Is America’s Criminal Justice System Racist?

Defund Murder Poster
Image courtesy of “#BLM Demonstration Hamburg 2020-06-05” (CC BY-NC 2.0) by Rasande Tyskar.


Is America’s criminal justice system racist? This question has resurged in interest following the police killing of George Floyd, an incident that appeared to be a wanton abuse of police power on an unarmed African-American man. The notion that the institution of criminal justice in America — from policing to convicting — is systemically racist has pervaded throughout social media, manifested in attention-grabbing videos and simplified infographics that are primarily spread by young people.

As a 16-year-old, I can attest to this. Within the past month or so, social media has provided me with a monolithic viewpoint: America’s justice system is indelibly and blatantly anti-black. Dare to question the validity of the claim, and you are an abettor of white supremacy. Teenagers, lacking a sense of purpose while being quarantined during COVID-19, have united together against the virus of racism.

Unlike many on the political Right — the side I identify with more — I consider myself to be open-minded, my beliefs driven by evidence and understood truths. Thus, I have tasked myself with analyzing the contention that systemic, anti-black racism is omnipresent within America’s criminal justice system.

I have strived to consider as many pertinent academic studies as possible. I am only a high school student; I am not a qualified expert or even a college graduate. With that being said, I believe I am intellectually capable of understanding a study and its findings. And, as will be seen below, the majority of my analysis comes from the authors of the studies themselves.

It is virtually impossible for me to examine every study done on the topic of racial disparities in the criminal justice system. However, I believe I have analyzed the most oft-cited, relevant, recent, and thorough ones, and I believe my analyses of them can be applied to other related, smaller studies. If I hear of a study that entirely nullifies my analyses or conclusions, I will ensure that this post is updated.

This post will be divided into the aspects of criminal justice that, from what I’ve gathered, receive the most attention from those who claim that America’s justice system is racist. A table of contents is provided below to allow easy navigation.


I will first begin with analyzing a claim that originates from a popular sentiment: police officers unjustly police African Americans more than any other group.

For this to be true, the rates at which African Americans are arrested for crimes would have to be higher than the rates at which African Americans are reported by victims as the offender of crimes. Is this the case?

Fortunately, the FBI’s National Incident-Based Reporting System allows the claim to be elucidated. In 2018, more than seven thousand law enforcement agencies reported their crime data through NIBRS, allowing data for both reported offenders and arrestees to be viewed.

Comparing the percentage of black reported offenders for each crime listed with the percentage of black arrestees for each crime gives the following:

CrimeReported Offenders (% Black) Arrestees (% Black)
Assault Offenses35.930.3
Homicide Offenses46.652.2
Human Trafficking Offenses50.750.6
Sex Offenses23.623.7
Burglary/Breaking & Entering19.226.8
Fraud Offenses20.327.0
Larceny/Theft Offenses23.225.2
Motor Vehicle Theft19.127.5
Stolen Property Offenses27.532.0
Animal Cruelty20.315.4
Drug/Narcotic Offenses23.725.3
Gambling Offenses21.745.5
Pornography/Obscene Material18.016.3
Prostitution Offenses32.229.5
Weapon Law Violations42.143.9

For most crimes, the two percentages are either relatively equivalent (a difference of 5% or less) or the percentage of arrestees who were black was lower. But what accounts for the crimes that had a significantly higher percentage of black arrestees than black reported offenders. such as burglary and gambling offenses?

Let’s take a look at those crimes again, this time with white reported offenders and white arrestees:

CrimeReported Offenders (% White) Arrestees (% White)
Homicide Offenses40.443.0
Burglary/Breaking & Entering40.568.1
Fraud Offenses45.166.6
Motor Vehicle Theft40.167.8
Gambling Offenses45.236.1

Interestingly, with the exception of gambling offenses, the percentage of arrestees who were white was also significantly higher than the percentage of reported offenders who were white for the crimes that had the same statistical effect for African Americans.

On a final note, let’s compare the change in percent — percent of reported offenders compared to percent of arrestees — for both races for the crimes in which significant increases occurred:

CrimeWhite Percent Change (Arrestees – Reported Offenders) (%)Black Percent Change (Arrestees – Reported Offenders) (%)
Homicide Offenses+2.6+5.6
Burglary/Breaking & Entering+27.6+7.6
Fraud Offenses+21.5+6.7
Motor Vehicle Theft+27.7+8.4

Based on this data, there is actually more validity to the claim that whites are overpoliced than to the claim that blacks are overpoliced. Of course, both claims are foolish. The significant increase in arrestees is probably because of the type of crime that the crimes are: crimes in which witnesses are far less likely to be present, such as motor vehicle theft and extortion. In these crimes, the racial identification of suspects may be unreliable, thus resulting in a significantly higher percentage of arrestees compared to the percentage of reported offenders.

Academic literature corroborates this finding. A 2003 study found that “the odds of arrest for white offenders were approximately 22% higher for robbery, 13% higher for aggravated assault, and 9% higher for simple assault than they are for black offenders.”

Supporting this finding, a 2019 study discovered that “the odds of arrest for black offenders were roughly 25.1% less than the odds of arrest for white offenders” and that “the odds of arrest for black co-offenders were roughly 42.2% less than the odds of arrest for white co-offenders.” The study found that the only event in which a black offender is more likely to be arrested than a white offender is in an interracial co-offender partnership — a dyad composed of a white and black offender — where “the odds of arrest for black offenders were 3.1% greater than the odds for white offenders.”

Thus, if the notion of overpolicing were true, black offenders would be more likely to be arrested than white offenders in virtually all cases. But there is only one instance where that occurs, and it is a small difference in likelihood (3.1%). If we are to view that as anti-black bias within the police, then we must also view the 25.1% less likelihood of black offenders to be arrested than white offenders and the 42.2% less likelihood of black co-offenders to be arrested than white co-offenders as conspicuous, anti-white racism.

Thus, there is a lack of evidence demonstrating that African Americans are overpoliced. Surprisingly, the inverse claim, that whites are overpoliced, may potentially have more validity to it.

Police Killings

A prevalent claim by many is that African Americans are unjustly killed more by police than whites. Incidents like the police killings of George Floyd and Breonna Taylor, two unarmed African Americans, further cement this claim.

On the surface, this claim appears to be true. Viewing the raw data of police killings reveals that African Americans are 2.5 to 3 times more likely than whites to be killed by police.

This figure would be indicative of severely unjust treatment if both races commit crime at the rate that their population levels should suggest. However, as demonstrated by the National Incident-Based Reporting System data above, African Americans, despite only composing 13.4% of the United States population, disproportionately commit more crime than whites, who constitute 76.3% of the American population (side note: the NIBRS considers Hispanic Americans as white; thus, I will use the figure of the population that includes white and Hispanic Americans, which is 76.3%). Thus, the appropriate question to ask to discern whether African Americans are unfairly killed by police is as follows: do African Americans get killed more by police than crime rates should warrant?

Sendhil Mullainathan, a then-professor of economics at Harvard University, examined this question in a 2015 editorial in The New York Times. He writes of several interesting conclusions in it.

He first establishes that, per the FBI’s Supplementary Homicide Report at the time of his editorial’s publication, “31.8 percent of people shot by the police were African-American.” While the figure may only consider police shootings and not police killings, it’s congruous with the data given by Mapping Police Violence that “Black people have been 28% of those killed by police since 2013.” Furthermore, The Washington Post, based on its own analysis, concluded that African Americans accounted for 24% of fatal police shooting victims in 2015.

Professor Mullainathan goes on to write that, at the time of his writing, 28.9% of national arrestees were African American. In 2018, they accounted for 24.7% of national arrestees. As either number — 28.9% or 24.7% — is minimally different from either figure of police killings — 28% or 24% — the then-Harvard professor concluded that “if police discrimination were a big factor in the actual killings, we would have expected a large gap between the arrest rate and the police-killing rate.”

Interestingly, there is evidence that potentially indicates the inverse: that whites, when accounting for homicide rates, are more likely to be killed by police than African Americans. That was the finding of Peter Moskos, a professor at John Jay College of Criminal Justice, who concluded that “adjusted for the homicide rate, whites are 1.7 times more likely than blacks to die at the hands of police.” Professor Moskos explained that the disparity was caused by the fact that police assigned to black majority neighborhoods with high crime rates experience “more political fallout when they shoot, and thus receive better training and are less inclined to shoot” and because they “are more accustomed to dangerous situations and thus are more likely to resolve them without resort to lethal force.”

A 2018 study that analyzed officer-involved fatal shootings in 2015 and 2016 arrived at results that found even more of a racial disparity. Using four sources for the criminal activity of African Americans and whites — the FBI’s Summary Report System, the National Incident-Based Reporting System, the Bureau of Justice Statistics’ National Crime Victimization Survey, and the CDC’s WONDER database — the study found:

  • Whites were 2.6 to 3.9 times more likely to be killed by police gunfire than African Americans when each group’s homicide rates were accounted for (the variation is due to the source that was used to determine the homicide rates for each group).
  • Whites were 2.5 times more likely to be killed by police gunfire than African Americans when each group’s assault rates were accounted for.

These conclusions were mirrored by the findings in three separate simulations involving almost all-white and nearly all-male police officers from the Spokane Police Department. Conducted by three researchers at Washington State University, the studies used “highly realistic police simulators, in which actors in various scenarios approach and respond to officers on large, high-definition video screens in an attempt to recreate critical situations on the street.” The findings were intriguing: white police officers — even those who had racial biases that were found through a series of written and oral tests, including the Harvard Implicit Association Test — were “three times less likely to shoot unarmed black suspects than unarmed white suspects.” Moreover, “officers took significantly longer to shoot armed black suspects than armed white suspects.”

Whether the results of the three simulations can be applied to real-life incidents is debatable. Lois James, one of the researchers behind the studies, acknowledged this but noted that “the police did not know that race was a factor in the project.” Additionally, her explanation of the results aligned closely with the explanation given by Professor Moskos as to why police officers are more hesitant to shoot African Americans: “the social and legal consequences of shooting a member of a historically oppressed racial group … paired with the awareness of media backlash that follows an officer shooting a minority suspect.” Thus, if the results of the simulations are indeed outliers, it is rather coincidental that they align so closely with the findings of real-world data.

Other simulations have found both similar and different results. A 2007 study involving officers from the Denver Police Department, for instance, found that “police do not show bias in their ultimate decisions” to shoot a target. However, the study found potential racial bias in the speed with which Denver police officers made a decision to shoot or not. A 2014 analysis on a decade of research studying racial biases in police shootings concluded that “although police are affected by target race in some respects, they generally do not show a biased pattern of shooting.” Thus, while simulations involving police officers may differ in some findings, a general consensus has been reached concerning the lack of evidence suggesting concordant bias in police shootings.

On the other side, there’s a study that appears to support the claim that unarmed African Americans are more likely to be shot by police than unarmed whites. Authored by University of California, Davis Professor Cody T. Ross, a 2015 analysis of police shootings concludes that “the probability of being black, unarmed, and shot by police is about 3.49 times the probability of being white, unarmed, and shot by police on average.” In my reading of the analysis, however, I noticed several significant problems — even from a layman’s perspective — that incline me to consider the results with skepticism.

The most glaring problem in Professor Ross’ study concerns the source of the data. The source is not an official database or even a database managed by a media outlet like The Washington Post or The Guardian: rather, it’s a crowd-sourced Google Form, where anyone can claim a police shooting occurred. He argues that the source he uses is more reliable than other databases, a strange claim that is largely unevidenced.

Additionally, as a criticism of his study included here points out, Professor Ross utilized weapon- and assault-related arrest rates to measure violent crime, “a seemingly poor proxy since people can be arrested for weapon-related incidents that don’t involve violence (e.g., having a gun stolen), and because black people are more over-represented among violent crimes like murder than they are assault.”

There are other critiques of the study mentioned here — including certain seemingly illogical conclusions of the study and how they compare with the conclusions of a similar study that uses sounder sources of data and methodologies — but I believe the two problems mentioned above are sufficient for the purpose of this post.

Even if I were to accept the results of the study, there may be separate factors unaccounted for in the analysis that may explain the racial disparity. One factor could be resisting arrest: several cities, including San Francisco, San Diego, and New York City, have released data showing that African Americans are significantly more likely to be arrested for resisting arrest than non-African Americans. In Chicago, African Americans accounted for 77% of all arrests for obstruction of justice and resisting arrest from September 2014 to September 2015. As demonstrated in the previous section, the notion of overpolicing in African-American communities is unfounded; thus, I am not inclined to believe that the arrests for obstruction of justice and resisting arrest are due to racism rather than criminality. If racism were at play, there would presumably be differences between black arrestees and white arrestees for other crimes that are incongruous with reported offenders for the crimes.

Thus, the claim that police officers are unjustly killing African Americans at higher rates than they are killing whites is unfounded. Surprisingly, there is potentially more validity to the claim that whites are disproportionately killed more by the police than African Americans when crime rates are accounted for.

Drug Arrests

Some argue that racism within the police is most exhibited in low-level, “victimless” arrests, such as drug arrests. On the surface, the logic makes sense: if police officers wanted the best chances to be racist without being caught, they presumably wouldn’t manifest it in killings but rather in insignificant arrests.

One claim circulated regularly is that African Americans are actually less likely to use drugs than whites and yet are arrested at far higher rates. Let’s begin by examining this claim.

The assertion that African Americans are less likely to use drugs than whites comes from self-reported studies, in which individuals report their drug history. There are numerous studies that have evaluated the validity of self-reported studies, such as a 2005 one, which sampled 600 adults from Chicago, comparing self-reports of drug use for cocaine, marijuana, and a combination of the two with drug test results. It concluded that “overall, the results replicate and extend a growing body of research suggesting that African Americans underreport substance use on surveys.” Studies with similar conclusions can be found here, here, here, and here.

Thus, the claim that African Americans use drugs at lower rates than whites is predicated entirely on self-reports, and, as concluded by a multitude of studies, African Americans generally underreport substance use more on self-reports than whites do.

The purchasing behavior of drugs has also been found to differ between African Americans and whites. A 2006 study found that, compared to whites, “African Americans are nearly twice as likely to buy outdoors, three times more likely to buy from a stranger, and significantly more likely to buy away from their homes.” These methods of purchasing drugs will assuredly increase the likelihood of being arrested.

Finally, the disproportionate crime rates among the African-American population may also contribute to blacks being more likely to be arrested for drugs. For instance, African Americans are disproportionately more likely to be arrested for burglary, motor vehicle theft, and counterfeiting than whites. It is very possible that, when being arrested for those crimes, the arrestees also had drugs on their person.

As a result, racism cannot be prescribed as the causation for the contemporary racial disparity in drug crime arrest rates between African Americans and whites until other factors, such as the purchasing behavior and whether another crime was the initial reason for arrest, are accounted for in analyses.

Police Stops and Searches

Similar to the logic concerning racism in drug arrests, many attest that routine police stops and searches exhibit a racial bias against African Americans.

A recent analysis of nearly 100 million traffic stops conducted in America appears, on the surface, to suggest that racism is involved in police stops. The analysis concluded that “black drivers were less likely to be stopped after sunset, when a ‘veil of darkness’ masks one’s race, suggesting bias in stop decisions.” This clearly looks like blatant evidence of racial profiling in policing.

However, I have noticed that when this study is cited, the rest of the analysis is — perhaps purposefully — left out. Let’s see what the authors of the reports go on to say about their findings:

The veil-of-darkness test is a popular technique for assessing disparate treatment but, like all statistical methods, it comes with caveats. Results could be skewed if race-specific driving behaviour is related more to lighting than time of day, leading the test to suggest discrimination where there is none. Conversely, artificial lighting (for example, from street lamps) can weaken the relationship between sunlight and visibility, and so the method may underestimate the extent to which stops are predicated on perceived race. Finally, if violation type is related to lighting, the test could give an inaccurate measure of discrimination. For example, broken tail lights are more likely to be detected at night and could potentially be more common among black drivers, which could in turn mask discrimination.”

Thus, factors other than racism — such as race-specific differences in driving behavior — could serve as equally logical explanations for the racial disparity in police stops. Until these factors are accounted for, they are as potentially likely to be causative for the disparity as racial bias is.

Another conclusion found in the study that many cite as evidence of racial bias is that “the bar for searching black and Hispanic drivers was lower than that for searching white drivers.” Again, let’s see what the authors note about this finding:

“As with all tests of discrimination, it is important to acknowledge limits in what one can conclude from such statistical analysis per se. For example, if search policies differ not only across, but also within, the geographic subdivisions we consider, then the threshold test might mistakenly indicate discrimination where there is none. Additionally, if officers disproportionately suspect more serious criminal activity when searching black and Hispanic drivers compared to white drivers (for example, possession of larger quantities of contraband), then lower observed thresholds may stem from non-discriminatory police practices.”

Once again, factors besides racial bias are as potentially likely to be causative of the racial disparities in police searches.

Does the study overwhelmingly find a racial disparity in police stops and searches between African Americans and whites? Yes. But does it also mention a multitude of factors other than racism that are as potentially causative of the disparity? Undoubtedly.

Let’s take a deeper look at some of these factors. A 2012 analysis of race-specific differences in speeding patterns found that “Black drivers speed more frequently and engage in more severe speeding compared to White drivers,” replicating the results of studies in states such as Pennsylvania, New Jersey, and North Carolina. Seat belt use has also been found to be chronically lower among African-American drivers, and non-moving violations, such as broken tail lights and expired registration tags, have found to be significantly higher among blacks.

Let’s now turn to what appears to be a scathing report of racism in policing: the Department of Justice’s 2015 Investigation of the Ferguson Police Department. The investigation found, among many other findings, that from 2012 until 2014, African Americans accounted “for 85% of vehicle stops, 90% of citations, and 93% of arrests made by FPD officers, despite comprising only 67% of Ferguson’s population,” concluding that the disparities “are unexplainable on grounds other than race and evidence that racial bias, whether implicit or explicit, has shaped law enforcement conduct.”

Several rebukes have been made toward the investigation’s methodology and conclusions. Peter Kirsanow, a member of the U.S. Commission on Civil Rights, described the report as “so buffoonish, so replete with conclusions unsupported by facts, so lacking in basic methodological rigor.” He proceeds to echo what I mentioned above:

“Several studies over the last 20 years (including data adduced before the U.S. Commission on Civil Rights) show that black drivers commit various types of traffic offenses — including speeding, driving under suspension, DUI, and running red lights and stop signs – more often than drivers of other races. Could it be possible these phenomena contribute to the ‘unexplainable’ disparities?”

The Manhattan Institute’s Heather Mac Donald mirrored this sentiment:

The Department of Justice’s claims “are meaningless as a measure of police behavior, unless one considers the underlying rate of traffic offenses. If blacks are disproportionately represented among speeders, red-light runners, and drivers without updated vehicle registration, say, then their higher rate of being stopped simply means that the police are applying the traffic laws neutrally to lawbreakers.”

Another oft-repeated conclusion of the report is that African Americans were more than twice as likely to be searched after traffic stops but were “found in possession of contraband 26% less often than white drivers.” Mac Donald, in her book The War on Cops, addresses the misleading nature of this finding:

“The report says that this disparity exists ‘even after controlling for the type of search conducted, whether a search incident to arrest, a consent search, or a search predicated on reasonable suspicion,’ but the report does not reveal how much of a disparity persisted in each type of search: automatic searches incident to an arrest should not be used to measure alleged police bias. The analysis also does not take into account differences in driver behavior following a stop that could increase an officer’s inclination to search.”

Several other studies have examined specific cities and localities. A recent analysis of D.C. Metropolitan Police Department data by the ACLU found that “Black people composed the vast majority of individuals subjected to stops or searches despite not violating the law.” The report offers three reasons as to why this racial disparity is suggestive of racial bias. I’ll respond to each one individually:

First Reason: “The vast majority of people who experienced the least justifiable stops were Black.” (The ACLU considered unjustifiable stops as stops that did not involve the issuance of a warning or ticket or the making of an arrest.)

Response: In a 2006 report submitted to the U.S. Department of Justice, police officers in Savannah, Georgia “were observed and debriefed after they became suspicious about an individual or vehicle.” Thus, the report was able to discern the leading reasons as to why officers form suspicions, even if they later turned out to be untrue.

The report found that “the main reason for forming suspicion was the behavior of the suspect(s).” The second largest reason was “information provided by either a dispatcher or fellow officer,” including “information provided by the department or fellow officers concerning characteristics of suspects, crimes, or vehicles thought to be related to specific crimes.” The third most prevalent reason was “related to time and place” and concerned “what activities should or should not be expected” at a particular place “after a particular time.”

I cannot speak for race-specific differences in behavior, but there are certainly racial disparities in crime in Washington, D.C. An analysis of D.C. murder suspects from 1998 through 2000, for instance, found that 94% of suspects arrested for homicide were African-American, despite African Americans composing 46% of the D.C. population. The disproportionate crime committed by African Americans results in more warrants for arrests — as they increase, so does the number of false suspicions and incorrect arrests. This is not a product of mere racism: rather, it is a logical product of the fact that unintended events will increase when the sample — in this case, aggregate crime rates for African Americans — also increases. It is an unfortunate fact that many black D.C. residents will be mistakenly stopped and arrested due to the nature of disproportionate crime rates, but it is not, by itself, evidence of racism.

Second Reason: “Racial disparities exist even in parts of the District where Black people make up a very small share of the residential population.” Per the ACLU’s findings, “86% of the people stopped outside the traffic context were Black, while only 5.6% were white.”

Response: I believe my explanation above can also be applied here. The ACLU does not have data as to why police officers stopped a disproportionate percentage of African Americans, and yet they decide to attribute officers’ rationales to mere racism, when other factors, as mentioned above, are as potentially causative.

Third Reason: “Black people are searched at higher rates than white people, even though searches of Black people and white people result in weapons seizures at nearly equal rates.”

Response: This reason follows the same logic that the DOJ’s Investigation of the Ferguson Police Department follows. Until crucial factors such as the type of search and the behavior of the individual searched are accounted for, they are as potentially causative for the racial disparity in searches as racial bias is.

Analyses of police data in states such as Tennessee and North Carolina yielded conclusions similar to the ones in the ACLU report: African Americans are far more likely to be stopped and searched by police, and yet contraband is more likely to be found among white drivers. But again, until factors such as race-specific differences in driving behavior, outstanding warrants, and the behavior of the driver after being stopped are accounted for, racism cannot be used as the explanation for the racial disparity in police stops and searches.

Finally, I want to briefly touch on Floyd v. City of New York, a set of cases heard by the U.S. District Court for the Southern District of New York concerning the New York Police Department’s former practice of stop-and-frisk. The sitting judge, Shira Scheindlin, ruled the practice to be unconstitutional and “racially discriminatory.” However, Judge Scheindlin was later removed from the case by a federal appeals court due to her giving “an appearance of bias by maneuvering to get the case and giving news interviews.” The Second U.S. Circuit Court of Appeals panel concluded that she “ran afoul of the Code of Conduct for United States Judges, and that the appearance of impartiality surrounding this litigation was compromised.” Thus, I am inclined to view the ruling of Floyd v. City of New York with skepticism.

Until other above-mentioned relevant factors are accounted for in studies, the claim that racism is the reason for the racial disparity in police stops and searches is simply an assumption and is unsubstantiated.

Pre-Trial Outcomes

Beyond policing, many claim that anti-black racism can be found in America’s court system. Let’s begin by examining racial disparities in charge bargaining and bail setting.

A 2011 research summary from the Bureau of Justice Assistance concluded that “studies have generally found a relationship between race and whether or not a defendant receives a reduced charge,” citing two studies.

However, a 2017 study on the same topic found the issue to be far more complex than just a simple racial disparity. When analyzing just the race of defendants to determine the odds of receiving a charge reduction due to pleading guilty, the study found:

“Pleading guilty increases the probability of a charge reduction by 50.1% for blacks, as opposed to 55.8% for whites … Blacks generally have slightly lower offense seriousness scores, more extensive prior records, and are detained at higher rates—all factors that decrease the likelihood of a charge reduction. This may partially explain the lower value blacks are getting for their plea.”

When both the race and gender of defendants were analyzed, however, the study yielded fascinating results:

“Pleading guilty increases the probability of a charge reduction by 46.1% for black males, compared to 58.1% for black females, 53.9% for white males, and 55.9% for white females.”

Surprisingly, black females were found to benefit most from pleading guilty, a phenomenon that surely would not occur in a racist criminal justice system. Moreover, there is a larger disparity between black males and black females than black males and white males, potentially providing validity to the claim that gender is a more influential characteristic than race in America’s criminal justice system.

This finding was also the conclusion of a 2012 study that examined racial disparities in bond amounts. When looking at just the race of defendants, the study concluded:

“Analyses of over 5,000 felony defendants from an urban Ohio jurisdiction revealed that significant main effects of a defendant’s race on release on one’s own recognizance (ROR), bond amounts, and prison sentences were rendered nonsignificant when controlling for legal factors, such as offense severity.”

However, when taking into account both race and gender, the study found:

  • African-American males had the lowest odds of being released on recognizance, while African-American females had the highest odds compared to every other group.
  • African-American males eligible for bond were assigned higher bond amounts relative to all other groups, with African-American females receiving just slightly higher bond amounts than white females.

Thus, a racial disparity in bond procedures only appeared when gender was categorized, with African-American women having the highest odds of being released on recognizance.

What could possibly explain this racial disparity among males? The authors of the report offer a potential explanation:

“If a larger “belt” of disadvantaged African American communities corresponds to higher crime rates in these communities, then there might develop a cultural orientation among non-residents (including court actors) of less tolerance toward offenders from those neighborhoods. This would not necessarily lead to a general race effect on pre-trial release decisions and sentencing, but instead would be consistent with the interaction effects reflecting poor community residents who are most often arrested (i.e., young, African American men). Prosecutors and judges may consider the quality of community life when prosecuting and sentencing offenders from particular neighborhoods. Young, African American males could be at higher risk of receiving harsher treatment by court actors because they are more likely to reside in high-crime communities and might be perceived by court actors as contributing to the deterioration of these neighborhoods.”

Assuming there aren’t other confounding variables that are driving this racial disparity, there is again potentially more validity to the claim that gender — i.e., being a male — is more likely to influence a criminal justice proceeding than race — i.e., being black. If the criminal justice system were racist, racial disparities would be present regardless of gender.

Jury Bias

Many claim that juries exhibit an unfavorable bias toward African Americans.

Jury members are selected through a process known as voir dire, in which potential jurors are questioned by a judge or a lawyer to determine their suitability for jury service. A 2011 study examined voir dire processes in 173 proceedings in North Carolina, finding that “even after controlling for other factors potentially relevant to jury selection, a black venire member had 2.48 times the odds of being struck by the state as did a venire member of another race.” The authors of the report testified (page 58) that “the probability of this disparity occurring in a race-neutral jury selection process is less than one in ten trillion.”

Ignoring the small possibility of perjury, prosecutorial discrimination towards African Americans in jury selection appears to be present. North Carolina passed the Racial Justice Act in 2009 to attempt to mitigate this disparity, but it was repealed in 2013 on the grounds that it served as a de facto prohibition on the death penalty.

Interestingly, another study on jury selection in North Carolina found that defense attorneys strike potential white jurors far more often than potential black jurors, excluding “22 percent of the available white jurors versus 10 percent of the available black jurors.” It appears that both races experience some form of racial bias in jury selection, with the race that experiences the bias depending on whether a state prosecutor or defense attorney is selecting the jurors.

Another question has to be answered, though, to determine if a racial disparity in jury selection means that African-American defendants are more likely to be unjustly found guilty than white defendants: is there evidence demonstrating that jurors generally demonstrate an anti-black bias when determining whether a black defendant is guilty?

A study from 2010 attempted to answer that question. In it, two groups of mock jurors were shown identical race-neutral photographic evidence from an armed robbery, with one group shown a picture of a “dark-skinned perpetrator” and the other group shown “an otherwise identical photo of a lighter-skinned perpetrator.” The study concluded:

“Participants who saw the photo of the dark-skinned perpetrator were more likely than participants who saw the photo of the lighter-skinned perpetrator to judge the evidence as tending to indicate criminal guilt and were also more likely to believe that the defendant was guilty of armed robbery.”

There are significant limitations to this study, however. For one, the authors note that the study’s population — 66 students at the University of Hawaii acting as mock jurors — is not indicative of jury members serving under legal obligation. Additionally, the authors cited a 2003 comprehensive review of a variety of studies involving mock juries that concluded that “no consensus has been reached regarding the influence of a defendant’s race on White mock jurors.” Interestingly, the 2003 report found that “Black mock jurors seem to be influenced by a defendant’s race regardless of the salience of racial issues at trial,” while white jurors appear to be less influenced.

Furthermore, a 2005 meta-analysis of 34 studies involving mock juries found that African-American mock jurors demonstrated an in-group bias in voting on the guilt of a defendant that was almost 15 times larger than the in-group bias seen among whites (page 628; 0.208 divided by 0.014). In other words, black mock jurors were far more likely to demonstrate a favorable bias toward black defendants than white mock jurors were to exhibit a favorable bias toward white defendants.

A similar 2014 meta-analysis “found evidence of a stronger outgroup bias for Black jurors than White jurors,” with Hispanic defendants being the one group that white jurors exhibited a stronger bias for than black jurors. Put another way, black mock jurors were more likely to exhibit an anti-white bias than white mock jurors were to demonstrate an anti-black bias.

Thus, while there is evidence that undoubtedly reveals an anti-black bias in jury selection, there is also evidence that suggests an anti-white bias in jury selection. More importantly, there does not appear to be sufficient evidence that indicates that the racial constituency of a jury affects whether an African-American defendant is unjustly found to be guilty. In fact, there is potentially more validity to the claim that black jury members demonstrate a significantly larger in-group bias and out-group bias than white jury members.

Convictions and Sentences

There are numerous studies that appear to suggest that black defendants are more likely to be unjustly convicted and receive harsher sentences than similarly-situated white defendants.

One analysis that is cited often is a 2015 report by The Sentencing Project, which found that studies on racial disparities in prison populations “concluded that between 61% and 80% of black overrepresentation is explained by higher rates of arrest … the remainder might be caused by racial bias, as well as other factors like differing criminal histories.” The report proceeds to write that racial differences in criminal histories “explain a substantial, but incomplete, portion of the differences in the prison population for non-drug crimes.” The report does not say what percentage of black overrepresentation in prison populations is due to relevant legal factors, such as racial differences in criminal histories. It does, however, offer “four key sources of unwarranted racial disparities in criminal justice outcomes,” two of which I will analyze below.

1.“Disparate racial impact of ostensibly race-neutral policies and laws.”

The report claims that African Americans are unfairly arrested for drug crimes, a claim which I analyzed above in the section “Drug Arrests.” It then claims that certain laws that punish certain offenses more harshly than others are racist, as African Americans commit the said offenses more than whites and thus are punished more — a contention that is backward in logic. Adding on to this, the report asserts that habitual offender policies are racist, a claim I examine below.

2. “Racial bias among criminal justice professionals.”

The report makes several claims here:

  • Experimental research has shown police officers “more quickly shot at armed black suspects than at armed white suspects.” I addressed this research — as well as research that came to opposite findings — in the section “Police Killings” above.
  • Police are more likely to pull over people of color, a claim I examined in the section “Police Stops and Searches” above.
  • “Prosecutors are more likely to charge people of color with crimes that carry heavier sentences than whites.” A 2010 study found that claim to be misleading, finding instead that age and gender were far more important factors in the imposition of charges and that “white defendants between thirty and thirty-nine years old were significantly less likely to have their cases dismissed as compared to young, Black defendants.”
  • “Judges are more likely to sentence people of color than whites to prison and jail and to give them longer sentences, even after accounting for differences in crime severity and criminal history.” A 2013 study found that the racial disparity in convictions and sentence lengths “was completely accounted for after including covariates for self-reported lifetime violence and IQ.”
  • “Studies of mock jurors have even shown that people exhibited skin-color bias in how they evaluated evidence,” a claim I examined in the section “Jury Bias” above.

The other two reasons the report provides concern the fact that low-income defendants tend to experience worse outcomes in the criminal justice system than their high-income counterparts, and that because African-Americans defendants are disproportionately more likely to be lower-income, they are more likely to be convicted. This may very well be true, but it indicates classism, not racism, in the criminal justice system.

Some point to a 2017 report by the U.S. Sentencing Commission which found that “Black male offenders received sentences on average 19.1 percent longer than similarly situated White male offenders.” Fortunately, the authors were candid about the report’s limitations and whether their findings can suggest racial bias:

“Some potentially relevant factors were not included, such as whether the offender’s criminal history included violent criminal conduct, the offender’s family ties, and the offender’s employment history … accordingly, the results presented in this report should be interpreted with caution and should not be taken to suggest race or gender discrimination on the part of judges.”

A 2014 study came to similar conclusions: “blacks receive sentences that are almost 10 percent longer than those of comparable whites arrested for the same crimes.” The report proceeds to offer multiple explanations besides racism that may explain the disparity:

“The black and white defendant pools differ on two key legally relevant dimensions … first, black defendants, on average, have more extensive criminal histories … second, there are differences in the distribution of arrest offenses … For example, black defendants are more likely than white defendants to be arrested for weapons offenses. Black arrestees are also more likely to have at least one aggravating factor noted in the written description of the arrest offense. In addition, there are differences in observables that, while not legally relevant, could be correlated with case outcomes, in particular, socioeconomic status. Black arrestees are more likely to be sufficiently poor to qualify for a publicly funded attorney and 43 percent of black arrestees are high school dropouts compared with only 29 percent of whites.”

A 2012 study on penal decisions made by judges found that these factors do serve as influences on a judge’s discretion, concluding that “judges are very interested to obtain information regarding offender’s personality, level of education, professional background, and psycho-social identity, being focused on external motivation of offender.”

Thus, until these factors are accounted for, they are as potentially causative for the racial disparity in sentencing lengths as racism is. The study was not actually able to find “comparable” white and black defendants, as certain relevant legal and extralegal factors were not controlled for.

This difficulty to find truly comparable white and black offenders recurs in many studies. For instance, a 2008 study on habitual-offender sentencing in Florida found that African-American offenders have a 28% greater likelihood to be charged as a habitual offender than comparable white offenders. The report notes that comparability was found by only considering offense seriousness and criminal history, ignoring relevant factors such as education and socioeconomic conditions. Until these variables are controlled for, racism cannot be established as a causation for the racial disparity.

A 2015 study on racial disparities in sentencing also, on the surface, found shocking results: “The average sentence received by whites for their first offense is about 270 days shorter than the average sentence received by blacks.” To make matters worse, the study additionally found that when disaggregating black sentences by skin color, light-skinned blacks have little to no disparity in sentencing compared to whites while “dark-skinned blacks receive sentences that are 400 days longer” than the average white person.

While those findings certainly suggest racial bias, the report’s authors do not instantly jump to racism as the explanation:

“As with disparities based on race alone, skin color disparities in justice outcomes also could be the result of factors that do not indicate overt color discrimination. Lightskinned blacks have been shown to have higher socioeconomic status, which might give judges and jurors more favorable opinions of them. Their higher economic status may also afford blacks with light skin the ability to afford a better defense. Finally, dark-skinned blacks might have longer criminal histories relative to those of their lighter counterparts.

As argued in the literature review, the differences in sentencing could be due to factors other than race, such as the inmate’s personal characteristics or offense severity.”

A 2014 report on racial disparities in probation revocation from the Urban Institute found that “the odds of revocation for white probationers were
between 18 and 39 percent lower than for their black counterparts after controlling for available factors.” Continuing the pattern of not accounting for other relevant factors, the report concedes the limitations of its findings:

“Data on some key factors likely related to revocations were not available for analysis. In no site was the data sufficiently populated regarding violation type, including whether violations were related to new crimes or technical violations of probation conditions. This is a very substantial limitation, as the type of violation is strongly related to the likelihood of revocation.

We also did not have the data necessary to parse out the contributions of different decision points and actors to the disparity. Probation revocations are a product of probationer conduct, probation officer discretion, judicial discretion, and supervision conditions, making these four factors important determinants of which probationers experience a revocation. Other processes, such as law enforcement practices (which could detect more or less probationer misconduct) or policies and statutes (which could limit discretion in responding to probation violations), also play a role in many jurisdictions.

Given these limitations, conclusions regarding the drivers of observed disparities in probation revocations are provisional and constrained by the data available for this study.”

Finally, let’s examine claims of racism in the most severe form of sentencing: the death penalty. A 2014 study analyzed capital punishments from 1981 until 2014 in the state of Washington, finding that African-American defendants were 4.5 times more likely to receive the death penalty than “similarly situated white defendants.” Once again, certain potentially relevant factors were not accounted for:

“Although the trial reports ask judges to supply information about a wide range of case, defendant and victim characteristics, we discovered through the coding process that many of the trial reports were incomplete. We were therefore unable to include a number of potentially relevant factors (such as defendant IQ and mental health status) in our analyses that may also influence the administration of capital punishment.”

This chapter in Handbook of Psychology, Second Edition gives a good overview of the impact the factors unaccounted for in the study have on whether a defendant receives the death penalty.

A 2006 Association for Psychological Science research report found that “defendants who were perceived to be more stereotypically Black were more likely to be sentenced to death only when their victims were White” than when their victim was of the same race. The authors offer a possible explanation for this disparity:

“One possibility is that the interracial character of cases involving a Black defendant and a White victim renders race especially salient. Such crimes could be interpreted or treated as matters of intergroup conflict. The salience of race may incline jurors to think about race as a relevant and useful heuristic for determining the blameworthiness of the defendant and the perniciousness of the crime.”

I am inclined to view the report’s results and explanation with skepticism for one reason: its understood premise — that the death penalty is far more likely to be received by an offender when the offender’s victim is white (known as the “white victim effect”) — has been found to be unsubstantiated by a more recent and enormously more thorough study.

The analysis, conducted in 2014, utilized data from more than 1,000 death penalty cases in North Carolina over nearly forty decades and controlled for 50 legal and extralegal factors, making it “one of the largest and most extensive studies in the literature looking at the ‘White victim effect’ as it relates to juror’s decisions to recommend the death penalty.” Its conclusions were twofold:

1. “After rigorously accounting for up to 50 legal and extralegal confounders, we found no significant or substantive differences in the likelihood of receiving the death penalty for cases with White victims (and defendants of any race) compared to cases with Non-White victims (and, again, defendants of any race).”

2. “After balancing cases on a myriad of potentially confounding legal and extralegal factors and accounting for the type of aggravators and mitigators accepted by the jury in addition to the number of aggravators and mitigators accepted, we found no statistical or substantive evidence that cases with White victims were more likely to result in the death penalty compared with cases involving similarly situated Non-White victims, even when the defendant is Black.”

Thus, the notion that the race of the victim is relevant to whether a defendant receives the death penalty is found to be unsubstantiated when a myriad of legal and extralegal factors are accounted for.

Until studies on racial disparities in convictions and sentences account for a multitude of relevant legal and extralegal factors that could very likely cause the disparity, racism cannot be used as the explanation. As demonstrated in the last study I discussed above, up to 50 factors may have to be considered until racial disparities disappear, revealing the complexity of the criminal justice system and the unlikelihood that racism is the reason for the disparity

Wrongful Convictions and Exonerations

This final section will be dedicated entirely to analyzing a 2017 report by the National Registry of Exonerations, as it appears to be the most thorough study on racial disparities in wrongful convictions and the origin for many prominent claims. The paper focuses on the three crimes “that produce the largest number of exonerations in the Registry”: murder, sexual assault, and drug crimes.

The report finds that “innocent black people are about seven times more likely to be convicted of murder than innocent white people,” that “a prisoner serving time for sexual assault is three-and-a-half times as likely to be innocent if he is black than if he is white,” and that “innocent black people are about 12 times more likely to be convicted of drug crimes than innocent white people.”

At the end of the report, the authors provide potential reasons for these tremendous disparities. Let’s look at each one:

  • The high homicide rate in the African American community. This is undoubtedly a reason — the more crime that occurs in a specific community, the more likely wrongful arrests will occur.
  • The risk of eyewitness misidentification in cross-racial crimes. Again, a very real possibility. The report notes that “most African American sexual assault exonerees were misidentified by white victims.” This is not a fault of the criminal justice system, though, but rather the nature of interracial crime.
  • Race-of-victim disparities. The report writes that even when accounting for errors in eyewitness identification, “investigations of murders in which African Americans killed white victims are less accurate than other murder investigations.” However, I do not believe this is indicative of racism — as I discussed in the section “Convictions and Sentences” above, the notion of a “white victim effect” has been found to be unsubstantiated by one of the largest and most thorough investigations into racial disparities in death sentences. Thus, until studies on the same scale also examine the “white victim effect” in murder, sexual assault, and drug convictions, I am inclined to believe that confounding variables — such as the type and number of aggravating and mitigating factors accepted by juries — may create the illusion of a “white victim effect.”
  • African Americans are more often stopped, questioned and searched than whites. I analyzed this phenomenon in the section “Police Stops and Searches” above.
  • Black suspects and defendants are more likely to be the targets of police and prosecutorial misconduct. Here the report discusses both individual exonerations and group exonerations:
    • Individual Exonerations – The report finds that “official misconduct occurred in fewer than two-thirds of murder exonerations with white defendants but more than three-quarters of those with black defendants.” It notes that the most common types of misconduct are the concealment of exculpatory evidence, witness tampering, and perjury by a state official.
      • The report, however, does not seem to control for very important factors — such as city and county — when comparing white defendants to black defendants. Cities and counties that are majority African-American could, just by happenstance, have a more corrupt and negligent police force, thus resulting in more blacks being wrongfully convicted due to police misconduct. This, by itself, does not suggest racism.
    • Group Exonerations – The report defines a group exoneration as “the exoneration of a group of defendants who were falsely convicted of crimes as a result of a large-scale pattern of police perjury and corruption.” The report finds 15 group exonerations from 1995 until 2017, some of which clearly involved racist police officers.
      • Fortunately, racially motivated police officers are exceptions, not the rule. Out of more than 800,000 law enforcement officers in the United States, there will assuredly be an infinitesimal minority that will be motivated by racial animus — that is not debatable. What is up for debate is whether an indelible, anti-black bias pervades the criminal justice system as a whole.
  • African-American exonerees spent more time in prison before they were released than did white exonerees. The authors of the report do not come to a conclusive explanation as to why this is the case, and yet — without examining the many other factors that could be causative for the disparity — write that “our best guess is that black sexual assault defendants who insisted on their innocence and refused to plead guilty were punished more harshly for doing so than innocent white defendants who followed the same course.” I could guess with equal confidence that factors such as socioeconomic conditions and offense severity cause the disparity, but if I too lacked evidence to support my guess, it would be as empirically useless as the guess that racism is the causation. For more of my analysis of racial disparities in sentences, see the section “Convictions and Sentences” above.
  • Many innocent black defendants encounter bias and discrimination throughout their ordeals. The report includes multiple cases in which a police officer clearly demonstrated racial bias against an innocent African American. Again, these are anecdotes, not empirical findings, and thus should not be construed as indicative of the aggregate criminal justice system.

Thus, out of all the reasons given by the report, it appears that the first two are the most likely to be causative of the racial disparity in wrongful convictions. Those reasons have nothing to do with the justice system itself but are rather a product of disproportionate crime rates and the nature of interracial crime — i.e., victims are more likely to commit a misidentification error when their offender is a stranger or of another race. Other factors must be accounted for in order for the other reasons to be indicative of racism in America’s criminal justice system.


I wrote this post for two main reasons:

1. To determine if claims of institutional racism in America’s criminal justice system can be substantiated.

2. To provide an in-depth analysis of academic studies on racial disparities in the criminal justice system that are frequently cited but are rarely examined.

I have strived to be as objective as possible in my analyses. There were several studies I debated as to whether I should mention — such as this one by a Harvard economist that found no bias in police shootings — but critiques of their methodologies prompted me to exclude them, even though they may have supported my overall conclusions.

My overall takeaways are:

  • By looking at the raw data of reported offenders and arrestees, the claim that African Americans are overpoliced is shown to be unfounded.
  • Similarly, the claim that African Americans are unjustly killed by police at a higher rate than whites is unfounded. Both the raw data and academic reports support this.
  • Factors such as purchasing method and the initial reason for arrest need to be accounted for in order to determine if racism is the primary reason for the racial disparity in drug arrests. I have been unable to find any studies that have controlled for these factors.
  • Likewise, there is currently insufficient evidence to indicate racial bias in police stops and searches. Many relevant factors that are unaccounted for in these studies — such as race-specific differences in driving behavior and the type of search — are as potentially causative for the racial disparity in stops and searches as racism is.
  • The gender of a defendant, not the race, appears to be the main factor in the perceived racial disparity in charge bargaining and bail setting. When not coding for gender, the racial disparity in bail setting disappears and is minimal in charge bargaining.
  • Racial bias appears to affect both white and black potential jurors in jury selection. Black mock jury members appear to demonstrate a larger in-group and out-group bias than white mock jurors.
  • Many oft-cited studies concerning racial disparities in convictions and sentences do not account for numerous relevant legal and extralegal factors. The so-called “white victim effect” is shown to be unfounded when controlling for these factors.
  • The two most likely reasons for the racial disparity in wrongful convictions are the disproportionate crime rates among the African-American population and the nature of interracial crime. Other reasons involve assumptions and anecdotal evidence.

On a broader note, reading dozens of studies on this topic has also shown me just how misleading citations of them can be. Virtually every study that appears to support the notion that America’s criminal justice system is racist warns of the limitations of its results and the numerous factors that weren’t accounted for in its methodology. Some, such as this one, even warn readers to not construe the findings as evidence of racial bias. This does not deter popular news websites and social media accounts from cherry-picking a few quotes from them to support a narrative, however. After all, who will actually spend the time to read a boring, 40-page report in its entirety?

In today’s social media age, there is a burdensome conflict between our desire to receive information instantly and easily and the nuances and complexities of the truth. The former will always win over the latter, as seen in the ability of colorful infographics and emotional social media posts to stimulate the youth’s eagerness for virtue and purpose. The desire for conformity and an objectively moral goal in a morally relativistic society causes teenagers and millennials on Instagram to post a picture of a black square, believing they are fighting a noble cause. It’s in many ways a new religion, its followers proselytizing pliable Americans and excommunicating those who dare to resist. But if the dogma of the religion — that America’s justice system is fundamentally racist — is rootless, how long will it be until its disciples abandon their aimless mission?

Daniel Schmidt is a 16-year-old paleoconservative political commentator and opinion writer. In his freshman year of high school, he founded The Young Pundit, a hard-hitting political commentary outlet that features young American nationalists critical of both the Left and the Right. To read more about Daniel, click here.

1 comment

Comments are closed.