Racial Impact Statements as a
Means of Reducing Unwarranted Sentencing
Disparities
Marc Mauer *
The extreme racial disparities in rates of incarceration in the United States
result from a complex set of factors. Among these are sentencing and drug
policies which, intended or not, produce disproportionate racial/ethnic effects. In
retrospect, it is clear that many of these effects could have been predicted prior to
the adoption of the legislation. In order to reduce the scale of unwarranted
disparities, policymakers should address the potential racial impact of proposed
legislation prior to enactment, rather than after the fact when any necessary
reform is more difficult to achieve. One means of accomplishing this would be
through the establishment of “Racial Impact Statements.” Similar to fiscal or
environmental impact statements, such a policy would enable legislators and the
public to anticipate any unwarranted racial disparities and to consider alternative
policies that could accomplish the goals of the legislation without causing undue
racial effects.
I. INTRODUCTION
One of the hallmarks of the “get tough” movement over the past three decades
has been the relative lack of evaluation regarding both the potential and actual
effectiveness of harsh criminal justice sanctions in controlling crime. Typically,
when new punitive sentencing legislation is enacted there is little funding or
attention devoted to assessing its likely effects, both intended and unintended. 1 In
addition to limited evaluation of the effects of sentencing policy on crime, there is
an even greater gap in addressing concerns relating to the dramatic racial
disparities that pervade the criminal justice system. 2 This essay proposes that one
way this problem can be addressed is by adopting racial impact statements as a
requirement for consideration of new sentencing legislation.
*
Marc Mauer is the Executive Director of The Sentencing Project, the author of RACE TO
INCARCERATE (1999), and the co-editor of INVISIBLE PUNISHMENT: THE COLLATERAL CONSEQUENCES
OF MASS IMPRISONMENT (2002). The author wishes to thank the following for review and/or
discussions of the concepts involved in this essay: Paul Butler, Kara Gotsch, Susan Katznelson, Ryan
King, Jolanta Juszkiewicz, Barbara Tombs, and Michael Tonry.
1
Alfred Blumstein, Making Rationality Relevant—The American Society of Criminology
1992 Presidential Address, 31 CRIMINOLOGY 1, 1 (1993) (describing the desire of criminologists to
inform public policy being thwarted by “policies that ignore well-founded criminological
knowledge”).
2
MICHAEL TONRY, THINKING ABOUT CRIME 221 (2004).

19

20

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

These issues can be most readily understood by examining the adoption of
federal mandatory sentencing laws for crack cocaine offenses. Just over twenty
years ago, members of Congress gathered in meeting rooms of the House and
Senate to pass emergency legislation. The meetings were occasioned by
widespread concern over a new drug, crack cocaine, that was beginning to wreak
havoc in many communities. 3 Sensational tales of the drug’s effect on its users—
how quickly they became addicted, how it led to violent behavior—were rampant,
and led to a virtual national frenzy over how to cope with this new threat. 4
The problem, at least as it was perceived, was not a generalized one, but
rather a racial one, pertaining to young black men. From the cover of Newsweek,
to stories on the nightly news, to pronouncements of members of Congress, the
image of a dark-skinned inner-city youth was the face of this new and seemingly
intractable drug problem. 5 Already emotionally laden, the issue achieved highprofile status in the nation’s capital following the tragic death of (black) University
of Maryland basketball star Len Bias in June of 1986. Following the NBA draft, at
which he had been selected as the number two pick by the Boston Celtics, Bias
died of a drug overdose that evening while celebrating with friends. News reports
indicated that he had overdosed on crack cocaine, an error that was not repudiated
until months later when it was reported that he had in fact died from freebasing
powder cocaine. 6
Bias’s death ignited momentum in Congress to “do something” about this new
drug scourge. 7 The hastily crafted response called for harsh prison terms, and only
harsh prison terms. According to the United States Sentencing Commission, the
adoption of the 1986 Anti-Drug Abuse Act “was notable for the speed of its
development and enactment.” 8 The Act established mandatory prison terms of
five years for possession or sale of just five grams of crack cocaine, the weight of
about two sugar packets. The same five-year term was also enacted for sale of
powder cocaine, but at a threshold of 500 grams, or 100 times the quantity of crack
cocaine. 9
Crack cocaine is derived from powder cocaine and is produced by mixing
cocaine with baking soda and then cooking it. 10 Thus, the two drugs are
3
Craig Reinarman & Harry G. Levine, Crack in Context: America’s Latest Demon Drug, in
CRACK IN AMERICA: DEMON DRUGS AND SOCIAL JUSTICE 1–2 (Craig Reinarman & Harry G. Levine
eds., 1997).
4
Craig Reinarman & Harry G. Levine, The Crack Attack: Politics and Media in the Crack
Scare, in CRACK IN AMERICA: DEMON DRUGS AND SOCIAL JUSTICE, supra note 3, at 18.
5
Id. at 19.
6
U.S. SENTENCING COMMISSION, SPECIAL REPORT TO THE CONGRESS: COCAINE AND FEDERAL
SENTENCING POLICY, 122–23 (Feb. 1995).
7
Id. at 121–23.
8
Id. at 121.
9
Id. at 116.
10
Id. at v–vi.

2007]

21

RACIAL IMPACT STATEMENTS

pharmacologically identical, yet crack cocaine offenders are punished far more
severely than persons convicted of powder cocaine crimes. While the effects of
any form of cocaine on the body are similar, the intensity and duration of the
effects differ according to the means of administering the drug. Crack cocaine can
only be smoked, but powder cocaine can be snorted, injected, or ingested. 11
Powder cocaine that is snorted takes longer to achieve its maximum effects than
crack cocaine, but smoking or injecting powder cocaine produces similar times of
onset and effects as crack cocaine. 12 Because crack is generally marketed in small
doses, it was initially heavily distributed in many low-income minority
neighborhoods. 13 In the twenty years since enactment of the law, more than 80%
of crack cocaine sentences have been imposed on African Americans. 14
Two decades after the fact, there is a great deal of discussion regarding the
extreme racial dynamics of the crack cocaine mandatory sentencing laws. 15 The
focus of this article will be to propose that we discuss the racial dimensions of
public policy before new legislation is enacted, rather than after the fact. One
means of accomplishing this would be through the establishment of “Racial Impact
Statements,” which would obligate policymakers to review data on racial effects
prior to adopting new legislation. Such a policy would not prohibit legislators
from enacting new laws that might exacerbate existing disparities, but it would
serve to focus discussion on these effects as well as to encourage consideration of
alternative policies that could accomplish the goals of the legislation without
causing undue racial effects. Overall, the goal of such a policy would be to
eliminate unwarranted disparity in the criminal justice system while also
promoting public safety, goals that should be complementary.
In order to establish the case for such a policy, I will cover the following
areas:
1) Racial disparities in the criminal justice system are in part a function
of differentials in crime rates, but they also reflect disparities in
criminal justice processing and decision-making.
2) Criminal justice policy is often premised on racial assumptions,
whether consciously stated or not, that limit consideration of
alternative social policy interventions.

11
12
13

Id. at 14.
Id. at 21–22.
MICHAEL TONRY, MALIGN NEGLECT—RACE, CRIME,

AND

PUNISHMENT

IN

AMERICA 105

(1995).
14
U.S. SENTENCING COMMISSION, REPORT TO THE CONGRESS: COCAINE AND FEDERAL
SENTENCING POLICY 15–16 (2007), available at http://www.ussc.gov/r_congress/cocaine2007.pdf.
15
See, e.g., TESTIMONY FROM THE PUBLIC HEARING ON COCAINE SENTENCING POLICY HELD BY
THE
U.S.
SENTENCING
COMMISSION
(Nov.
2006),
available
at
http://www.ussc.gov/hearings/11_15_06/testimony.pdf.

22

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

3) Racial impact statements as a public policy hold the potential for
serving as a mechanism to reduce racial disparities while
contributing to a more effective public safety response.
II. OVERVIEW OF RACE AND INCARCERATION
While it now seems commonplace that there are extreme racial disparities in
the use of incarceration in the United States, in fact the extent of these disparities
has varied significantly over time and place. In his history of the development of
Southern prisons, David Oshinsky describes the evolving racial dynamics of
imprisonment. 16 Prior to the Civil War, prison populations in the South had been
comprised almost entirely of whites for the simple reason that African Americans
had been essentially imprisoned as slaves. But once blacks gained their freedom,
Southern prisons rapidly became transposed. 17 And under the vicious “convict
leasing” system, prison systems leased out the labor of inmates in ways that eerily
resembled the days of slavery. 18
Even in the early decades of the 20th century, the black rate of imprisonment
was not nearly at the level we see today. In tracing admissions to prison, for
example, blacks constituted 21% of admissions in 1926; by 1986 this had doubled
to 44%. 19 This is not to suggest that these earlier times were somehow a benign
period for race relations, since social mechanisms for repressing blacks were
clearly ubiquitous among varied institutions. But the broad-scale use of
incarceration, ostensibly as a means of crime control, had not yet become
prevalent. 20
By the early 21st century, the scale of incarceration for African Americans had
reached dramatic proportions. Projections by the Department of Justice show that
if current trends continue, a black male born today has a one in three (32.2%)
chance of spending time in state or federal prison in his lifetime. Comparable
figures for Latino males are one in six (17.2%) and, for white males, one in
seventeen (5.9%). 21 The overall figures for women are lower, but the racial
disparities parallel those for men.
These figures derive not only from the growing proportion of African
Americans among the prison population, but also from the unprecedented
16
DAVID M. OSHINSKY, WORSE THAN SLAVERY: PARCHMAN FARM AND THE ORDEAL OF JIM
CROW JUSTICE (1996).
17
Id. at 34.
18
Id. at 34–53.
19
PATRICK A. LANGAN, BUREAU OF JUSTICE STATISTICS, U.S. DEP’T OF JUSTICE, RACE OF
PRISONERS ADMITTED TO STATE AND FEDERAL INSTITUTIONS, 1926–86, at 5 (May 1991).
20
MARC MAUER, RACE TO INCARCERATE 17–19 (1999).
21
THOMAS P. BONCZAR, BUREAU OF JUSTICE STATISTICS, U.S. DEP’T OF JUSTICE, PREVALENCE
OF IMPRISONMENT IN THE U.S. POPULATION, 1974–2001, at 1 (2003), available at
http://www.ojp.usdoj.gov/bjs/pub/pdf/piusp01.pdf.

2007]

RACIAL IMPACT STATEMENTS

23

expansion of the prison system overall in the past three decades. 22 With
sentencing and incarceration policies firmly institutionalized in many respects,
reducing racial disparities will require a sustained proactive strategy of policy
reform.
III. ASSESSING THE RISE IN BLACK IMPRISONMENT
If we want to develop policies and practices to reduce the unprecedented rates
of imprisonment of African Americans, clearly we need to assess the causes of
those trends. This is a complex process, and this analysis will only attempt to
provide an overview of those dynamics, with a view toward assessing their
implications for criminal justice policy alternatives. The key issues to be
examined are:
• Disproportionate crime rates
• Disparities in criminal justice processing
• Overlap of race and class effects
• Impact of “race neutral” policies
A. Disproportionate Crime Rates
Many people would assume that if African Americans are incarcerated at
higher rates than other racial/ethnic groups, this merely reflects greater
involvement in crime, and is a natural, if perhaps unfortunate, consequence of that
involvement. In fact, crime rates explain some, but only some, of these disparities.
The most sophisticated national estimates of the racial dynamics of
incarceration stem from two national studies conducted by criminologist Alfred
Blumstein. Comparing imprisonment and arrest rates by offense initially for the
1979 prison population and subsequently for the 1991 population, Blumstein
concluded that higher rates of involvement in crime (as estimated by arrest rates)
explained 80% of racial disparity in the first study and 76% in the second.23
According to Blumstein, the unexplained disparity might be a function of a number
of factors, including racially biased decision-making in the court system, but also
sentencing-relevant factors such as prior criminal histories of offenders. Notably,
though, the decline in the crime-involvement explanation for the disparity from
1979 to 1991 was primarily a function of the increasing incarceration of drug
offenders. 24 In this regard, Blumstein concluded that due to the racially skewed

22

MAUER, supra note 20, at 19.
Alfred Blumstein, On the Racial Disproportionality of United States’ Prison Populations,
73 J. CRIM. L. & CRIMINOLOGY 1259, 1280 (1982); Alfred Blumstein, Racial Disproportionality of
U.S. Prison Populations Revisited, 64 U. COLO. L. REV. 743, 751 (1993) [hereinafter Blumstein,
Prison Populations Revisited].
24
Blumstein, Prison Populations Revisited, supra note 23, at 750–754.
23

24

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

nature of drug law enforcement, only 50% of the racial disproportionality for drug
incarceration was explained by differential arrest rates. 25
While Blumstein’s findings are notable, their national scope may obscure
variation among the states. In fact, an assessment of this relationship at the state
level finds broad variation in the extent to which higher crime rates among African
Americans may explain disproportionate imprisonment. Robert Crutchfield and
colleagues’ examination of this relationship concluded that at the state level there
is a broad variation in the degree to which criminal involvement explains
incarceration trends. 26
As will be discussed later, to the extent that crime rates among African
Americans explain some portion of incarceration rates, we should note that this is
not necessarily a racial effect, but rather a function of the overlap of race and class.
Since African American communities are more likely to be low-income, as well as
more likely to experience the disadvantages resulting from concentrated poverty
than low-income whites, 27 what may appear to be a race effect is often one of
social class.
B. Disparities in Criminal Justice Processing
While crime rates explain some degree of variation by race, it is also clear that
racism within the criminal justice system, whether conscious or not, contributes to
disproportionate incarceration. This is not to suggest that the demonstrable
changes within the criminal justice system in recent decades have not been of
significance. Indeed, the greater diversity of leadership within the system, the
growing attention to issues of racial disparity, and changes in political discourse on
race are all welcome developments. Nevertheless, a wealth of scholarship and
experience documents that unwarranted racial disparities persist at every level of
criminal justice decision-making, albeit sometimes in more subtle ways than in
previous eras. 28
In the area of law enforcement, evidence of widespread racial profiling has
been well documented in recent years. 29 In states such as New Jersey and Florida,
state troopers and local law enforcement personnel have stopped African American
drivers for alleged traffic infractions at far higher rates than white drivers. 30 At the
25

Id. at 751.
Robert D. Crutchfield, George S. Bridges, & Susan R. Pitchford, Analytical and
Aggregation Biases in Analyses of Imprisonment: Reconciling Discrepancies in Studies of Racial
Disparity, 31 J. RES. CRIME & DELINQ. 166, 179 (1994).
27
DOUGLAS S. MASSEY & NANCY A. DENTON, AMERICAN APARTHEID: SEGREGATION AND THE
MAKING OF THE UNDERCLASS 1–14 (1993).
28
LEADERSHIP CONFERENCE ON CIVIL RIGHTS & LEADERSHIP CONFERENCE EDUCATION FUND,
JUSTICE ON TRIAL: RACIAL DISPARITIES IN THE AMERICAN CRIMINAL JUSTICE SYSTEM (2000).
29
See, e.g., DAVID A. HARRIS, PROFILES IN INJUSTICE: WHY RACIAL PROFILING CANNOT WORK
(2002).
30
Id. at 21–23, 53–60.
26

2007]

RACIAL IMPACT STATEMENTS

25

national level, research by the Department of Justice reveals that while black
drivers may be stopped at similar rates to white drivers, they are three times as
likely to be subject to a search after being stopped. 31 Since racial profiling has
been intimately tied to the war on drugs, 32 more frequent searches of black drivers
will inevitably turn up higher rates of drug infractions, just as would be the case if
white drivers were searched more frequently.
Racial profiling has taken place not only on highways, but in black
neighborhoods as well. In 1990, the Boston Police Department was cited for
unconstitutional stops and searches by the Civil Rights Division of the
Massachusetts Attorney General’s office. 33 The action came after the police
agency had carried out a large indiscriminate policy of searching on sight “known
gang members” in the black neighborhood of Roxbury. 34
At a policy level, the inception of the modern day “war on drugs” in the 1980s
inaugurated a wholesale shift in the allocation of law enforcement resources. 35
Over the past two decades, this initiative has been the most significant factor
contributing to the disproportionate incarceration of African Americans in prisons
and jails, with increasing effects on Latinos as well. 36 This has come about
through two overlapping trends. First, the escalation of the drug war has produced
a remarkable rise in the number of persons serving sentences or awaiting trial for
drug offenses in prisons and jails from about 40,000 in 1980 to nearly a half
million by 2005. 37 This represents the combined impact of the tripling of drug
arrests from 580,900 in 1980 to 1.8 million in 2005, 38 along with the adoption of
mandatory minimum sentencing at the federal and state levels that has produced
both more and longer drug offense prison terms. 39

31

MATTHEW R. DUROSE, ERICA L. SCHMITT, & PATRICK A. LANGAN, BUREAU OF JUSTICE
STATISTICS, U.S. DEP’T OF JUSTICE, CONTACTS BETWEEN POLICE AND THE PUBLIC: FINDINGS FROM THE
2002 NATIONAL SURVEY 4, 8 (2005), available at http://www.ojp.usdoj.gov/bjs/pub/pdf/cpp02.pdf.
32
See DAVID COLE, NO EQUAL JUSTICE: RACE AND CLASS IN THE AMERICAN CRIMINAL JUSTICE
SYSTEM 47 (1999) (noting that “One of the most common rationales for stopping and questioning
citizens is the ‘drug-courier’ profile”).
33
Id. at 25–26.
34
Id.
35
Bruce L. Benson & David W. Rasmussen, Relationship Between Illicit Drug Enforcement
Policy and Property Crimes, 9 CONTEMP. POL’Y ISSUES 106, 106–07 (1991).
36
RYAN S. KING & MARC MAUER, THE SENTENCING PROJECT, DISTORTED PRIORITIES: DRUG
OFFENDERS
IN
STATE
PRISONS
14
(2002),
available
at
http://www.sentencingproject.org/Admin/Documents/publications/dp_distortedpriorities.pdf.
37
Marc
Mauer,
Incarceration
Nation,
TOMPAINE.COM,
Dec.
11,
2006,
http://www.tompaine.com/articles/2006/12/11/incarceration_nation.php.
38
TINA L. DORSEY, MARIANNE W. ZAWITZ, & PRISCILLA MIDDLETON, BUREAU OF JUSTICE
STATISTICS, U.S. DEP’T OF JUSTICE, DRUGS AND CRIME FACTS 60 (2006), available at
http://www.ojp.usdoj.gov/bjs/pub/pdf/dcf.pdf.
39
TONRY, supra note 13, at 92–93.

26

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

The escalation of drug prosecutions has coincided with a large-scale law
enforcement emphasis on drug policing in communities of color. Nationally,
African Americans represented 33.9% of drug arrests in 2005, 40 considerably
higher than their 14% share of current drug users. 41 Among drug offenders
sentenced to prison, 53% are African American. 42 While this figure arguably may
incorporate factors such as higher levels of involvement in the drug trade or a more
substantial criminal history, scholars such as Michael Tonry have concluded that
“[a]lthough disadvantaged young people of all races and ethnicities have been
affected by the drug wars, the greatest attention has been on Hispanics and
blacks,” 43 and, as a result, the war on drugs “forseeably and unnecessarily blighted
the lives of hundreds of thousands of young disadvantaged black Americans and
undermined decades of effort to improve the life chances of members of the urban
black underclass.” 44
Unwarranted disparities continue at the level of prosecution, although this
area is probably the least analyzed among all criminal justice stages. Nevertheless,
research evidence suggests that decision-making at the charging and plea
negotiation stages may contribute to racial disparity as well. 45 An assessment of
plea bargaining practices in 700,000 criminal cases published by the San Jose
Mercury News in 1991, for example, concluded that “[a]t virtually every stage of
pretrial negotiation, whites are more successful than non-whites.” 46 Of 71,000
adults charged with felonies and with no prior record in the study, one third of
whites had charges reduced to misdemeanors or infractions, while only one quarter
of blacks and Hispanics received these outcomes. 47
In the area of sentencing, the most dramatic findings exist in application of
the death penalty. Studies by David Baldus and others dating back to the 1980s
demonstrate that race of both victim and offender are key components of the
difference in sentencing outcomes between the death penalty and life

40

FEDERAL BUREAU OF INVESTIGATION, U.S. DEP’T OF JUSTICE, CRIME IN THE UNITED STATES,
2005 tbl.43 (2006), available at http://www.fbi.gov/ucr/05cius/data/table_43.html.
41
Data calculated from the OFFICE OF APPLIED STUDIES, SUBSTANCE ABUSE AND MENTAL
HEALTH SERVICES ADMINISTRATION, DEP’T OF HEALTH AND HUMAN SERVICES, RESULTS FROM THE
2005 NATIONAL SURVEY ON DRUG USE AND HEALTH: NATIONAL FINDINGS 234 tbl.1.28B (2006),
available at http://www.drugabusestatistics.samhsa.gov/nsduh/2k5nsduh/2k5results.pdf.
42
PAIGE M. HARRISON & ALLEN J. BECK, BUREAU OF JUSTICE STATISTICS, U.S. DEP’T OF
JUSTICE,
PRISONERS
IN
2005,
at
9
tbl.12
(2006),
available
at
http://www.ojp.usdoj.gov/bjs/pub/pdf/p05.pdf.
43
TONRY, supra note 13, at 82.
44
Id.
45
See, e.g., CORAMAE RICHEY MANN, UNEQUAL JUSTICE: A QUESTION OF COLOR (1993).
46
Christopher H. Schmitt, Plea Bargaining Favors Whites as Blacks, Hispanics Pay Price,
SAN JOSE MERCURY NEWS, Dec. 8, 1991, at 1A.
47
Id.

2007]

RACIAL IMPACT STATEMENTS

27

imprisonment. 48 In the realm of non-capital sentencing, the research is more
complex, but a survey conducted for the National Institute of Justice found that
“race and ethnicity do play an important role in contemporary sentencing
decisions. Black and Hispanic offenders—and particularly those who are young,
male, or unemployed—are more likely than their counterparts to be sentenced to
prison.” 49 In recent decades, it is not so much race alone but rather race in
combination with other factors, such as gender and employment, that translate into
unwarranted racial disparities at the sentencing stage. 50
Thus, despite progress in creating a more diverse criminal justice system,
unwarranted racial disparities still persist and contribute to the overrepresentation
of African Americans in the prison system. While there is far less comprehensive
data available on Latinos and other minority groups, these dynamics may carry
over to these communities as well.
C. Overlap Between Race and Class Effects
It has been suggested by some scholars that the disparities we see in the
criminal justice system reflect relative disadvantages associated with social class
rather than race. Indeed, the degree of access to resources does play a significant
role in determining outcomes in the court system.
This can be seen most prominently in regard to the quality of defense counsel.
While many public defenders and appointed counsel provide high quality legal
support, in far too many jurisdictions the defense bar is characterized by high
caseloads, poor training, and inadequate resources. 51 In an assessment of this
situation, the American Bar Association concluded that “[t]oo often the lawyers
who provide defense services are inexperienced, fail to maintain adequate client
contact, and furnish services that are simply not competent.” 52 Horror stories of
indigent defense representation abound, combined with shameless levels of public
support. In Virginia, for example, counsel representing persons charged with
felonies that can result in a life sentence can receive no more than $1,096 in fees
for the case. 53
48
David C. Baldus, Charles Pulaski, & George Woodworth, Comparative Review of Death
Sentences: An Empirical Study of the Georgia Experience, 74 J. CRIM. L. & CRIMINOLOGY 661, 707
(1983); Jennifer L. Eberhardt et al., Looking Deathworthy: Perceived Stereotypicality of Black
Defendants Predicts Capital-Sentencing Outcomes, 17 PSYCHOL. SCI. 383, 384 (2006).
49
Cassia C. Spohn, Policies, Processes, and Decisions of the Criminal Justice System, Thirty
Years of Sentencing Reform: The Quest for a Racially Neutral Sentencing Process, 3 CRIM. JUST. 427,
428 (2000), available at http://www.ncjrs.gov/criminal_justice2000/vol_3/03i.pdf.
50
Id. at 461.
51
A.B.A. STANDING COMM. ON LEGAL AID AND INDIGENT DEFENDANTS, GIDEON’S BROKEN
PROMISE: AMERICA’S CONTINUING QUEST FOR RACIAL JUSTICE (2004), available at
http://www.abanet.org/legalservices/sclaid/defender/brokenpromise/fullreport.pdf.
52
Id. at iv.
53
Editorial, Justice Denied in Virginia, WASH. POST, Feb. 10, 2004, at A22.

28

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

Access to private resources also plays a role in the quality of justice one
receives at various points in the criminal justice process. In regard to pretrial
release, for example, owning a telephone is one factor used in making
recommendations to release a defendant prior to trial. 54 But if low-income
defendants are less likely to have a telephone in the home, this seemingly
innocuous requirement becomes an obstacle to pretrial release.
While these dynamics illustrate the interplay among race, class, and criminal
justice processing, they do not necessarily lead to the conclusion that social class
has replaced race as the determining factor in producing disparities. As we have
seen in the case of racial profiling by law enforcement personnel, motorists are not
stopped because they are low-income African Americans but because they are
African American. 55 A black banker wearing blue jeans while driving on the
weekend is no more immune to being stopped than is an unemployed wage laborer.
Secondly, black poverty is different than poverty in general. African
Americans who are poor are far more likely than whites to live in communities of
concentrated poverty and disadvantage. Because of the confluence of racism and
segregated housing patterns that persist to this day, low-income blacks are far more
likely to live in neighborhoods that are segregated both by race and class. 56 As a
result, the disadvantages that accrue from poverty are magnified as well. These
include limits on job opportunities and social linkages that develop through
informal networks among residents of communities with access to resources. 57
D. Impact of “Race Neutral” Policies
Sentencing and related criminal justice policies that are ostensibly “race
neutral” have in fact been seen over many years to have clear racial effects that
could have been anticipated by legislators prior to enactment. In some instances,
the purported race neutrality has merely been a thin cover for racial intent. In the
area of felony disenfranchisement, for example, following the Reconstruction
period, some Southern states tailored their statutes with the specific intent of
restricting the voting rights of African Americans. 58 The means by which this was
accomplished, not coincidentally during the same period in which poll taxes and
54

BRIAN A. REAVES & JACOB PEREZ, BUREAU OF JUSTICE STATISTICS, U.S. DEP’T OF JUSTICE,
PRETRIAL RELEASE OF FELONY DEFENDANTS, 1992, at 3 (1994), available at
http://www.ojp.gov/bjs/pub/pdf/nprp92.pdf (reporting on the results of a national survey showing that
“[a]bout two-thirds of conditional releases included an agreement by the defendant to maintain
regular contact with a pretrial program through telephone calls and/or personal visits”).
55
COLE, supra note 32, at 36 (noting that “[m]any black professionals tell of being stopped
simply because they were black and driving a fancy car”).
56
MASSEY & DENTON, supra note 27 at 1–14.
57
Id.
58
Marc Mauer, Mass Imprisonment and the Disappearing Voters, in INVISIBLE PUNISHMENT:
THE COLLATERAL CONSEQUENCES OF MASS IMPRISONMENT 50, 51–52 (Marc Mauer & Meda
Chesney-Lind eds., 2002).

2007]

RACIAL IMPACT STATEMENTS

29

literacy requirements were being adopted, was to disenfranchise persons convicted
of crimes believed to be committed disproportionately by blacks, while not taking
away the right to vote for persons convicted of offenses believed to be committed
by whites. 59 As a result, an Alabama man convicted of beating his wife would lose
the right to vote, while a man convicted of killing his wife would not. 60 Such was
the racial logic of the day. Lest we think that these laws were just a relic of 19th
century Southern racism, we should note that these policies were in place for
nearly one-hundred years until finally being struck down by the U.S. Supreme
Court in 1985. 61
Recent research on the development of punitive sentencing policies sheds
light on the relationship between harsh sanctions and public perceptions of race. In
a study of the relationship between public attitudes and sentencing policies, Ted
Chiricos and colleagues found that among whites, support for harsh sentencing
policies was correlated with the degree to which a particular crime was perceived
to be a “black” crime. 62
A variety of sentencing policies now have unmistakable racial impacts despite
being race neutral on the surface.
1. Crack Cocaine Sentencing Laws
The federal mandatory sentencing laws enacted for crack cocaine offenses in
1986 and 1988 do not, of course, directly target African Americans for severe
punishment. But as has been well documented, this has clearly been the outcome
over time, with more than 80% of crack defendants being African American—well
above their proportion of users of the drug—and subject to harsh prison terms. 63
2. School Zone Drug Laws
A number of states and the federal government have adopted policies that
penalize drug offenses that take place near a school zone more harshly than other
drug crimes. 64 Under the goal of protecting children from the drug trade, these
laws have also had severely disproportionate racial impacts, due to residential

59

Id.
Id. at 52.
61
Hunter v. Underwood, 471 U.S. 222 (1985).
62
Ted Chiricos, Kelly Welch, & Marc Gertz, Racial Typification of Crime and Support for
Punitive Measures, 42 CRIMINOLOGY 359, 374 (2004).
63
U.S. SENTENCING COMMISSION, supra note 14, at 15–16.
64
For an overview of these laws, see JUDITH GREENE, KEVIN PRANIS & JASON ZIEDENBERG,
JUSTICE
POLICY
INSTITUTE,
DISPARITY
BY
DESIGN
5
(2006),
available
at
http://www.justicepolicy.org/images/upload/06-03_REP_DisparitybyDesign_DP-JJ-RD.pdf.
60

30

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

housing patterns. 65 Typically, such laws target drug transactions that take place
within 1,000 or 1,500 feet of a school zone. 66 States such as New Jersey have
gone further, passing legislation that provides enhanced penalties for drug
transactions near public housing, libraries, and museums as well. 67 Since urban
areas are more densely populated than suburban or rural areas, city residents are
much more likely to be within a short distance of a school or public housing site
during much of the day. And since African Americans are more likely to live in
urban neighborhoods than are whites, African Americans convicted of a drug
offense are subject to harsher punishments than whites committing the same
offense in a less populated area. 68 A state commission analysis of the New Jersey
law documented that nearly all (96%) of the persons serving prison time for drug
free zone offenses were African American or Hispanic. 69
3. Habitual Offender Laws
Sentencing policies, such as habitual offender and “three strikes” laws that
penalize repeat offenders more harshly, also produce racially disparate effects even
though not expressly conceived to do so. These policies are premised on the idea
that persons with prior criminal histories shall receive harsher punishments. In
itself, this is not necessarily objectionable since many sentencing frameworks and
philosophical approaches endorse such a system. But the current generation of
harsher sentencing policies based on prior record will clearly have a racially
disproportionate effect, and one that is likely to be far more severe than in previous
times. This racial impact is due to the fact that African Americans are more likely
to have a criminal record than other groups. 70 Some believe that this is due to
greater involvement in criminality, 71 others that it is due to greater scrutiny and
racist behavior by actors in the justice system. 72 Regardless of the cause, though,
an African-American defendant being sentenced is more likely than a white
defendant to have a substantial criminal record, and therefore receive a harsher
punishment.
65

Id. at 14 (showing that ninety-six percent of persons serving terms for school zone drug law
violations in New Jersey are African American or Hispanic).
66
Id. at 1, 5.
67
Id. at 22.
68
Id. at 4.
69
THE NEW JERSEY COMMISSION TO REVIEW CRIMINAL SENTENCING, REPORT ON NEW
JERSEY’S DRUG FREE ZONE CRIMES AND PROPOSAL FOR REFORM 23 (2005), available at
http://sentencing.nj.gov/dfz_report_pdf.html.
70
As of 2001, 16.6% of African American males had served time in state or federal prison,
compared to 4.9% for white males and 7.7% for Hispanic males. Similar disparities exist for women
as well. BONCZAR, supra note 21, at 1.
71
John J. DiIulio, My Black Problem, and Ours, CITY JOURNAL, Spring 1996, at 14.
72
HARRIS, supra note 29, at 21.

2007]

RACIAL IMPACT STATEMENTS

31

While the impact of a criminal record has affected sentencing practice over
time, with policies such as “three strikes” these effects are magnified substantially.
In California, African Americans represent 31.3% of the inmate population but
44% of persons serving three strikes sentences. 73 Persons convicted of such
offenses in California prisons include a man serving a term of twenty-five years to
life for stealing golf clubs and another serving a term of fifty years to life for
stealing videotapes. 74 Thus, the “race neutral” penalty is increasingly prominent
over time as sentencing polices become more severe.
Similar effects may be seen in regard to changes in parole policies as well. In
recent years, many states have adopted more restrictive parole release policies,
particularly for long-term and life-sentenced prisoners. Whereas life sentences in
most states previously permitted the possibility of parole release, states such as
Pennsylvania, Louisiana, and Michigan have now adopted “life means life”
policies, essentially eliminating this possibility. 75 In this area as well, while there
is no stated racial intent in the new policies themselves, there will almost
inevitably be a racial outcome. This comes about because African Americans are
disproportionately incarcerated for violent offenses, 76 and thus will comprise a
disproportionate degree of lifers. As will be discussed later, one may or may not
think that such an outcome is problematic when talking about violent offenses, but
it is nonetheless a contributor to racial disparity.
IV. RACIAL IMPACT STATEMENTS AS A MEANS OF CONTROLLING DISPARITY
A. Impact Statements
As we have seen, to at least some extent unwarranted racial disparities in the
criminal justice system result from policy initiatives of recent years. In regard to
some of these initiatives, such as mandatory sentencing laws pertaining to crack
cocaine or school zone drug laws, the racial impact could easily have been
foreseen had policymakers undertaken an analysis prior to adoption of the law.77
Because of this potential for addressing disparities in a proactive way, the adoption
of racial impact statements offers a means by which policymakers can avoid some

73
RYAN S. KING & MARC MAUER, THE SENTENCING PROJECT, AGING BEHIND BARS: THREE
STRIKES
SEVEN
YEARS
LATER
13
(2001),
available
at
http://www.sentencingproject.org/Admin/Documents/publications/inc_aging.pdf.
74
Erwin Chemerinsky, Life in Prison for Shoplifting: Cruel and Unusual Punishment, A.B.A.
HUM. RTS. MAG., Winter 2004, available at http://www.abanet.org/irr/hr/winter04/shoplifting.html.
75
MARC MAUER, RYAN S. KING & MALCOLM C. YOUNG, THE SENTENCING PROJECT, THE
MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT 6–7 (2004), available at
http://www.sentencingproject.org/Admin/Documents/publications/inc_meaningoflife.pdf.
76
In 2003, African Americans constituted 562,100 (44.7%) of the 1,256,400 persons
incarcerated for a violent offense in state prisons. HARRISON & BECK, supra note 42, at 9 tbl.12.
77
TONRY, supra note 13, at 104–15.

32

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

of the mistakes of the past and develop crime policy that is both constructive and
fair.
Before discussing a proposed framework for the use of racial impact
statements, it is useful to conduct a brief review of the ways in which impact
statements have become incorporated as a standard means of assessing both the
intended and unintended effects in other areas of public policy. Environmental
impact statements, for example, have been required by law since the adoption of
the National Environmental Policy Act (NEPA) of 1969. Under NEPA, federal
agencies are required to prepare an environmental impact statement for “proposals
for legislation and other major Federal actions significantly affecting the quality of
the human environment.” 78 Not only must the statement include an assessment of
the environmental impact and any adverse impacts which may ensue, but it must
also examine alternatives to the proposed action. 79 The Act has become a standard
component of federal environmental policy; in 2005, the U.S. Forest Service alone
prepared 153 impact statements, with many others prepared by the Army Corps of
Engineers, Federal Highway Administration, National Park Service, and other
agencies. 80
Other types of pro-active assessments are now routine in public policy
consideration as well. For example, fiscal cost estimates are prepared by the
Congressional Budget Office (CBO) for every bill reported by committee. The
CBO assessment of HR 5, the College Student Relief Act of 2007, for example,
concluded that the bill’s provisions regarding interest rates and fees on lenders
would reduce direct spending by $65 million over the 2007–2012 period and by
$7.1 billion over the 2007–2017 period. 81
In other nations, Health Impact Assessments are common as well, focusing on
the possible health effects of new initiatives and related costs. For example, the
British National Health Service describes the goal of these assessments as
considering “potential health impacts before a policy is implemented—and thus
making adjustments that will maximize the beneficial effects and minimize any
harmful effects on health.” 82
There are also areas in which data analysis is used to assess the effects of
criminal justice policy retrospectively that are instructive. Due to increasing
concern over racial profiling by law enforcement agencies in the 1990s, a number
of jurisdictions have adopted practices of compiling data on traffic stops and other
78

42 U.S.C. § 4332(c) (2007).
Id.
80
Council on Environmental Quality, Calendar Year 2005 Filed EISs—552 Total,
http://ceq.eh.doe.gov/nepa/EIS/Filed_%20EIS_2005_by_Agency.pdf (last visited Sept. 10, 2007).
81
College Student Relief Act, H.R. 5, 110th Cong. CONGRESSIONAL BUDGET OFFICE,
http://www.cbo.gov/showdoc.cfm?index=7729&sequence=0 (last visited Sept. 10, 2007).
82
NHS EXECUTIVE LONDON, A SHORT GUIDE TO HEALTH IMPACT ASSESSMENT: INFORMING
HEALTHY
DECISIONS
5
(2000),
available
at
http://www.nice.org.uk/media/hiadocs/Copy_of_Londons_SHORT_GUIDE_TO_HIA%20.pdf.
79

2007]

RACIAL IMPACT STATEMENTS

33

police encounters by race in an effort to monitor inappropriate actions. 83 And in
the area of the death penalty, the state of Kentucky adopted the Racial Justice Act
in 1998, which permits defendants to challenge their prosecution on a capital
charge if it can be demonstrated that race played a role in seeking the death
penalty. 84 Racial analyses, therefore, are not new concepts for assessment within
the justice system, but offer the possibility of deliberation on potential racial
effects prior to, not after, the implementation of policy initiatives.
B. Goals and Scope of Racial Impact Statements
The public policy goal of requiring racial impact statements is quite direct: to
encourage lawmakers to examine the racial effects of changes in sentencing and
related policy that affect prison populations, and when necessary, to consider
alternative means of achieving public safety goals without exacerbating
unwarranted racial disparities.
The use of racial impact statements by policymakers should be guided by two
principles: reducing unnecessary racial disparities in the use of incarceration and
promoting public safety. While these goals will appear unobjectionable in
themselves to most people, their synergistic relationship may be less obvious to
some. Specifically, why will reducing unwarranted racial disparities—a worthy
goal in itself—also contribute to public safety? Two factors stand out in particular.
First, law enforcement and sentencing policies that exacerbate unwarranted
racial disparities are generally also ineffective in contributing to public safety
goals. The evidence on racial profiling is now clear that not only are such
practices racist and unfair, but they do not produce good results for crime
control. 85 Law enforcement agents produce better results when they rely on
investigative techniques that yield information about specific individuals engaged
in illegal activity. As David Harris notes, “They’re [police] focusing on
appearance when they should be focusing on behavior. When they’re not
distracted by race, they’re actually doing a more accurate job” of picking out the
right people. 86
Similarly, the federal crack cocaine penalties have resulted in thousands of
prosecutions of lower-level African-American defendants, yet have produced no
demonstrable effects on substance abuse or sales. As noted by the United States
83
See, e.g., Racial Profiling Data Collection Resource Center at Northeastern University,
Background
and
Current
Data
Collection
Efforts,
http://www.racialprofilinganalysis.neu.edu/background/jurisdictions.php (last visited Sept. 10, 2007).
84
Racial Justice Act, KY. REV. STAT. ANN. §532.300 (West 2007), available at
http://www.deathpenaltyinfo.org/article.php?scid=5&did=255.
85
HARRIS, supra note 29, at 12–14.
86
Kim Zetter, Why Racial Profiling Doesn’t Work, SALON, Aug. 22, 2005, available at
http://dir.salon.com/story/news/feature/2005/08/22/racial_profiling/index.html (quoting HARRIS,
supra note 29).

34

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

Sentencing Commission, an analysis of cocaine price trends “appears inconsistent
with a finding that federal cocaine penalties established under the 1986 Act and
incorporated into the guidelines have had a deterrent effect on cocaine trafficking,”
and that “it is unlikely that the federal cocaine sentencing policy had a significant
deterrent effect on users.” 87
Second, growing racial disparities contribute to a lack of confidence in the
criminal justice system in many African American communities.
Law
enforcement and prosecution can only be effective when they work in concert with
communities, and so building trust is a key ingredient in producing overall
safety. 88 Judge Reggie Walton, a former high ranking official in the Office of
National Drug Control Policy, describes the perceived injustice of the crack
cocaine laws as having a “coercive impact on the respect many of our citizens have
about the general fairness [of] our nation’s criminal justice system,” and that
some people desire not to serve on juries when crack cocaine is involved
because of the negative attitudes they have about the crack and powder
cocaine sentencing disparity or have refused to convict crack offenders,
despite the quality of the government’s evidence, because of their
attitudes about the current sentencing structure. 89
As applied to policy changes that would directly affect the number of people
in prison, racial impact statements could be applied to the following:
• Sentencing statutory changes
• Sentencing guidelines adjustments
• Legislation creating new substantive crimes
• “Truth in sentencing” policies
• Parole release policies
• Parole revocation policies
• “Early” release policies, such as participation in drug treatment or
other programming
C. Preparation of Racial Impact Statements
The government agency charged with preparing racial impact statements will
vary by jurisdiction, but in virtually all states, as well as in the federal system,
mechanisms already exist to incorporate this function. Essentially, the agency
charged with analyzing data and estimating prison capacity needs will generally be
87

REPORT TO THE CONGRESS: COCAINE AND FEDERAL SENTENCING POLICY, U.S. SENTENCING
COMMISSION 72 (2002), available at http://www.ussc.gov/r_congress/02crack/2002crackrpt.pdf.
88
TESTIMONY OF REGGIE B. WALTON, THE UNITED STATES SENTENCING COMMISSION ON
SENTENCING DISPARITY FOR CRACK AND POWDER COCAINE OFFENSES 7–8 (Nov. 14, 2006), available
at http://www.ussc.gov/hearings/11_15_06/JudgeWalton-testimony.pdf.
89
Id.

2007]

RACIAL IMPACT STATEMENTS

35

the most appropriate one to perform this analysis. These agencies would include
the following:
1. Sentencing Commissions
In the federal system, sentencing policy development and oversight falls
under the jurisdiction of the United States Sentencing Commission. 90 At the state
level, twenty-one states and the District of Columbia have a sentencing
commission as well. 91 These agencies maintain sophisticated databases of
sentencing data, trends, and policy impacts, and generally incorporate relatively
complete data on race, gender, and offense demographics. 92 Since prison
populations are essentially a function of the number of people sentenced to prison
and their length of stay, projecting impacts of particular policy changes by race
will usually fall well within the parameters of these systems’ capabilities.
Some states, including North Carolina and Virginia, maintain legislative
requirements that their sentencing commissions produce impact statements for
legislation that may affect the size of the prison population. 93 The North Carolina
Sentencing and Policy Advisory Commission offers a model of how such a process
might work. 94 By law, the Commission is mandated to produce a fiscal impact
assessment of any bill with possible court or correctional impact on
appropriations. 95 The fiscal impact is related to an assessment of changes to time
served in prison. Thus, this could include sentencing enhancements, parole release
policies, or programmatic diversions from prison. 96 Additional costs, such as court
or community corrections costs, are added to the final impact statement by other
state agencies. 97
The North Carolina commission generally produces a report at the request of
the legislature’s fiscal research staff when a proposed bill has been sent to that
office.
The Commission employs a sophisticated simulation model that
90
United States Sentencing Commission, An Overview of the United States Sentencing
Commission, http://www.ussc.gov/general/USSCoverview_2005.pdf (last visited Sept. 10, 2007).
91
National
Association
of
State
Sentencing
Commissions,
Contact
List,
http://www.ussc.gov/states/nascaddr.htm (last visited Sept. 10, 2007).
92
NEAL B. KAUDER ET AL., STATE SENTENCING POLICY AND PRACTICE RESEARCH IN ACTION
PARTNERSHIP,
SENTENCING
COMMISSION
PROFILES
(1997),
available
at
http://www.ncsconline.org/WC/Publications/KIS_SentenSenCommProfiles.pdf.
93
N.C. GEN. STAT. § 164-43 (2007); VA. CODE ANN. § 30-19.1:4 (2007).
94
Email from Susan Katznelson, Executive Director, North Carolina Sentencing and Policy
Advisory Commission, to Marc Mauer, Executive Director, The Sentencing Project (Nov. 3, 2006,
11:46 a.m. EST) (discussing North Carolina approach) (on file with author); Telephone Interview
with Susan Katznelson, Executive Director, North Carolina Sentencing and Policy Advisory
Commission (Nov. 16, 2006) (discussing North Carolina approach) (on file with author).
95
Id.
96
Id.
97
Id.

36

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

incorporates case-based data on convictions and sentencing to produce a fiscal
impact statement that projects out for ten years. 98
2. Departments of Correction
Both state and federal corrections agencies track current and projected prison
populations on an ongoing basis, and generally have a range of available data
broken down by race and offense. 99 These projections are typically used to
forecast budget and prison space needs and, given the necessary data, could be
applied to produce racial impact statements as well.
3. Budget and Fiscal Agencies
In jurisdictions where either a sentencing commission or corrections agency
may not be positioned to conduct the impact analysis, other governmental agencies
may have the capacity to do so. Fiscal staff of legislative bodies regularly produce
various types of analyses for legislative initiatives, and with proper guidance and
support could be delegated to produce a racial impact statement as well.
D. Components of Racial Impact Statements
Policy initiatives establishing a process for producing racial impact statements
could use the following as a guide for the parameters of the statements: racial
impact statements shall be prepared for sentencing or corrections legislation or
policy initiatives that may affect the number of the incarcerated population. As
noted above, the process would be triggered primarily by proposed sentencing
legislation or policy changes that would affect time served in prison, including
parole release policies.
Let us consider two instances of legislative proposals that would be subject to
a racial impact statement in order to focus on the issues involved in evaluating
“disparity.” The first case would be the classic one of the crack cocaine sentencing
laws. Had Congress required that an impact statement be produced, it would have
demonstrated that an estimated 4000 defendants a year would be sentenced to five
and ten-year mandatory prison terms, 100 80% of whom would have been African
American. 101 A modest amount of additional data from government agencies
98

Id.
See, e.g., PATRICIA A. BIGGS ET AL., KANSAS DEPARTMENT OF CORRECTIONS, STATISTICAL
PROFILE:
FY
2002
OFFENDER
POPULATION
(2002),
available
at
http://www.dc.state.ks.us/research/StatProfile/Stat_Profile_2002.pdf.
100
During Fiscal Year 2003, for example, 4170 defendants convicted of a crack cocaine
offense received either a five-year or ten-year mandatory minimum sentence. See UNITED STATES
SENTENCING COMMISSION, 2003 SOURCEBOOK OF FEDERAL SENTENCING STATISTICS 88 tbl.43
(2003), available at http://www.ussc.gov/ANNRPT/2003/table43.pdf.
101
Id. at 79 tbl.34.
99

2007]

37

RACIAL IMPACT STATEMENTS

would have documented that these rates were far higher than the black proportion
of crack users or sellers in the general population. The question for policymakers
would then have been whether the disparity was “unwarranted” because of the
racial effects or “warranted” due to the need to provide public safety resources for
the African-American community.
A second type of case would further the discussion of warranted and
unwarranted disparities. Suppose that a state legislature was considering a
proposal to impose a mandatory minimum prison sentence for third-time auto theft.
Under current sentencing laws and practices, let us assume that there are 100 such
convictions annually, 70 of which result in a prison sentence (see table below). Of
this population, 35 are black, 30 are white, and 5 are Latino. Under the proposed
legislation, all 100 offenders would be sentenced to prison; of these, 45 would be
black, 45 white, and 10 Latino.
Race/Ethnicity

Current
Prison Sentences
#
%

New Law Prison
Sentences
#
%

Rate of
incarceration, all
offenses (per
100,000)
Current law

Black
White
Latino
TOTAL

35
30
5
70

50%
43%
7%

45
45
10
100

45%
45%
10%

New law

1750
250
750

How, then, to evaluate the disparity issues here? Here are two ways to
consider this:
1. Proportional Disparity
As a proportion of the current prison population for this offense, blacks
represent 50%, whites 43%, and Latinos 7%. Under the proposed legislation, this
would change to 45% black, 45% white, and 10% Latino. Thus, one might say
that the change would disproportionately disadvantage Latinos (and whites to a
small extent), but “benefit” blacks due to a declining share of the incarcerated
population.
2. Population Disparity
In comparison to the proportional disparity which shows blacks with a
declining share of the incarcerated population for auto theft, an overall population
analysis demonstrates that high rates of incarceration of blacks overall would
become even higher under the proposed change (along with increases for both

1760
265
755

38

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

white and Latino rates as well). Thus, the black rate of incarceration of 1750 per
100,000 would increase to 1760 per 100,000 under the new law. While in itself
not a dramatic change, the overall rate (seven times that of the white rate) would be
exacerbated under the proposal. Therefore, one could say that based on this
measure, minorities would be harmed by escalating rates of incarceration.
People of good will might disagree about which of the above methods is a
more appropriate tool for analysis, but the point is that they provide a context for
assessing racial and ethnic impacts. And as such analyses are developed,
policymakers can begin to determine the relative advantages of each method for
their goals of public safety and fairness. In addition, to the extent that impact
statements received public attention, they would facilitate a broader conversation
among policymakers and their constituents regarding proposed policies and their
implications.
E. Technical Issues
While the process of producing racial impact statements is theoretically
straightforward, there will be challenges in developing such systems. These
include the following:
1. Limited Data
In some situations current sentencing data may not exist. For example, a
legislative change that would make certain types of domestic assaults felonies
rather than misdemeanors would prove to be a challenge for sentencing forecasts.
Many states will not have sophisticated data on case processing and sentencing for
misdemeanor cases. 102 It will also be necessary to develop assumptions regarding
the proportion of current cases that would be charged as felonies under a proposed
law. One means of developing such an analysis is the model used by the North
Carolina Sentencing and Policy Advisory Commission, in which state analysts
survey prosecutors to develop reasonable assessments of how case processing
would change under proposed legislation. 103
2. Availability of Racial/Ethnic Data
While most states maintain reasonably good data on sentencing and
incarceration by race, there is a broad range of comprehensiveness in terms of the

102
For example, national datasets of the Bureau of Justice Statistics only report on felony
offenses in detail. See Bureau of Justice Statistics, U.S. Dep’t of Justice, State Court Sentencing of
Convicted Felons—Statistical Tables, http://www.ojp.usdoj.gov/bjs/abstract/scscfst.htm (last visited
Sept. 10, 2007).
103
Telephone Interview with Susan Katznelson, supra note 94.

2007]

RACIAL IMPACT STATEMENTS

39

availability of ethnic data on Latinos. 104 Some states have produced reasonably
accurate counts in this regard, while others categorize Latinos as “other” or
significantly undercount this population. 105 Such gaps are problematic not only for
producing racial/ethnic impact statements, but for assessing criminal justice
processing issues overall. In addition, while racial data are often fairly complete
for blacks and whites, there is also broad variation in the degree of accuracy for
Native Americans, Asian Americans, and other racial groups. 106 These gaps are
problematic not only for the preparation of racial impact statements, but also for
assessment of criminal justice policy overall, and so clearly need to be addressed
for a variety of reasons.
3. Quantity of Impact Statements to be Produced
In a given year, dozens of sentencing bills may be introduced in a state or
federal legislative body. Most will never receive a committee hearing, let alone a
vote by the full House or Senate. 107 Therefore, should racial impact statements be
required for all such proposals? In at least one state, North Carolina, there are
sufficient staff resources to permit a sentencing impact analysis to be made for all
relevant legislative proposals, with the idea being that one cannot know in advance
which bills are most likely to receive serious consideration. 108 But in jurisdictions
in which such a practice is not feasible, an alternative process could be to produce
a racial impact analysis for any sentencing legislation that has been passed out of
committee and prior to floor consideration.
4. Racial Impact as a Result of Multiple Decision-Making Points
In the case of the crack cocaine legislation, the racial disparities that have
ensued are the result not only of the sentencing provisions but of law enforcement
practices as well, which have disproportionately targeted low-income minority
104

See, e.g., PAIGE M. HARRISON & ALLEN J. BECK, BUREAU OF JUSTICE STATISTICS, U.S.
DEP’T OF JUSTICE, PRISON AND JAIL INMATES AT MIDYEAR 2005 11, tbl.14 (2006) (documenting that
eleven states do not report data on the number of incarcerated Hispanics), available at
http://www.ojp.usdoj.gov/bjs/pub/pdf/pjim05.pdf.
105
Id.
106
For example, in regard to Native Americans, the Bureau of Justice Statistics reports that
“representative statistical data about Native Americans are difficult to acquire and use,” and that such
limitations “severely circumscribe the depth and generalizability of data on American Indians and
inhibit the Nation’s ability to know much of the details about victims, offenders, and the
consequences of crime for both.” LAWRENCE A. GREENFELD & STEVEN K. SMITH, BUREAU OF JUSTICE
STATISTICS, U.S. DEP’T OF JUSTICE, AMERICAN INDIANS AND CRIME 34 (1999), available at
http://www.ojp.usdoj.gov/bjs/pub/pdf/aic.pdf.
107
Glen S. Krutz, Issues and Institutions: ‘Winnowing’ in the U.S. Congress, 49 AM. J. POL.
SCI. 313 (2005).
108
Telephone Interview with Susan Katznelson, supra note 94.

40

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

communities for drug law enforcement. 109 Thus, a racial impact statement would
try to assess these dynamics, in the process incorporating available research on law
enforcement in this area. Following that, a policy discussion should incorporate an
analysis of the relative degree to which the crack cocaine sentencing disparity was
the result of law enforcement or sentencing policy, an analysis that would both
inform the sentencing discussion but also point toward other means of addressing
disparities in the justice system.
5. Courtroom Dynamics
In the hypothetical auto theft legislation described above, the dynamics of
mandatory sentencing virtually assure that, in certain cases, courtroom personnel
will engage in decision-making designed to avoid imposing the mandatory
sentences called for by the proposed statute whether out of concern for managing
court dockets, through the dynamics of plea negotiations, or because of
prosecutorial assessments of the inappropriateness of the penalty in selected cases.
Therefore, a sophisticated sentencing forecast model needs a means of
incorporating such processes into its estimates.
6. Limitations on Projections
While some state sentencing commissions or other bodies can produce
sophisticated projections over a period of ten years or more, other states do not yet
have this capability. 110 This should not be a major impediment to the development
of racial impact statements, though. In such cases, state policymakers can rely on
current sentencing data to produce descriptive analyses of proposed policy
changes. For instance, in the auto theft example described above, a simple
description of the racial/ethnic composition of persons convicted of this crime
would yield a ballpark estimate of the overall impact of changes in sentencing or
corrections policy.
These issues are all challenging ones, but the key point is that analysts in
various states have been assessing these issues for a number of years and have
developed reasonable means of incorporating their assessments into prison and
fiscal projections. Once such a model is established, software used by the state or
federal agency can generally incorporate sentencing and corrections data on race
and ethnicity in addition to other information in order to produce a racial impact
analysis.

109

TONRY, supra note 13, at 105–06.
Telephone Interview with Barbara Tombs, Director of the Center of Sentencing and
Corrections, Vera Institute of Justice (Nov. 7, 2006).
110

2007]

RACIAL IMPACT STATEMENTS

41

F. Implementing Racial Impact Statements
After a racial impact statement is prepared, a legislative body would receive it
prior to a vote by committee or consideration in a floor vote, depending on the
means by which the implementing legislation is drafted. Let us assume that a
committee is considering a sentencing enhancement bill and is informed by the
racial impact statement that it will result in a disproportionate effect on African
Americans. How will the legislative process be affected by such a finding?
Committee consideration of such data should be guided by two questions.
First, do the crime control benefits of such a policy outweigh the consequences of
heightened racial disparity? And second, are there alternative policy choices that
could address the problem at hand without such negative effects?
In the case of school zone drug laws, for example, the public safety question
becomes whether such laws contribute to public safety—reducing drug selling to
schoolchildren—more than existing statutes already do. 111 In this regard,
policymakers would want to note not only the current statute, but how it is
implemented as well. Consider two scenarios in a jurisdiction that does not have
school zone drug laws, for example. In the first case, a drug seller has been
convicted of a drug sale to a consenting adult at 2 a.m. in a school yard. In the
second, a seller has been found guilty of selling to a high school student on the
school playground during lunch recess. Is there a judge in the country who would
not treat the high school case more seriously than the middle of the night sale to an
adult?
Therefore, in such a case, a legislative body might decide to leave current
penalties intact or else to define the new statute in a more targeted way. Such
targeting could include focusing the legislation on direct sales to children, selling
during school hours, or selling directly on (as opposed to near) school property.
Any such priorities would both direct sentencing policy more specifically toward
the area of concern and would almost inevitably reduce the racial disparities that
would ensue under the expanded concept of “school zone.”
In the notorious case of the federal crack cocaine sentencing laws, imagine
that Congress had required the preparation of an impact statement prior to
consideration of the legislation in 1986 and 1988. It is possible that a finding that
80% of the affected population would be African American would not have
affected eventual passage of the legislation. Some lawmakers no doubt would still
have argued that harsh penalties imposed on black offenders would be of great
value to “law-abiding” members of the black community. But it is also possible
that such a finding might have encouraged the exploration of other issues,
including: was a five-gram threshold for prosecution of crack cocaine offenses
appropriate for federal courts, as opposed to being handled under state
111

GREENE ET AL., supra note 64, at 44 (concluding that “[a] substantial body of evidence from
research and policy studies indicates that drug-free zone laws, as they are typically configured, are
not effective in reducing the sale or use of drugs, or in protecting school children”).

42

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

jurisdiction?; was there an appropriate mix of supply-side—interdiction and
enforcement—approaches, and demand-side—prevention and treatment—
initiatives, to the emerging crack problem?; 112 and, was the perceived association
of crack cocaine with violent behavior a function of the drug itself or the new
markets emerging with the drug?
In the case of crack cocaine laws, the impact of a policy change on racial
disparities would have been very substantial. A recent assessment of potential
policy change examined the race and incarceration impacts of eliminating the
100:1 quantity disparity, essentially sentencing crack cocaine offenders as if they
had been convicted as powder cocaine offenders. Using data from a 1997 survey
of federal prison inmates, Eric Sevigny concluded that such a change would have
averted a cumulative total of 24,000 prison years imposed, 90% of which would
have benefited African Americans. 113
A more problematic case would be sentencing enhancement legislation for
persons convicted of violent offenses. This might include sentencing commission
proposals to raise the grid level for armed robbery to include more time in prison
or legislation designed to restrict release from prison for those convicted of a
violent offense. At current rates, African Americans are disproportionately
incarcerated for violent offenses, representing about 45% of persons incarcerated
in state prisons for this category. 114 Therefore, any proposed increases in
sentencing for violent offenses would be likely to have a disproportionate effect.
In such a case, policymakers might well decide that the hoped for reductions
in violent offending would take priority over the goal of reducing disparity, and
that the disparity is a “warranted” one. But it is also conceivable that the racial
analysis could contribute to a broader discussion of the effect of sentence length on
deterrence. In fact, there is a good deal of research suggesting that deterrent
effects are much more a function of certainty than severity of punishment. 115
In other areas of social policy, impact statements are often required to include
alternative options that would provide fewer harmful effects on the social indicator
of concern. Environmental impact statements, for example, frequently include an
assessment of loss of habitat or farmland resulting from proposed development

112
Blumstein, supra note 1, at 9, (stating that “basic observation about drug markets is that
they are inherently demand driven. As long as the demand is there, a supply network will emerge to
satisfy that demand”).
113
Eric L. Sevigny, The Tyranny of Quantity: How the Overemphasis on Drug Quantity in
Federal Drug Sentencing Leads to Disparate and Anomalous Sentencing Outcomes 138 (2006)
(unpublished
Ph.D.
thesis,
University
of
Pittsburgh),
available
at
http://etd.library.pitt.edu/ETD/available/etd-07182006-154904/unrestricted/sevignyericl_etdPitt2006.
pdf.
114
HARRISON & BECK, supra note 42, at 9 tbl.12.
115
See, e.g., Daniel S. Nagin & Greg Pogarsky, Integrating Celerity, Impulsivity, and
Extralegal Sanction Threats into a Model of General Deterrence: Theory and Evidence, 39
CRIMINOLOGY 865 (2001).

2007]

RACIAL IMPACT STATEMENTS

43

projects along with proposals to permit development in ways that are less
harmful. 116
Such a goal might or might not be included as an element of criminal justice
legislation, but the policy discussion related to alternative measures could certainly
be stimulated by the preparation of impact statements. In the challenging area of
sentencing for violent offenders, for example, an assessment could be completed
regarding the efficacy of various measures of reducing violence. Suppose that a
proposed penalty for armed robbery were to be increased from five years to ten
years in prison. For each offender subject to this provision, the state would spend
about $125,000, assuming an estimated annual incarceration cost of about $25,000.
Multiplying this by the number of offenders subject to such a penalty would
produce a figure for the total state investment in addressing crime through such an
initiative. That could then lead to a discussion of the crime control impact
(whether through incapacitation or deterrence) of the additional five years in prison
per offender, compared to projected public safety impacts through investing a
similar sum in policing, drug treatment, preschool programs, or any other
intervention believed to be effective in this regard.
While such a process might seem to some to be wishful thinking in the current
world of public policy, the Washington state legislature has created just such an
institutional mechanism for permitting comparisons of this type. The Washington
State Institute for Public Policy is a legislative agency that prepares assessments of
the effects of various policy initiatives. 117 In the area of crime control, the agency
produces periodic assessments of the cost-effectiveness of a range of program
interventions designed to address crime and violence. 118 A 2006 Institute
publication, for example, assessed the potential of using intervention programs,
prevention programs, and sentencing options as a means of reducing the need for
future prison construction. 119 Using meta-analysis techniques the institute
concluded that programs such as vocational education in prison produced cost
benefit savings of more than $13,000 per program participant, while Scared
Straight programs had negative cost benefits of more than $14,000. 120 Such
evidence-based assessments hold the potential for rational development of public
policy and more effective crime control programming.
116

See the enabling statute of the National Environmental Policy Act, 40 C.F.R. § 1502.1
(2007), available at http://www.nepa.gov/nepa/regs/ceq/1502.htm.
117
The Institute “conducts research using its own policy analysts and economists, specialists
from universities, and consultants. Institute staff work closely with legislators, legislative and state
agency staff, and experts in the field to ensure that studies answer relevant policy questions.”
Washington State Institute for Public Policy, http://www.wsipp.wa.gov/ (last visited Sept. 10, 2007).
118
Id.
119
STEVE AOS, MARNA MILLER & ELIZABETH DRAKE, WASHINGTON STATE INSTITUTE FOR
PUBLIC POLICY, EVIDENCE-BASED PUBLIC POLICY OPTIONS TO REDUCE FUTURE PRISON
CONSTRUCTION, CRIMINAL JUSTICE COSTS, AND CRIME RATES 2 (2006), available at
http://www.wsipp.wa.gov/rptfiles/06-10-1201.pdf.
120
Id. at 9.

44

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

VI. EXPANDING THE USE/SCALE OF RACIAL IMPACT STATEMENTS
Policymakers adopting the use of racial impact statements would have before
them a range of options by which to make use of the data analysis prepared for
sentencing legislation. At the most basic level this would include an assessment of
the degree to which any unwarranted racial disparities are exacerbated by proposed
legislation. But the statements offer a tool for further analysis as well.
In recent years, for example, lawmakers have enacted a host of bills that
extend the range of collateral consequences that result from a felony conviction,
with many of these applying specifically to drug offenses. 121 Persons applying for
federal financial aid for higher education, for example, are excluded from access to
funds if they have been convicted of a drug felony while receiving aid in
college. 122 Here, too, a seemingly race neutral policy clearly has differential social
class, and likely racial, effects. Wealthy persons with a drug conviction are
unlikely to need financial aid to attend college, but low-income students certainly
will. And to the degree that people of color are disproportionately represented
among the low-income population, there will be a racial effect as well. Thus,
expanding the use of racial impact statements to other areas of social policy related
to sentencing could help to alleviate the expansion of racial disparities to these
collateral penalties.
Lawmakers might also want to consider adopting a model of policy tradeoffs
to neutralize the effects of initiatives that might exacerbate disparities. Such a
process has parallels in other policy areas. On environmental issues, the Kyoto
Protocol established provisions for “carbon trading,” whereby industrial companies
that reduce their production of greenhouse gases can sell the “rights” to emit
quantities that exceed permissible limits to other firms. 123 The process, while
criticized by many environmental groups, 124 nonetheless represents a means of
addressing a critical policy concern through incentives and tradeoffs.
One can envision such tradeoffs in sentencing policy, resulting in part from
racial impact statements. The model in this regard relates to the operating
assumptions of most sentencing commissions. Guided by a concern for the
effective use of prison resources, these commissions place a priority on using

121

Mauer, supra note 58, at 5.
Id.
123
United Nations Framework Convention on Climate Change, Emissions Trading,
http://unfccc.int/kyoto_protocol/mechanisms/emissions_trading/items/2731.php (last visited Sept. 10,
2007).
124
Jeff Goodell, Capital Pollution Solution?, N.Y. TIMES MAG., July 30, 2006, available at
http://www.nytimes.com/2006/07/30/magazine/30carbon.html?ex=1311912000&en=7a1aac348e ad2
98b&ei=5088&partner=rssnyt&emc=rss. (“Commissions in Minnesota, Oregon, and Washington
intended to reduce imprisonment sentences for property offenders and increase them for violent
offenders.”)
122

2007]

RACIAL IMPACT STATEMENTS

45

prison space for people convicted of violent offenses while encouraging the use of
community sanctions for people convicted of non-violent offenses. 125
In the realm of race and sentencing dynamics, suppose that a sentencing
commission believed it necessary to enhance prison terms for violent offenses and
concluded that doing so would have disproportionate racial effects. Such a finding
might trigger a search for ways of offsetting that impact. Might there be a set of
offenses for which a reduction in prison time would be appropriate and would
disproportionately advantage minority offenders? Just to be clear, such a
discussion should not be undertaken if it conflicts with public safety concerns
regarding the utility of incarceration for a given set of offense and offender
characteristics. But as a catalyst for such a discussion, a legislative body or
sentencing commission might find a periodic assessment of sentencing policies
that incorporates such an analysis to be a useful policy tool. And given the
massive scale of incarceration in the United States at present, 126 it is not unlikely
that a serious consideration of sentence lengths for various offenses might produce
constructive options for change.
VII. CONCLUSION
While some might argue that racial impact statements are “injecting race” into
considerations of public policy, in fact they merely bring to light data on the
already existing racial dynamics of criminal justice policy. By doing so, they
create the possibility of a policy dialogue on race that acknowledges the
complexity of issues of race and justice, but provides a way to reduce unwarranted
disparities while producing better public safety outcomes.
Far from being just a visionary notion, the concept of racial impact statements
has received support from distinguished quarters. Sentencing scholar Michael
Tonry has called for all sentencing proposals to be “accompanied by or subjected
to impact analyses that project their differential effects for women and for
nationality and ethnic groups.” 127 The American Bar Association’s Justice
Kennedy Commission has gone further in recommending that government bodies
conduct racial and ethnic disparity impact analyses not only of proposed legislation
but of existing statutes as well, along with a call for policymakers to “propose
legislative alternatives intended to eliminate predicted racial and ethnic disparity at
each stage of the criminal justice process.” 128 And in at least one state, Oregon, a
125

E.g., DALE PARENT ET AL., NATIONAL INSTITUTE OF JUSTICE, U.S. DEP’T OF JUSTICE, KEY
LEGISLATIVE ISSUES IN CRIMINAL JUSTICE: THE IMPACT OF SENTENCING GUIDELINES 4 (1996)
available at http://www.ncjrs.gov/pdffiles/sentguid.pdf.
126
The United States has become the world leader in its rate of imprisonment. MAUER, supra
note 20, at 9.
127
TONRY, supra note 2, at 221.
128
A.B.A. JUSTICE KENNEDY COMMISSION, REPORT TO THE HOUSE OF DELEGATES 1 (2004),
available at http://www.abanet.org/media/kencomm/rep121b.pdf.

46

OHIO STATE JOURNAL OF CRIMINAL LAW

[Vol 5:19

legislative proposal to require the preparation of “racial and ethnic impact
statements that assess impact [sic] of prison-related legislation . . . on racial and
ethnic profile of prison population” has been introduced. 129
The rate of incarceration for African Americans in the United States is now at
a level that is seriously affecting life prospects for the generation of black children
growing up today. In addition, the ripple effects of current policy now extend the
impact of incarceration beyond just the individual in prison, but to families and
communities as well. 130 While the criminal justice system has an obligation to
promote public safety, there is also an obligation to promote fairness and justice.
There should not be any inherent contradiction in promoting effective crime
control policies while reducing unwarranted racial disparity, and, in fact, the two
goals are best addressed simultaneously. Racial impact statements provide a tool
for policymakers and the general public to begin to grapple with how to develop
public policy that can be both effective and fair. One would hope that such a
policy would be embraced by all concerned policymakers.

129
H.B.
2933,
74th
Leg.
Reg.
Sess.
(Or.
2007),
http://www.leg.state.or.us/07reg/measpdf/hb2900.dir/hb2933.intro.pdf (last visited Sept. 10, 2007).
130
Mauer, supra note 58, at 1–12.