level data can lead to skewed results 88

Outcomes, The Journal of Legal Studies vol. XXIV (June 1995)(arguing against political party as
a predictor of judicial decisions in federal civil rights litigation).
See, e.g. Frank B. Cross and Emerson H. Tiller, Judicial Partisanship and Obedience to Legal
Doctrine: Whistleblowing on the Federal Court of Appeals, 107 Yale L. J. 2155 (1998).
86 Max M. Schanzenbach and Emerson H. Tiller, Reviewing the Sentencing Guidelines: Judicial
Politics, Empirical Evidence and Reform, 75 U. Chicago L. Rev. 715, 731 (2008).
87 Schanzenbach, Max M., and Emerson H. Tiller, Strategic Judging Under the U.S. Sentencing
Guidelines: Positive Political Theory and Evidence. Journal of Law, Economics, and
Organization 23 (1): 24–56 (2007).
88 See Alma Cohen and Crystal S. Yang, Judicial Politics and Sentencing Decisions, American
Economic Journal: Economic Policy 2019, 11(1): 160-191 (2019) (arguing that court level data
can lead to biased results “if courts with different compositions differ in ways that affect all judges
in the district court, or if the partisan composition of a court is correlated with unobservables that
affect sentencing.”)

Electronic copy available at:

Republicans judges (coded as judges appointed by Republican presidents) gave higher sentences
than Democratic judges (those appointed by Democrat presidents). The authors acknowledged
that the implications of the study were limited by the inability to connect sentencing decisions to
individual judges. The follow up study addressed this issue by examining individual sentencing
outcomes. Other researchers have noted that the Schanzenbach and Tiller studies were the first
known to look at how political party affiliation of the appointing president impacts sentencing
Attempting to ascertain the effect of United States v. Booker on sentencing outcomes,
another group of researchers reviewed sentencing decisions by federal court judges between 1997
and 2008 and specifically studied the decision of the judge to depart from the sentencing
guidelines.90 The research found that that political party of the president appointing the judge
affects the probability of voting for defendants in sentencing deviation cases.91 Judges associated
with the Republican party were less likely to making sentencing departures in favor of defendants
than judges associated with the Democratic party.92
B. The Harvard Political Sentencing Study
Following these studies, Cohen and Yang (the authors of the Harvard political sentencing
study) examined over five hundred thousand cases of individual judge-level sentencing outcomes
between 1999 and 2015.93 The cases were drawn from across the federal courts and included 1,398

Lydia Tiede, Robert Carp and Kenneth Manning, Judicial Attributes and Sentencing Deviation
Cases: Do Sex Race and Politics Matter?, The Justice System Journal, Vol. 31 Number 3, p. 129
(2010); see also Alma Cohen and Crystal S. Yang, Judicial Politics and Sentencing Decisions,
American Economic Journal: Economic Policy 2019, 11(1): 160–191 (2019).
90 Id.
91 Id.
92 Id.
93 Alma Cohen and Crystal S. Yang, Judicial Politics and Sentencing Decisions, American
Economic Journal: Economic Policy 2019, 11(1): 160–191 (2019).

Electronic copy available at:

judges.94 Using databases from the United States Sentence Commission, the Transaction Access
Clearinghouse and the Federal Judicial Center, the authors were able to collect data for a variety
of offender, case and judge characteristics. Cohen and Yang investigated whether the political
affiliation of a judge impacts sentencing outcomes.95 Specifically, the authors hypothesized that
judicial political party affiliation can explain the racial and gender disparities in sentencing.96 The
study looked at how judges appointed by a Republican president sentence black offenders
compared to nonblack, or female versus male offenders, relative to judges appointed by a
Democratic president. 97 Cohen and Yang confirm prior research by finding “economically
meaningful and statistically significant evidence that judge political affiliation is a source of
disparities in federal sentencing.”98 The study concentrated on racial and gender disparities and
was able to control for other judicial characteristics such as judge race, gender, former
prosecutorial experience, or proxies for racial bias. 99 Importantly, the study finds that the
sentencing disparity both between Republican judges and Democrat judges and the race and gender
disparities increase the with greater sentencing discretion.100 Researchers were able to isolate this
finding by looking at federal sentencing outcomes in the pre-Booker and post-Booker settings.101
C. Research on State Court Judges and Political Affiliation

96 Id.
97 Id.
98 Id.. at 162.
99 Id.
100 Id.
101 Id. at 163 (Booker, 542 U.S. 220 (2005) held that mandatory sentencing guidelines violated the
Sixth Amendment right to trial and, to avoid unconstitutional sentencing outcomes, determined
that the application of U.S. Sentencing Guidelines was solely advisory.)

Electronic copy available at:

Research on the political influences of judicial decision making in state courts has focused
on the impact of politics generally on judicial behavior. These studies have not examined
individual judge level sentencing outcomes related to party affiliation. In one comprehensive
study of state court judges, Yale University researcher Gregory Huber and Sanford Gordon looked
at over 22,000 sentencing outcomes in Pennsylvania to see if the election cycle had an impact on
judicial decision-making.102 The study found robust evidence that elected judges become more
punitive the closer they get to reelection.103 The authors attribute 1,818 to 2,705 additional years
of incarceration were attributable to judicial reelection for the cases they examined.104 In a similar
study, Carlos Berdejó and Noam Yuchtman researched the impact of re-election on sentencing
judges in Washington State.105 Relying on over two hundred thousand criminal cases heard by 265
trial court judges in Washington from 1995-2006, researchers concluded that sentences were 10
percent longer at the end of a trial judge’s political cycle than at the beginning.106 In another study
Emory Law School’s Joanna Shepherd and Michael Kang conclude that state supreme court
justices are more likely to rule in favor of the state in criminal cases when faced with the threat of
future attack ads.107

Gregory A. Huber and Sanford C. Gordon, Accountability and Coercion: Is Justice Blind When
it Runs for Office, American Journal of Political Science, Vol. 48, No. 2 (Apr. 2004), pp. 247-263.
103 Id.
104 Id.
105 Carlos Berdejó and Noam Yuchtman, Crime, Punishment, and Politics: An Analysis of Political
Cycles in Criminal Sentencing, 95 Rev. Econ. & Stat. 741, 755 (2013).
106 Id.
107 Joanna Shepherd & Michael S. Kang, Skewed Justice: Citizens United, Television Advertising
and State Supreme Court Justices’ Decisions in Criminal Cases (2014), available at
Decisions.pdf (last viewed July 19, 2020).

Electronic copy available at:

Researchers also compared the two selection systems in Kansas to measure the influence
of the selection method on the behavior of state court judges and their criminal sentencing
decisions.108 Using quantitative analysis to assess the preference heterogeneity versus reelection
incentives in determining sentencing decisions, the study determined that the sentencing behavior
of elected judges is far more variable than that of appointed judges.109 The study also found that
sentencing severity of elected judges is strongly related to the political ideology of the voters in
their districts, while that of appointed judges is not.110 Furthermore, appointed judges’ preferences
are far more homogenous than those of their elected counterparts.111 Like other research focused
on state court judges, this study did not use individual level judge data, but focused on sentencing
decisions in the aggregate. 112 Additionally, consistent with the majority of other studies, this
research explored the interaction of sentencing decisions and elections.113 However, these studies
are instructive because the results establish the foundational hypothesis for the Harvard political
sentencing study on federal court judges and the current study on state court judges and the impact
of political affiliation.
Research collected from individual judge level sentencing outcomes in the state of Ohio
confirms the influence of political party affiliation on sentencing outcomes at the state level. Ohio
provides an interesting study model, in that its judges are elected in non-partisan elections and
therefore theoretically more removed from political influence than appointed federal judges and


Claire S.H. Lim, Preferences and Incentives of Appointed and Elected Public Officials:
Evidence from State Trial Court Judges, American Economic Review 103(4): 1360–1397 (2013).

Electronic copy available at:

state court judges selected through directly political processes.114 However, while Ohio judicial
elections are in labeled as non-partisan, recent legislative changes permit judicial candidates to
receive party endorsements and to identify party affiliations on their campaign advertising.115 As
a result, Ohio judges are selected by an outwardly non-partisan method that is inwardly and
inherently political.
As we report below, similar to appointed federal judges, Ohio’s elected trial court judges
sentence defendants to disparate terms of imprisonment based on their political party affiliation,
with Republican-affiliated judges sentencing defendants more harshly than Democratic-affiliated
judges. In this regard, our research extends the findings of the Harvard political sentencing study
to state courts.
A. Study Methodology
The current sentencing study examines the impact of judicial political party
affiliation on sentencing outcomes for state court judges. Consistent with the findings reported by
the Harvard political sentencing study and others, we hypothesized that sentencing outcomes for
Republican judges would be more severe than for Democratic judges. This study is unique in its
analysis of judicial sentencing and political affiliation, as the vast majority of studies rely on
readily available data collected and maintained at the federal level. The lack of comprehensive


0at,or%20appointed%20to%20the%20Court (last viewed July 19, 2020).
See Ohio Code of Jud. Conduct Canon 4. Widespread change to judicial election rules took
place in the wake of Republican Party of Minnesota v. White, 536 U.S. 765 (2002), which found
a First Amendment violation when judicial candidates are barred from speaking about their
political party affiliations. See also Note, Voting and Democracy, 119 Harv. L. Rev. 1133 (Feb.
2006) (discussing impact of White on judicial elections and arguing that “doctrinal developments
will affect more than just the long-running--and perhaps insoluble--debate about the proper
balance of democratic accountability and judicial impartiality … [and] will also play a large role
in distributing power among groups vying for influence in judicial elections.”).

Electronic copy available at:

available data to review and measure judicial sentencing outcomes appears to be a common
problem for many states, including Ohio.116 The current study is also distinct in that we were able
to connect individual judges to sentencing outcomes, which addresses a gap in the research
literature related to the sentences practices of particular judicial officials.117
This study examined the sentencing practices of 40 sitting Ohio trial court judges, 20 from
each political party, in at least 20 cases per judge.118 For each of the 20 cases for each judge, the
case docket for the individual defendant was accessed through the county clerk’s publicly available
website.119 For each case, the docket was examined for the sentencing entry, which we looked at
to determine the length of sentence, which was then recorded for the sentencing judge. 120 Each
sentencing decision was coded on a nine-point scale using one for probation and nine for a life

The sentence decision coding reflects an approximation of potential sentencing

See Neal B. Kauder and Brian J. Ostrom, Sentencing Guidelines Profiles and Continuum,
National Center for State Courts (2008). Some judges in Ohio have recognized the lack of a
centralized repository of information on sentencing in the state; however, the Ohio Supreme Court
has yet to take corrective action or to mandate consistent data collection practices in the state trial
courts. Michael P. Donnelly, Associate Justice Ohio Supreme Court and Ray Headen, Judge, 8th
Ohio District Court of Appeals, Guest Columnist, Create Centralized Criminal Sentencing
Database to Reduce Mass Incarceration in Ohio,, Jan. 8, 2020, available at: (last viewed July 21,
117 See e.g. Max M. Schanzenbach, Racial and Sex Disparities in Prison Sentences: The Effect of
District-level Judicial Demographics, 34 J Legal Stud 57, 85-90 (2005) (relying on variation of
sentencing outcomes at the district level.); Alma Cohen and Crystal S. Yang, Judicial Politics and
Sentencing Decisions, American Economic Journal: Economic Policy 2019, 11(1): 160–191
(2019) (noting that efforts to estimate the impact of judge political affiliation on sentencing
decisions have been complicated by the lack of data linking judge identifiers to defendant
characteristics and case outcomes.)
See Research Notes and Excel Data Collection Spreadsheets (on file with authors).
119 Id.
120 Id.

Electronic copy available at:

outcomes by degree of offense according to Ohio sentencing legislative directives. 121 The
considerable investment required to access quality data is a significant barrier to the analysis of
state judges sentencing outcomes.
A sample of 796 court cases were randomly selected for inclusion in the current study.
Selection was conducted using multistage-cluster sampling techniques in which we first identified
the 391 common pleas court judges in Ohio and then categorized each judge as either Republicanaffiliated or Democratic-affiliated based on political party endorsement.122 Next we randomly
selected 40 judges based on their political affiliation to have an equal representation of Republicanaffiliated (n = 20) and Democrat-affiliated (n = 20) judges. Then we selected approximately 20
cases at random from each of the 40 judges during a three-year time span ranging from October
2017 to May 2020. Cases were selected for inclusion in the data set based on case filing rather
than sentencing date.123 We were able to achieve random case selection by relying on Ohio’s use
of the random assignment system for criminal cases in the common pleas division.124
The resulting sample included 395 cases that were sentenced by Republican-affiliated
judges and 401 cases sentenced by Democrat-affiliated judges. 125 Most of the cases (i.e.,
approximately 98%) ended in a plea deal.126 Only 20 of the 796 cases included in the sample went

Sentencing options in Ohio are proscribed by statute and dictated by offense level. Ohio
Revised Code 2929.13. Our research categorized sentencing outcomes based upon the ranges set
forth in the state sentencing statutes.
122 As discussed supra, Ohio judicial candidates are permitted to receive and advertise a party
endorsement, although the ballot entries remain non-partisan.
123 See Max M. Schanzenbach and Emerson H. Tiller, Reviewing the Sentencing Guidelines:
Judicial Politics, Empirical Evidence and Reform, 75 U. Chicago L. Rev. 715, 731 (2008).
124 Rules of Superintendence for the Courts of Ohio 36.01 et seq.
125 See Research Notes and Excel Data Collection Spreadsheets (on file with authors).

Electronic copy available at:

to trial.127 The average age of defendant was approximately 34 years (SD = 10.40 years).128 A
large majority (i.e., 81%) of the defendants were male.129
At the outset, a number of obstacles made our research more difficult than that undertaken
by the Harvard political sentencing study. For example, Ohio lacks a uniform sentencing database
from which to analyze sentencing outcomes. Each of the 88 counties maintains its own databank
linked only to each individual criminal defendant.130 Many demographic factors are not reported
and, if the factors are captured, there is no uniformity in reporting methods.131 Collecting the data
necessary to study sentencing outcomes therefore necessitated individual records searches on nonuniform websites operated at the individual county level.
This study attempted to collect race, gender and age information; however, the lack of
consistency and accuracy in reporting race data across county clerk of court systems made it
impossible to report race with precision.132 As a result, the Ohio study was unable to replicate the
findings of the Harvard political sentencing study relative to racially disparate sentencing
outcomes based on judicial party affiliation.


See, e.g., (Hamilton County, Ohio criminal case public access database); (Summit County, Ohio records search database).
For example, compare the search results for “John Smith” obtained through the Hamilton
County, Ohio public access database (available by searching the name feature in the “records
search” tab at and through the Summit County, Ohio public access database
(available by using the “records search” tab at
However, the overrepresentation of Black people in Ohio prisons is well
documented. Black people constitute 13% of the state population, but comprise 34% of the county
jail population and an astounding 45% of the state prison population. See, Incarceration Trends
in Ohio, Vera Institute (2015), available at (last viewed July 19, 2020).

Electronic copy available at:

B. Study Results
The major outcome variable in this study is the judges’ sentencing decision, which was
coded using a 1 (probation) to 9 (life in prison) scale with higher scores reflecting a more severe
sentencing decision.133 As seen in Table 1, the distribution of sentencing was positively skewed
(see Table 1), with a majority of the defendants (i.e., approximately 49%) receiving the minimum
sentencing (i.e., probation).134 The average severity of sentencing score across all judges and cases
in the sample was 2.65 (SD = 2.13) on the 9-point scale. The median score, representing the
midpoint of the distribution of all sentencing decisions, was 2, which reflects a sentence of 0-6
Table 1. Frequency table describing sentencing decisions within the sample of 796 court cases
Sentencing Decision

Number of Cases

Percent of Cases






0-6 months




7-12 months




13-24 months




25-35 months




3-5 years




6-10 years




10 years or more







See Research Notes and Excel Data Collection Spreadsheets (on file with authors).

Electronic copy available at:

The average prison sentence across Republican-affiliated and Democratic-affiliated judges
was 4.25 on the 9-point scale, with a median score of 4, which represents a prison sentence of 1324 months..136 This is consistent with data reported by the Bureau of Justice Statistics of the
average time served by state prisoners. 137 As seen in Table 2, additional frequency analyses
examining the most frequently occurring sentencing decision to Republican-affiliated and
Democrat-affiliated judges where the defendant was sentenced to serve time in prison revealed
that while the median sentencing score was the same for both groups (i.e. a prison sentence of 1324 months), the most frequently occurring sentencing decision among the Democrat-affiliated
judges was less severe than those made by Republican-affiliated judges. More specifically, while
the most frequently occurring sentencing decision for Democrat-affiliated judges was 0-6 months,
the most frequently occurring sentencing decision for Republican-affiliated judges was 7-12
Table 2.

Frequency table describing Republican-affiliated and Democrat-affiliated judges

sentencing decisions that resulted in prison time.
Number (and Percent) of Cases
Sentencing Decision




0-6 months

45 (19%)

45 (27%)


7-12 months

61 (26%)

26 (15%)


13-24 months

41 (17%)

33 (20%)


25-35 months

14 (6%)

6 (4%)


3-5 years

44 (19%)

38 (23%)


6-10 years

17 (7%)

13 (8%)

See Research Notes and Excel Data Collection Spreadsheets (on file with authors).
Danielle Kaeble, Time Served in State Prison, 2016, Bureau of Justice Statistics (Nov. 29,

Electronic copy available at:


10 years or more

6 (3%)

4 (2%)



7 (3%)

4 (2%)




After examining the frequency of the judge’s sentencing decisions, a series of preliminary
correlations were conducted to examine the extent to which the defendant’s age and gender, as
well as his/her decision to go to trial, were associated with the judges’ sentencing decisions.
Results revealed that although there was no association between the defendants’ age and the
severity of the defendants’ sentence (r = .03, p = .53), males received more severe sentences than
females (r = .12, p = .001), and defendants who went to trial received more severe sentences than
those who decided to take a plea deal (r = .24, p < .001).138 These findings confirm those of the
prevailing literature.139
To determine the extent to which the political affiliation of the judge explained differences
in the judge’s sentencing decisions, and directly test the hypothesis of the study a one-way Analysis
of Covariance (ANCOVA) was conducted on the judge’s sentencing decisions controlling for
gender of the defendant and his/her decision to take a plea deal.140 As seen in Figure 1, results
were consistent with our prediction. After controlling for any variance in the judge’s sentencing

See e.g. Sonja B. Starr, Estimating Gender Disparities in Federal Criminal Cases, Law and
Economics Working Papers, 57 (2012)(studying federal sentencing outcomes and finding that
female offenders are sentenced less harshly than male offenders); Charles W. Ostrom, Brian J.
Ostrom and Matthew Kleiman, Judges and Discrimination: Assessing the Theory and Practice of
Criminal Sentencing, NIJ Grant No. 98-CE-VC-0008 (2004) (determining that female offenders
are sentenced more harshly than males at the state level); Steven P. Grossman, Making the Evil
Less Necessary and the Necessary Less Evil: Towards a More Honest and Robust System of Plea
Bargaining, 18 Nev. L.J. 769 (2018)(discussing the research findings on severity of sentence
driven by case resolution of plea or trial).
See Research Notes and Excel Data Collection Spreadsheets (on file with authors).

Electronic copy available at:

decisions that can be attributed to the defendant’s gender and whether the defendant went to trial,
Republican-affiliated judges were more severe in their sentencing (M = 2.94, SD = 2.17) than
Democratic-affiliated judges (M = 2.38, SD = 2.05), F(1, 787) = 13.05, p < .001, ηp2 = .02.141

Severity of Sentence

Republican-Affiliated Judges

Democrat-Affliatiated Judges

Figure 1. Average sentencing of Republican- and Democrat-affiliated judges. Error bars
represent standard error of the mean.
Although not central to the purpose of the current study, a series of exploratory independent
samples t-test were conducted to examine the extent to which Republican and Democrat-affiliated
judges differed in their sentencing of male and female defendants. As seen in Figure 2, results
revealed that there was no difference between male and female defendants among Democrataffiliated judges, t(394) = 1.59, p = .11. However, and in contrast to their Democrat-affiliated
counterparts, Republican judges sentenced male defendants (M = 3.11, SD = 2.23) much more
severely than female defendants, t(393) = 3.03, p = .001.142


It should be noted here that this effect may skewed by the fact that there were significantly
more male defendants in this data set than females. However, this finding does confirm the
Harvard study finding that Republican judges give female defendants less prison time than
similar male defendants compared to Democratic judges. Cohen and Alma, supra p. 162.

Electronic copy available at:

Severity of Sentencing







Republican-Affiliated Judges
D Male Defendants

Democrat Affiliated Judges
D Female Defendants

Figure 2. Average sentencing of male and female defendants as a function of political affiliation
of judge. Error bars represent standard error of the mean.
C. Study Conclusions
The main findings of the current study confirm the research of the Harvard political
sentencing study as well as that of Professors Schazenbach and Tiller. This study extends the
findings of this previous research by determining that at the state level, Republican-affiliated
judges sentence criminal defendants more severely than Democrat-affiliated judges. Consistent
with the Harvard political sentencing study’s conclusion regarding the influence of political
affiliation on federal sentencing outcomes, this study finds a similar political bias in state court
judges when it comes to sentencing outcomes in criminal cases.
The statistically significant findings of this study shed light on an unexplored area of
judicial decision making and politics – the state courts. State courts handle more than 90% of the
cases the United States.143 This analysis suggests that the influence of political party affiliation on

Court Statistics Project, Examining the Work of State Courts: An Overview of 2015 State Court
Caseloads, CONF. ST. CT. ADMINS. & NAT'L CTR. FOR ST. CTS. 3 (2016),

Electronic copy available at:

judicial decision making is prevalent in the vast majority of criminal cases in the United States.
While more research is needed in the state court arena, increasing the sample size and controlling
for additional variables, including race and the seriousness of the crime, the conclusions of this
study contribute significantly to the research on the influence of political party affiliation and
sentencing outcomes.
The combined findings of the current state court sentencing study and the Harvard political
sentencing study are troubling in light of the focus of modern legislative sentencing reform efforts
to create sentencing consistency and fairness. In addition to legislative efforts to bring consistency
to judicial decision-making, scholars have expended substantial effort in making determinations
about which method of judicial selection creates the most fairness on the bench. 144 The
politicization of the federal court selection process has been well documented. 145 From the
political nature of presidential nomination process to the partisan divide in Senate confirmation
hearings political bias continues to be an issue in the federal judiciary.146 In an effort to curb the
effect of these political influences on the judiciary, judicial elections were originally conceived as
a way to address the partisanship of judicial appointments.147 However, as discussed above, prior [
144 See e.g. Peter D. Webster, Selection and Retention of Judges: Is There One “Best” Method?
23 Fla. St. U. L. Rev. 1 (1995-1996); Chris W. Bonneau and Melinda Gann Hall, In Defense of
Judicial Elections, Routledge (2009); F. Andrew Hanssen, The Effect of Judicial Institutions on
Uncertainty and the Rate of Litigation: The Election Versus Appointment of State Court Judges,
The Journal of Legal Studies Vol. 28, Number 1 (1999).
145 David Weiden, Judicial Polticization, Ideology and Activism at the High Courts of the United
States, Canada and Australia, Political Research Quarterly, Vol. 42 Issue 2 (2011).
146 John Ferejohn, Judicializing Politics, Politicizing Law, 65 Law & Contemp. Probs. 41, 66
(2002); see also Thomas L. Jipping, From Least Dangerous Branch to Most Profound Legacy:
The High Stakes in Judicial Selection, 4 Tex. Rev. L. & Pol.. 365 (2000) (discussing the growing
political nature of the federal bench).
147 Jed Handelsman Schugerman, The People’s Court, Harvard University Press (2012).

Electronic copy available at:

research has demonstrated that judicial sentencing decisions have been overtly influenced by the
presence of elections.148 In addition, popular media as well as the American Bar Association have
called attention to the partisanship of judicial elections noting that the selection of state court
judges “. . . has become increasingly politicized, polarized, and dominated by special interests . .
..”149 If consistency and the elimination of discrimination are sentencing values the Legislature
seeks, the current sentencing study suggests that the focus should be less on judicial selection
methodology or direct legislative solutions and more on empowering the public with information
and mechanisms to ensure judicial decision making consistent with these values.150
Given that political sentencing disparities persist despite legislative sentencing reform
efforts – a finding confirmed by both the Harvard political sentencing study and the current
sentencing study – further sentencing reform efforts should be focused on eliminating sentencing
disparities arising from judicial political affiliation. Three possible solutions – one focused on
educating the judiciary, one on informing the public and one focused on arming the parties in
criminal cases with the ability to ensure more moderate sentencing outcomes – may advance the
underlying goals of modern sentencing reform: fairness, consistency, and impartiality.

See supra note 105.
American Bar Association, Rethinking Judicial Selection, (March 1, 2016), available at
er/2016/volume-24-number-1/rethinking_judicial_selection/; Adam Liptak, Judges Who are
Elected Like Politicians Tend to Act Like Them, The New York Times, (Oct. 3, 2016), available
at: (last
viewed July 19, 2020).
150 Many scholars have advocated that sentencing be moved from the purview of judges to juries;
however, this approach has also been criticized as yielding more disparate and severe sentences
than those crafted by judges. See Nancy J. King and Roosevelt L. Noble, Jury Sentencing in
Noncapital Cases: Comparing Severity and Variance with Jud. Sentences in Two States, 2 Journal
of Empirical Legal Studies 331, 332 (2005).

Electronic copy available at:

A. Data-Based Solutions
The use of data to inform sentence decisions has gained popularity in the form of risk/needs
assessments and other predictive algorithms.151 However, less attention has been given to the use
of judicial sentencing data at the individual level to shape sentencing outcomes. While there is
robust data collected regarding the offender and judicial characteristics related to sentencing in
federal courts, there is a lack of data linking judicial identifiers to defendant-specific characteristics
and case outcomes.152 In the states, the availability of data on sentencing outcomes connected to
individual judges varies greatly, and, as a result, the ability to connect sentencing outcomes to
individual judges is largely absent. 153 Requiring individual jurisdictions to collect relevant
offender characteristics, such as race, gender, and age, for each criminal case and to organize
specific offender according to sentencing judge is a necessary first step. Making this information
available to future researchers will allow further exploration on the significance of judicial
characteristics, including political affiliation, on sentencing decisions at the judge specific level.
In addition, placing individual level judicial sentencing data organized by offender characteristics
in the hands of individual judges would allow judicial actors to become informed about their
sentencing practices relative to other judges in their county and state.

Possession of this

information would provide an opportunity for judges to self-correct implicit and potential


See, e.g., Ohio Rev. Code § 5120.114 (mandating use of single validated risk assessment tool
in Ohio criminal cases).
152 Cohen and Alma, supra, note ___, at 161.
153 Neal B. Kauder and Brian J. Ostrom, State Sentencing Guidelines Profiles and Continuum,
National Center for State Courts (2008); Ohio Criminal Sentencing Commission, The Data
Disconnect: Adult Criminal Justice Data in Ohio, (Jan. 2019), available
at (last
viewed July 19, 2020).

Electronic copy available at:

unrecognized racial, gender, and proportionality biases in their own sentencing practices.154 In
theory, judges who are better informed as to their sentencing practices in the aggregate and
potential implicit biases inherent in sentencing outcomes will make more principled decisions.
Assuming the disparate sentencing practices of judicial actors is the product of unconscious
bias, shining light on an individual’s patterns and practices would help inform judicial behavior.155
To help place this information in context, aggregate sentencing data organized by offense level for
sentencing outcomes across the state and within individual judicial jurisdictions should be made
available. To further the impact of this data, for judges who substantially deviate from sentence
averages, court rules should require counseling and training to assist judges in correcting
sentencing biases.156
There is some evidence in the juvenile justice context that improved data collection around
racial disparities in the criminal justice system can lead to systemic improvement.


specifically, the Juvenile Justice and Delinquency Prevention Act of 2002 (“the JJDPA”) required
states accepting grant money to track disproportionate minority contact and to implement plans
designed to reduce any disparities.157 States that fail to address disproportionate minority contact
stand to lose up to 20 percent of grant funding in subsequent years.158 While the results of the
JJDPA are mixed, at least 34 states had implemented strategies to reduce disparities revealed from
the JJDPA-mandated disproportionate minority contact data collection, and four states had tested


See, e.g., Olatunde C.A. Johnson, Disparity Rules, 107 Colum. L. Rev. 374 (March 2007).
Jeffrey J. Rachlinski, Sheri Johnson, Andrew J. Wistrich and Chris Guthrie, Does Unconscious
Racial Bias Affect Trial Judges,? 84 Notre Dame L. Rev. 1196 (2008-2009)(finding that while
judges are influenced by implicit bias, they are motivated to avoid it); see also Nancy J. King and
Roosevelt L. Noble, Felony Jury Sentencing in Practice: A Three State Study, 57 Vand. L. Rev.
886 (2004).
156 Id.
See 34 U.S.C. § 11101 et seq.
158 Id.

Electronic copy available at:

the efficacy of those strategies.159 While not a perfect example, the JJDPA reveals that measurable
progress can be made towards eliminating criminal justice disparities following the improvement
of data collection practices.
B. Public Information
The public availability of information on individual judicial sentencing outcomes could
also serve as a mechanism for reform in its own right, allowing the electorate to make more
informed choices in judicial elections. Judicial elections were in part formulated as a way to create
judicial accountability.160 However, the premise inherent in voter participation in the selection of
judges is that voters understand the rule of law and appreciate how the judicial decisions in their
jurisdictions fit within that framework. The problem is that voters are woefully uninformed in
almost all elections and particularly in judicial elections.161 Voter ignorance in judicial elections
abounds because the majority of the work conducted by judges is done in the courthouse outside
of the public eye, the way judges make decisions is mysterious to the public, and there is a “lack
of useful cues and heuristics that allow voters to compensate for their lack of relevant
knowledge.” 162 Voters specifically lack information about the identity of the judges in their
jurisdiction, but they are also unaware of what judges do and unable to compare what is actually
happening in courtrooms to what should be happening under Constitutional and other judicial
mandates.163 Other scholars have proposed the implementation of judicial evaluations and the


Megan Mason, Judges’ Role in Correcting the Overrepresentation of Minority Youth in the
Juvenile Justice System, 28 Geo. J. Legal Ethics 719, 720-21 (Summer 2015).
160 See e.g. Glenn R. Winters, Selection of Judges—An Historical Introduction, 44 Tex. L. Rev.
1081, 1082 (1966); Minnesota v. White, 536 U.S. 765, 785 (2002) (finding that judicial elections
were manifested in the state as a response to democratic concerns); Chris W. Bonneau & Melinda
Gann Hall, In Defense of Judicial Elections 8 (2009).
161 Dmitry Bam, Voter Ignorance and Judicial Elections, 102 Ky. L. J. 554 (2013).
162 Id. at 565-66.
163 Richard H. Fallon, Jr., Legitimacy and the Constitution, 118 Harv. L. Rev. 1787, 1825 (2005).

Electronic copy available at:

creation of judicial commissions to aggregate data on characteristic related to judicial
temperament, promptness and impartiality.164 The dissemination of this information would surely
be helpful to voters. However, the collection and distribution of data on the objective metrics of
average sentence length by offense level, race, and gender compared with other judges in the state
would provide voters with impartial and targeted information to inform voting decisions. This
kind of information is readily intelligible by lay people and fits into a framework that the average
voter already uses to make decisions.
C. Judicial Peremptory Strikes
An additional solution to sentencing disparities created by judicial political affiliation
would be to provide the parties with a small number of judicial peremptory strikes. This idea has
historical roots, dating back to a California civil procedure rule adopted nearly a hundred years
ago.165 The specific text of the rule stated:
Any party or his attorney to any cause or proceeding of any nature pending in a
superior or municipal court, except the people or district attorney in a criminal case,
may make and file with the clerk of the court in which the action is pending, and
serve on the opposite party, a peremptory challenge in writing of the judge assigned
to try or hear the cause or pending matter. Thereupon, without any further act or
proof, the presiding judge in those counties where there is a presiding judge who
assigns causes for hearing or trial, or the chairman of the judicial council in other
counties, shall assign some other judge to try the cause or hear the pending matter,
and such cause shall be continued on the calendar until the judge so secured or
assigned can try the cause or hear the matter. If it is necessary to secure a judge
from another county, the chairman of the judicial council shall assign such judge.

See, e.g., Stephen J. Choi & G. Mitu Gulati, Choosing the Next Supreme Court Justice: An
Empirical Ranking of Judge Performance, 78 S. Cal. L. Rev. 23, 31–34 (2004).
See Austin v. Lambert, 11 Cal.2d 73, 77 P.2d 849 (Cal. 1938) (declaring California rule allowing
judicial peremptory strikes unconstitutional on separation of powers grounds).

Electronic copy available at:

Although the rule was ultimately declared unconstitutional on separation of powers grounds, it
gained significant early scholarly support, particularly as an alternative to inefficient and
cumbersome methods for seeking judicial recusal.166
The idea of judicial peremptory challenges has resurged with new force over the past two
decades167 and a sizeable number of states in which judges are elected now allow some form of
peremptory judicial exclusion.168 Alaska, for example, requires mandatory disqualification of a
judge if a party files an affidavit alleging that judge cannot be fair and impartial.169
The expansion of judicial peremptory strikes in criminal cases, particularly following the
collection and publication of sentencing data for individual judges, would make a measurable
difference in reducing sentencing disparities. It is reasonable to assume that prosecutors would
strike judges with the most lenient sentencing practices, and criminal defendants would strike
judges with the most severe sentencing practices, thus causing sentencing outcomes to move
towards the middle. This practice would also bolster the information feedback described above.
Judges who are routinely disqualified from participating in criminal cases would be compelled to
reflect on and to evaluate their case decisions in the context of this feedback from the parties. Care
should be taken in crafting judicial peremptory challenge policies to ensure that parties do not


See, e.g., Note, Disqualification of Judges by Peremptory Challenge, 47 Yale L. J. 1403, 1408
(1938) (“If the issue [of judicial bias] should arise in states accustomed to disqualify their judges
by simple affidavit, the courts might well approve a shift to peremptory challenge as a
comparatively minor simplification of the process.”).
See, e.g., Debra Lyn Bassett, Judicial Disqualification in the Federal Appellate Courts, 87 Iowa
L. Rev. 1213, 1251 (May 2002) (proposing the use of judicial peremptory strikes in the federal
appellate courts).
Roy A. Schotland, New Challenges to States’ Judicial Selection, 95 Geo. L. J. 1077, 1102 (April
Alaska Stat. § 22.20.022; Marla L. Greenstein, Judicial Disqualification in Alaska Courts, 17
Alaska Law Review, 53 (2000) (for a comprehensive overview of the Alaska rules of

Electronic copy available at:

make use of their challenges in discriminatory ways, thereby unintendedly exacerbating the very
kinds of sentencing discrepancies judicial peremptory challenges are intended to avoid.170

Judicial political affiliation undoubtedly plays a role in driving criminal sentencing
outcomes. Empirical evidence of sentencing outcomes in both state and federal court shows that
Republican-appointed and Republican-affiliated judges sentence defendants to longer prison
sentences than Democratic-appointed and Democratic-affiliated judges.

This is a critically

important finding in a criminal justice system which grants wide discretion to judges in sentencing
determinations. While legislative actors have attempted to enact guardrails around sentencing
outcomes to achieve consistency and proportionality in sentencing, sentencing disparities persist.
This is at least in part due to the failure of legislative sentencing reform to account for individual
judicial political differences and the lack of creative statutory solutions – like expanded data
collection and judicial peremptory strikes – to reduce sentencing disparities between judges with
differing political leanings. Understanding that a primary driver of sentencing disparities is related
to political party affiliation – even in determinate sentencing schemes where judicial sentencing
discretion is highly guided - can aid policy makers in more effectively addressing these partisan


For a discussion of the ways in which the exercise of judicial peremptory challenges may
perpetuate racial and gender bias, see Nancy J. King, Batson for the Bench? Regulating the
Peremptory Challenge of Judges, 73 Chi. Kent. L. Rev. 509 (1998).

Electronic copy available at: