One gene which plays a role in Black’s high crime rate is the monoamine oxidase -a gene (MAO-A). This gene produces an enzyme by the same name. The enzyme MAO-A breaks down a class of neurotransmitter called mono-amines in the brain. These neurotransmitters include ones which are well known to effect behavior such as dopamine and serotonin. Some versions of the MAO-A gene lead to lower levels of MAO-A the enzyme and, therefore, more mono-amine activity in the brain.
The molecular difference between different versions of the MAO-A gene lies in the number of times a certain section for its promoter region is repeated. Three repeat alleles lead to lower than average levels of the enzyme MAO-A. The two repeat allele leads to even lower levels of MAO-A.
Beginning in the 1990’s, researchers began to suspect that variation in MAO-A levels might be related to traits like aggression and criminal behavior. This suspicion arose following a study on the particular family which carried a rare mutation that lead to extremely low levels of the enzyme MAO-A. This family had rapists, arsonists, and other violent criminals (Brunner et al. 1993).
Later, geneticists raised rates with an artificial version of the MAO-A gene which lead to low MAO-A production. It was found to cause the rats to act far more aggressive than is typical (Cases et al. 1995, Scott et al. 2008, Vishnivetskaya et al. 2007, and Mejia 2002).
The first direct link between common MAO-A alleles and violence in humans came from Caspi et al 2002.
This study found that people with either the 2R or 3R version of the MAO-A gene tended to be more aggressive than average, but only if they had difficult upbringings. Based on these results, it was hypothesized that low activity MAO-A genes might cause people to have stronger reactions to negative experiences than people with high activity versions of the gene. Since then, a meta-analyses has confirmed that this interaction between MAO-A genotype and childhood trauma has been well replicated in male subjects (Bryd and Manuch 2013).
Since then, researchers have found that low activity MAO-A also predict higher levels of anti-social personality even among those who have not experienced child abuse (Ficks and Waldman 2014). These versions of the gene have also been linked to high levels of credit card debt and obesity (Camarena et al. 2004, Neve and Flower 2009, Need et al. 2005, and Feummeler 2008). Low activity versions of the gene have also been shown to significantly increase the risk that someone will end up in prison (Roux 2014, Vaughn et al. 2009, Beaver et al. 2009, and Beaver et al. 2013). Given this, it seems plausible that low activity versions of the MAO-A gene leads to increased impulsivity, likely due to increased dompanie activity, which in turn leads to an elevated risk for criminal activity, aggression, obesity, and credit card debt.
Racial differences in MAO-A genotypes are well established. Below you can see a review of the literature in the United States. The vast majority of the research shows that Blacks are more likely than Whites to carry both versions of the low repeat allele.
|Study||Sample Size||Sample Type||Sex||Black2R %||White 2R %|
|Beaver et al. (2013)||2574 Whites164 Blacks||Nationally Representative Sample (ADD Health)||Male||5.2||0.1|
|Widom and Brzustowicz (2006)||180 Whites98 Non-Whites||Abuse victims||Male||6.12||0.6|
|Widom and Brzustowicz (2006)||196 Whites143 Non Whites||Abuse Victims||Female||2.6||7|
|Reti et al. (2011)||365 Whites232 Blacks||Psychiatric Patients||Both||4.7||.5|
|Choe et al. (2014)||106 Whites86 Blacks||N/A||Male||6.0||0|
|Study||Sample Size||Sample Type||Sex||Black 3R%||White 3R%|
|Beaver et al. (2013)||2574 Whites164 Blacks||Nationally representative sample (Add Health)||Male||52.3||37|
|Widom and Brzustowicz (2006)||180 Whites98 Non-Whites||Abuse victims||Male||35.7||41.1|
|Widom and Brzustowicz (2006)||196 Whites143 Non-Whites||Abuse victims||Female||74.1||62.8|
|Rosenberg et al. (2006)||274 Whites37 Blacks||N/A||Male||54.1||36.1|
|Reti et al. (2011)||365 Whites232 Blacks||Psychiatric patients||Both||48.7||34.1|
|Choe et al. (2014)||106 Whites83 Blacks||Low income||Male||42.2||30|
|Sabol, Hu, and Hamer, (1998)||539 Whites52 Blacks||Various combined samples||Both||59.1||33.1|
In summary, Blacks have higher levels of low repeat MAOA alleles than Whites do, these differences constitute strong evidence that genetics plays a role in racial differences in aggression, self control, and crime.
Poverty and Single Motherhood Probably Don’t Cause Crime
Though it is true that poverty and family structure correlate with crime, neither has been shown to cause crime. A non-causal relationship between crime and these variables is suggested by the fact that changes in a population’s degree of poverty and single motherhood overtime do not predict change’s in that population’s crime rate in the way that you would expect.
For instance, looking at the last 50 years of American history, we see that crime actually tended to decrease when the poverty rate increased (1).
This evidence coheres well with a review of 28 studies in the Handbook of Crime Correlates which found that the majority of studies showed that crime actually rises when the economy improves.
Thus, most of the longitudinal data available does not support the notion that poverty causes crime. When more people become poor, crime does not rise. This suggests that the correlation between poverty and crime is best explained by the view that there is a certain segment of the population which tends to be poor and criminal not because one variable causes the other but, rather, because some other variable, or set of variables, causes people to be both poor and criminal.
Unfortunately, I am not aware of much high quality data on the longitudinal relationship between family structure and crime. However, if we look at how these two variables changed between 1960 and 2009 we clearly do not see strong support for a causal relationship. Luscombe (2012).
In order to address the lack of good evidence on this question, we analyzed data from the FBI and the census on how violent crime, family structure, poverty, and racial demographics changed in 197 American cities between the years 2006 and 2014. (See notes 3 and 4 for data links).
As can be seen below, changes in the proportion of a city that was Black was the only significant predictor of changes in violent crime. Moreover, both single motherhood and poverty were negatively correlated with violent crime over time, meaning that they tended to decrease when when crime rose and rise when crime fell.
Thus, the relevant longitudinal evidence does not seem to support the contention that differences in family structure, or poverty, cause differences in crime.
Race Predicts Crime After Controlling for Poverty and Family Structure
Even if family structure and poverty do have a causal role in crime, and therefore do contribute something to the Black/White crime gap, they do not explain anywhere near the totality of the gap. We know this because the proportion of an area which is Black continues to predict its crime rate even after regional differences in family structure and poverty are controlled for.
Land, McCall, and Cohen (1990) collected data on the homicide rates of cities, standard metropolitan statistical areas (SMSAs), and states for the years 1960, 1970, and 1980. In each year they included all 50 states and every city and SMSA included in the census. They then looked at how well the following 11 variables predicted crime variation between these areas: population size, population density, percent black, percentage aged between 15 and 29, percent divorced, percent of kids without two parents, median family income, the poverty rate, income inequality, the unemployment rate, and whether or not the city/SMSA/State was in the south. All of these variables were entered into a single regression model, meaning that the estimated effect size for each variable held all other 10 variables constant. This analysis thus produced 9 total models explaining crime variation in cities, SMSAs, and states, across 3 decades. In all but one model, the proportion of an area that was Black continued to have a strong relationship with crime even after the other 10 variables in the study were held constant.
Similarly, Kposowa, Breault, and Harrison (1995) analyzed crime variation across 2,078 U.S counties and found that the proportion of the county that was Black continued to predict crime even after controlling for county differences in poverty, divorce rates, income inequality, religiosity, population density, age, and education.
In the first analysis, using data from the Census and the FBI, data on the racial demographics, violent crime, poverty and education levels, and family structure of 306 American cities in the year 2014 was analyzed. Violent crime rates were put through a log transformation to correct for non-normality. The percent of a city that was Black continued to predict its violent crime level after holding poverty, education, and family structure constant (2).
Similarly, in the second analysis data on 231 American cities in the year 2006 was analyzed. Once again, the proportion of a city that was Black continued to predict its crime rate even after controlling for city differences in poverty, education, and family structure (3).
In both models, the introduction of these multivariate controls reduced the bivariate relationship between % Black and violent crime by about half. Some might interpret this to suggest that half of the Black/White crime gap can be causally explained by single parenthood and poverty. However, this would be incorrect. % Black’s association with crime may be reduced by controlling for poverty and single motherhood because these variables cause crime, because these variables are caused by crime, or because Blacks tend to be poor, criminal, and from single parent homes due to some confounding variable. Thus, these analyses only provide us with an estimate of the maximum proportion of the relationship between race and crime which is attributable to single parenthood and poverty, and that maximum estimate is half.
Another source of evidence comes from the relationship between racial differences in wealth and crime over time. Over the 20th century, the Black/White wage gap has lessened while the Black/White crime gap has widened (4).
Finally, Zaw and Darity (2016) looked at how crime varied among individuals of different races and varying amounts of net wealth. As can be seen, the results of this analysis also cast serious doubt on the idea that poverty can explain racial crime gaps.
Given the uniformity of the results across these analyses, we can say with a reasonable degree of confidence that a significant proportion of the relationship between race and crime cannot be explained by poverty or family structure.
Racial Differences in Family Structure, Crime, and Poverty are geographically and temporally consistent
Talking about gangs as a cause of Black crime rates we can find some agreement: it is almost certainly true that Blacks would be less criminal if they were not involved in gangs. However, this proposition naturally leads us to wonder why it is that Blacks are so involved with gangs in the first place.
Some would likely say that this has to do with poverty and family structure but, as we’ve already seen, these variables cannot explain the Black/White crime gap.
It would also be reasonable for us to wonder why it is that Blacks are so much more likely to have children outside of marriage and to be poor. Neither of these facts about American life are new, nor are they exclusively American. The same can be said of crime.
If we look back in American history we can see that, relative to African Americans, Whites have always, though to varying degrees, been richer:
Less likely to have a child outside of marriage:
And more likely to be criminal (5):
This mirrors what we see around the world. Black nations have higher rates of both crime and poverty than White and East Asian nations do:
Similarly, African nations have the highest rates of single parent households:
As it turns out, the explanation for why Blacks, around the world, have such high crime rates also explains part of why Blacks, around the world, have high rates of poverty and single parenthood.
The Alternative Hypothesis
This Alternative Hypothesis is that, for partially genetic reasons, Blacks are more likely than Whites to exhibit various personality traits which in turn causes them to have high rates of crime, poverty, and single parenthood.
Those unfamiliar with behavioral genetics may be surprised by such an explanation because they are uncomfortable with the idea that behavioral differences between people have a genetic basis. However, twin studies done over several decades have shown that variation in virtually all human traits is partly caused by genetics (Polderman et al. 2015). Traits studied in these studies include criminality, income, and family structure.
|Violent Crime||55%||Frisell et al. (2012)|
|Anti-social personality||50%||Mason and Frick (1994)|
|Income||42%||Hyytinen et al. (2013)|
|Divorce||32%||Jerskey et al. (2010)|
Others will be uneasy with the idea that there are genetic differences between the races. We all recognize genetic differences in non mental traits such as height and skin color. For political reasons, we resist applying this same thinking to mental traits. From a genetics standpoint, this inconsistency is even more problematic that it might at first seem because the genes that differ the most between races are ones involved in the brain.
With that being said, let’s start with racial differences in personality traits related to crime and then work our way back to genes.
Racial Differences in Aggression
The personality trait most obviously associated with crime is aggression. There are several lines of evidence suggesting that Blacks are more aggressive than Whites. First, Blacks are more likely than Whites to get into fights at school.
Secondly, Blacks are more likely than Whites to bully people at school (Wang, 2013).
Thirdly, the more violent a crime is the more heavily African Americans tend to be overrepresented in it.
Thus, there is good reason to think that Blacks are more aggressive than Whites are.
Racial differences in self control
Blacks are also more impulsive than Whites are. Self control is experimentally measured by having people choose between a reward they can get now and a larger reward they can get later. People who choose to receive a larger reward later have a greater ability to delay gratification and, so, more self control.
Such studies have shown that low self control predicts criminality and poverty even after controlling for IQ and parental socio-economic status (Moffitt et al. 2010). Moreover, racial differences in self control have been found many times. For instance:
Michel (1958) conducted the first experiment on self control on a sample of 53 children aged 7-9 living in Trinidad. He found Black children had less self-control than Asian children.
Herzberger and Dweck (1978) looked at a sample of 100 4rth grade American school children and found that Blacks had lower self-control than Whites even after controlling for socio-economic status.
Warner and Pleeter (2001) took advantage of a semi-natural experiment which came about due to the military. In the mid 1990’s the U.S Government offered sufficiently experienced military personnel two options when they retired: they could take a large lump sum of money now or agree to get a yearly payment from the military for the rest of their lives which, over time, would add up to far more than the lump sum. Warner and Pleeter were able to find data on the choices of 66,000 individuals and found that Blacks were 15% more likely than non-Blacks to take the lump-sum while.
Wany, Rieger, and Hens (2011) utilized a sample of 5,291 university students from 45 countries and gave participants a chance to choose an immediate monetary reward or a larger long term reward. The chart below shows the proportion of people from different regions that chose the larger and less immediate reward:
Castillo, Ferraro, Jordan, and Petrie (2011) had a sample consisting of 82% of the student population of 4 middle schools in a poor Georgia school district. As they write “In our experiment, subjects are asked, orally and in writing, to make twenty decisions in total. For each decision, subjects are asked if they would prefer $49 one month from now or $49+$X seven months from now. The amount of money, $X, is strictly positive and increases over the twenty decisions.” Using this design, they were able to measure at what point people began to prefer the later reward and, thus, the strength of their preference for immediate gratification. Blacks were found to have significantly less self-control than Whites.
Borgo (2013) looked at data on 25,820 American households and found that Black homes had lower savings rates than White homes even after controlling for differences in income, age, family size, education, region of residence, and marriage. Normally, using savings rates as a measure of self-control would be problematic because it is obviously easier to save money if you have a large income, but this study utilized an impressive set of controls and so can be taken as a valid measure of self-control.
Finally, Andrade and Petry (2014) looked at a sample of 317 individuals with gambling problems and found that White gambling addicts had more self-control than Black gambling addicts even after controlling for education, drug problems, and income.
Thus, racial differences in self control have been shown to exist across many studies. Said differences exist around the world, in children and adults, and cannot be explained by racial differences in socio-economic status. These differences likely contribute to racial differences in crime, poverty, and non-marital birth rates.
Racial Differences in Psychopathic Personality
Blacks also score higher than Whites on measures of psychopathic personality. Lynn (2002) described a common measure of psychopathic personality, the Psychopathic Deviate Scale:
“This was constructed by writing a number of questions, giving them to criterion groups of those manifesting psychopathic behaviour and ‘‘normals’’, and selecting for the scale the questions best differentiating the two groups.
The criterion group manifesting psychopathic behaviour consisted of 17–24 year olds appearing before the courts and referred for psychiatric examination because of their ‘‘long histories of delinquenttype behaviours such as stealing, lying, alcohol abuse, promiscuity, forgery and truancy’’ (Archer, 1997, p. 20).
The common feature of this group has been described as their failure to ‘‘learn those anticipatory anxieties which operate to deter most people from committing anti-social behaviour’’ (Marks, Seeman, & Haller, 1974, p. 25).
The manual describes those scoring high on the scale as follows: irresponsible, antisocial, aggressive, having recurrent marital and work problems, and underachieving (Hathaway & McKinley, 1989).
A number of subsequent studies have shown that the Psychopathic Deviate scale differentiates delinquents and criminals from nondelinquents and non-criminals (e.g. Elion & Megargee, 1975).”
Lynn then reviewed 5 studies comparing racial groups on this measure. In White, Asian, and African countries it was found that Blacks scored higher than average in psychopathic personality:
Two later meta-analyses on Black/White differences in psychopathic personality came to the conclusion that Blacks did have higher levels of psychopathy than Whites but that this difference, while statistically significant, was not large enough to be practically significant (Skeem et al., 2004; McCoy and Edens, 2006). However, whether or not this difference is practically significant on its own tells us nothing about whether or not it is one among the many factors which contributes to racial crime disparities. Further more, all the samples used in these meta-analyses were either clinical or correctional in nature. Because of this, the reported mean differences in these studies may be significantly less than what would be found were the sample taken from the general population.
In sum, differences in psychopathic personality likely make some contribution to racial differences in crime, poverty, and non-marital birth rates.
Racial Differences in Stress and Anxiety
Another feature shared by both Blacks and criminals is low levels of stress. Criminologists have explained the association between low levels of stress and crime by positing that many people would be deterred from committing a crime even if they wanted to by the level of anxiety, stress, and nervousness, that attempting to commit said crime would cause, which is why having a less active stress response would predispose people towards criminality.
Having a slow heart beat and a low level of cortisol are both biological markers of this phenomena which have been independently linked to criminality and shown to be more common among Blacks than among Whites. (Aggarwal, 2013; Easton, 2000; “Race and Stress“; Murray et al., 2016; Higginbotham et al., 1991; Gillum, 1988). Blacks also self report generally having lower levels of stress than White people do.
Masi et al. (2004)
In conclusion, several aspects of Black personality, including higher levels of aggression, impulsivity, and psychopathy, as well as lower levels of stress, likely predispose them to crime.
With this understanding of the proximate causes of Black criminality in place, let’s now turn to analyzing more distal or ultimate causes.
Racial Differences in Testosterone
We’ve already seen that the hormone cortisol may play a role in racial crime differences. Blacks have lower levels of cortisol and this, in turn, makes them have lessened biological stress responses. Low levels of cortisol may also predispose Blacks towards crime because cortisol blocks the action of the hormone testosterone. Having less cortisol means that testosterone will be more active in Blacks.
Testosterone activity has been linked many times to aggression and crime. Meta-analyses show that testosterone is correlated with aggression among humans and non human animals (Book, Starzyk, and Quinsey, 2001). Women who suffer from a disease known as congenital adrenal hyperplasia are exposed to abnormally high amounts of testosterone and are abnormally aggressive. Artificially increasing the amount of testosterone in a person’s blood has been shown to lead to increases in their level of aggression (Burnham 2007; Kouri et al. 1995). In fact, a study of Rhehus monkeys found that injecting female fetuses with testosterone caused them to behave just as aggressively as young males (Book, Starzyk, and Quinsey, 2001). Finally, people in prison have higher than average rates of testosterone (Dabbs et al., 2005). Thus, testosterone seems to cause both aggression and crime.
To return to race, Richard et al. (2014) meta-analyzed data from 14 separate studies and found that Blacks have higher levels of free floating testosterone in their blood than Whites do. Thus, both testosterone levels and cortisol’s interaction with testosterone likely contributes to crime.
Exacerbating this problem even further is the fact that Blacks are more likely than Whites to have low repeat versions of the androgen receptor gene. The androgen reception (AR) gene codes for a receptor by the same name which reacts to androgenic hormones such as testosterone. This receptor is a key part of the mechanism by which testosterone has its effects throughout the body and brain.
All versions of this gene have several repeats of a CAG nucleotide sequence. Variants between people differ in the number of these repeats. Versions of this gene with fewer than average numbers of repeats lead to increased androgen activity and has been associated many times with crime and aggression. For instance:
Rajender et al. (2008) analyzed data on 645 men including 241 convicted rapists, 107 convicted murders, and 26 men convicted for both rape and murder. They found that those incarcerated for both rape and murder had the shortest average number of CAG repeats followed by murderers, followed by rapists. The non criminal controls had the longest CAG repeats.
Butovskaya et al. (2013) conducted a study in Tanzania in which an association between anger, aggression, and AR genotype was found among 138 adult men.
Commings et al. (2003) found an association between AR genotype and the probability that a person’s parents were divorced and that their father was present in the first 7 years of life. They also found the expected correlations between AR genotype and agression, but, possibly due to small sample size, they were not statistically significant
Similarly, Pichard et al. (2007) found that AR genotype predicted anti-social personality in a sample of 1,007 men
Hurd et al. (2010) found a significant correlation between AR genotype an anger and physical aggression, and insignificant but positive relationships with verbal aggression and hostility in a sample of 188 men.
Jonsson et al. (2001) found non-significant relationships with aggression and impulsivity in a sample of 335 Swedes.
Cheng et al. (2006) study found that violent criminals were more likely than controls to have short CAG repeats. However, the difference in means, while in the expected direction, was not statistically significant.
What we see in this literature then is a mix of positive and significant effects and positive and insignificant effects for the relationship between shorter AR genes and aggression/crime. This distribution of effect sizes strongly suggests that the true mean effect is real and positive. On a theoretical level, this finding makes a great deal of sense in light of the role of testosterone in aggression and crime and what we know about the function of the AR gene.
Four separate studies have shown that low repeat alleles of the AR gene are more common among Blacks than among Whites (Irvine et al 1995, Wang et al. 2013, Bennet et al. 2002, and Shibalev et al. 2013). Given this, it is highly likely that low repeat AR alleles increase Black crime rates by increasing the impact of testosterone. Due to testosterone’s impact on sexuality, these same mechanisms may also explain why Blacks have such high rate of single parenthood. This is especially likely in light of the fact that, as seen above, low repeat AR alleles predict family structure.
Racial Differences in IQ Related Genes
Racial differences in IQ related genes are also relevant to this discussion. There is a significant genetic correlation between IQ and self control, meaning that the same genes which explain variation in self control also explain variation in IQ (Polderman et al., 2009). Moreover, self control and IQ are positively correlated (Shamosh and Gray, 2008). In conjunction, these facts should lead us to suspect that genes which predispose people towards low IQ will often also predispose them towards low self control.
Piffer (2015) was the first major attempt to compare the frequency of IQ related genes across the races. It was found that, for every gene examined, Whites were more likely than Blacks to possess the allele associated with higher intelligence.
Given this, it seems likely that the genes which predispose Blacks towards low IQ also predispose them towards low self control which in turn predisposes them towards crime, poverty, and single parenthood.
A Unifying Explanation
It is not my view that genes explain the entirety of the Black/White crime gap. Environmental variables, such as lead exposure and child abuse rates, likely play a role too. However, I doubt that these variables explain as much of the Black/White crime gap as genes do.
A genetic explanation has the benefit of offering an extremely parsimonious explanation for a wide range of data. Racial differences in crime are seen around the world. As I noted earlier, Black nations have higher crime rates than European nations do. Moreover, official crime statistics have replicated the African/White crime disparity in many nations all over the earth including, but not limited to, Canada, Britain, Australia, France, Israel, Brazil and Japan. (Ellis, Beaver, and Wright, 2009; Macimo, 2004; Ministry of Justice, 2012; Murray et al., 2013; Ray, 1985).
The racial crime gap is also known to start early in life. This can most obviously be seen by looking at juvenile arrest rates by race:
Further evidence of this comes from data on school punishments. Getting in trouble in school is similar to getting in trouble with the police, both involve rule breaking and, often, violence. The Department of Education has repeatedly shown that Blacks are far more likely than Whites to get in trouble even as early as preschool.
Further still, as already noted, Blacks are more likely than Whites to bully and get into fights in school.
As also already noted, the racial crime gap has existed for a long time. It even shows up in studies using data from the 19th century (Ellis, Wright, and Beaver 2009).
This general picture fits a genetic explanation well. It is hard to see how culture could cause a difference that follows a population through time and around the world. (Unless, of course, this cultural difference itself had genetic origins). It is even harder to think of a more parsimonious explanation for Black crime. This is especially true given that, in light of the genetic evidence presented above, total egalitarianism about race and crime is not a reasonable position to take.
This hypothesis, that Blacks are genetically predisposed towards violence and low self control, also explains why poverty and weak families are present among Africans all over the world. It also explains why people don’t become more criminal when the economy tanks. After all, the economy tanking doesn’t magically change people’s character. However, there will always be an aggressive and impulsive fraction of society which, due to these personality traits, are poor and are having children out of wed lock. This segment of society will then cause correlations between these variables when you look at the total population.
In conclusion, we do not think that poverty and family structure play a very huge role in racial crime differences. Even if they do play some role, a large difference remains when their influence is statistically controlled for and this is an estimate of the largest possible contribution they could have.
The immediate cause of racial differences in crime significantly consists of racial differences in aggression, anxiety, and self control. These differences, in turn, are probably partially caused by genetic differences between Blacks and Whites. Other environmental variables are likely involved as well, but this post have concentrated on genetic causes because they are the causes most often ignored.
2. Crime data was taken from the 2014 Uniform Crime Report Table 8 and all other data came from the Census Fact Finder. The data came from 1 year estimates of the American Community Survey. “Geographies” was set to “All places within the United States”
3. Crime data was taken from the 2006 Uniform Crime Report Table 8 and all other data came from the Census Fact Finder. The data came from 1 year estimates of the American Community Survey.“Geographies” was set to “All places within the United States”
6. Data was taken from the World Bank for the year 2014. The regions were defined by the World Bank.