Last Updated on July 20, 2022

In a previous post, I argued that systemic racism and socioeconomic status (SES) were poor explanations of disproportionate levels of black crime. I concluded with the claim that the intelligence gap (or IQ gap), not the SES gap, is the primary cause of the disproportionate levels of black crime. In other words, the IQ gap is the primary cause of the disparities in crime between blacks and whites. I will defend this claim in the first half of this post. Another significant cause of the black-white crime gap seems to be differences in family structure between blacks and whites, e.g. racial differences in the rates of single-mother households and out-of-wedlock birth rates. I will explore the importance of family structure in the second half of the post.

Differences in IQ

I believe the IQ gap, not the SES gap, is the primary cause of the black-white crime gap for the following reasons:

1. The mean IQ for blacks is substantially lower than the mean IQ for whites.
2. Low-IQ individuals are significantly more likely to commit crimes than high-IQ individuals.
3. The black-white IQ gap explains the majority of the black-white crime gap.
4. The black-white IQ gap explains the majority of the black-white SES gap.
5. The black-white SES gap does not explain much of the black-white IQ gap.

1. The mean IQ for blacks is substantially lower than the mean IQ for whites

Research on race and IQ has fairly consistently shown that the average IQ score for Blacks in the US is about one standard deviation (about 15 points) lower than the average IQ score for Whites (Neisser et al. (1996) [archived], page 93). This is an uncontroversial finding. Studies usually put the average Black IQ slightly above 85 and the average White IQ slightly above 100. I go into more detail on the degree and importance of IQ disparities in a previous post.

2. Low-IQ individuals are significantly more likely to commit crimes than high-IQ individuals

• Gottfredson (1998) [archived] reports that youths with IQ of 75–90 are 7 times more likely to be incarcerated than those with IQ of 110–125 (7% vs 1%, page 28). She also notes that ‘‘no other trait or circumstance yet studied is so deeply implicated in the nexus of bad social outcomes – poverty, welfare, illegitimacy and educational failure – that entraps many low-IQ individuals and families.”
• Beaver et al. (2013) investigated whether IQ influenced the probability of being arrested and incarcerated. The results revealed a strong negative association between IQ scores and the probability of being arrested/incarcerated for both black and white males (Figures 2 and Figures 3). What’s particularly impressive about this study is that it showed the effect of IQ even after controlling for low self-control and lifetime delinquency. The study concludes “when combined with previous literature, there is good reason to believe that IQ is related to criminal involvement regardless of the sample analyzed, the measurement of crime, and the inclusion of controls for potentially confounding factors, such as executive functions. There is likely not another individual level variable that is so consistently associated with crime and other forms of antisocial behaviors than IQ” (page 285).

3. The black-white IQ gap explains the majority of the black-white crime gap

Murray and Herrnstein (1994) found that, among 29-year-olds, 13% of Blacks had been interviewed in a correctional facility (a proxy for incarceration) compared to 2% of Whites. Among 29-year-olds of average IQ (IQ=100), the number drops to 5% for Blacks and 2% for Whites (page 338). So controlling for IQ reduces the incarceration disparity from 11 percentage points to just 3 percentage points, accounting for 73% of the gap. As shown in a previous post, incarceration disparities track crime disparities very well, so we can use this data to infer that controlling for IQ also reduces the majority of the crime gap.

4. The black-white IQ gap explains the majority of the black-white SES gap

• The Armed Forces Qualification Test (AFQT) is a good proxy for IQ (AFQT scores and IQ scores correlate with r=0.8). After controlling for age and AFQT scores, Johnson and Neal (1998) [archived] found that AFQT disparities explain about two-thirds of the wage gap for young men and all of the gap for young women. The wage gap reduces from 24% to 9% for men after controlling for AFQT. For young women, Black women go from a 17% disadvantage to a 5% advantage after controlling for AFQT scores (page 4).
• Using the same dataset, Jencks and Phillips (1998) [archived] (page 6) report that, after controlling for age and controlling for those with above-average AFQT scores, the average black male earns 96% of the average white male, and the average black woman earns more than the average white woman. Thus, the wage gap is virtually eliminated without even controlling for geography, education, parental socioeconomic status, or even occupation.
• Pew Research released data [archived] showing that about 75% of Whites born into the bottom quintile of income transition out of that quintile, whereas only about 56% of Blacks do the same (figures 3a and 3b). However, among those with a median AFQT score, there is almost no difference in the likelihood of transitioning out of the bottom quintile – 81% for Whites and 78% for Blacks (figure 13). So, the relative lack of economic mobility for Blacks is mostly the result of the relative lack of cognitive ability.
• Murray and Herrnstein (1994) also found that 90% of the wage gap is erased after controlling for IQ. Controlling only for age, the mean black worker’s wages are 80% of the mean white worker’s wages. After controlling for age and IQ, the mean black worker’s wages are 98% of the mean white worker’s wages. On the other hand, controlling for age, education and parental SES (but not IQ), brings the number up to only 86% (page 324).

These are just a few components of the SES gap that are caused by the IQ gap. In a previous post, I provide data showing that the IQ gap also explains black-white differences in education, occupational prestige, and poverty.

5. The black-white SES gap does not explain much of the black-white IQ gap

None of what I said thus far is sufficient to show that the SES gap is not ultimately responsible for black crime. I’ve only shown that IQ differences are statistically associated with differences in crime. But it could be the case that the SES gap is the cause of both the IQ gap and the gap in crime. However, studies suggest that the SES gap is not the cause of the IQ gap. Again, to show this claim, I will copy some points that I made in a previous post:

• From the introductory chapter [archived] of The Black-White Test Score Gap, Jencks and Phillips report the following: a two-year reduction of the black-white education gap among mothers would only reduce the IQ gap by about a point for children (page 22), eliminating Black-White income differences would reduce the IQ gap by about a point (page 23), and marital status had no independent effect on a child’s success (page 23).
• Murray and Herrnstein (1994) show that the BW IQ gap persists across all SES levels, with the gap the largest at the highest SES levels (page 288).
• Rushton and Jensen (2005) [archived] (page 267) report that controlling for SES differences between Blacks and Whites only reduces the IQ gap by about one-third. Thus, SES differences can at most explain about 5 points of the IQ gap.
• In a review of intelligence research by experts in the field, Neisser et al. (1996) [archived] (page 94) reported that “the Black/White differential in test scores is not eliminated when groups or individuals are matched for SES (Loehlin et al., 1975). Moreover, the data reviewed in Section 4 suggest that–if we exclude extreme conditions–nutrition and other biological factors that may vary with SES account for relatively little of the variance in such scores. Finally, the (relatively weak) relationship between test scores and income is much more complex than a simple SES hypothesis would suggest. The living conditions of children result in part from the accomplishments of their parents: If the skills measured by psychometric tests actually matter for those accomplishments, intelligence is affecting SES rather than the other way around. We do not know the magnitude of these various effects in various populations, but it is clear that no model in which “SES” directly determines “IQ” will do.”

So we know that the racial IQ gap is not the result of the racial SES gap. Therefore, the fact that the racial crime gap is mostly the result of the racial IQ gap suggests that the racial crime gap is not the result of the racial SES gap. SES (particularly neighborhood SES) may play some causal role in the crime gap between blacks and whites. For example, the crime rate of low-IQ individuals in low-SES environments (e.g. high rates of poverty, unemployment, and crime) is probably significantly higher than the crime rate of low-IQ individuals in high-SES environments crime. However, insofar as SES is the cause of the crime gap, this is still evidence that the IQ gap is ultimately the cause of the crime gap, because the IQ gap is primarily responsible for the SES gap (excluding differences in family structure, which I explore in the next section).

Differences in family structure

While the IQ gap seems to be the primary cause of the black-white crime gap, I also believe differences in family structure, e.g. rates of single-mother households and out-of-wedlock birth rates, between blacks and whites are also a significant cause of the crime gap. I believe this for the following reasons:

1. Certain family structures are relatively criminogenic (i.e. more likely to result in children that engage in crime).
2. Black people are more likely to have relatively criminogenic family structures.
3. Differences in family structure explain much of the black-white poverty gap.
4. The black-white IQ gap and SES gap does not explain much of the differences in family structure.

1. Certain family structures are relatively criminogenic

Most people are aware that family structure has a strong impact on crime. Specifically, children in families headed by single mothers and children born out of wedlock are more likely to engage in crime. For example, Harper and McLanahan (2004) [archived] shows that children from mother-only households were 3 times as likely as children from mother-father households to be incarcerated (Table 2).

One might say that this is because single-mother households are typically low-SES and low-SES households tend to predict crime. However, several studies find that the association between crime and measures of family structures is stronger than the association between crime and other measures of SES. This suggests that family structure impacts crime independently of SES. Consider the following studies:

• Land, McCall, and Cohen (1990) [archived] reviewed 21 studies and 44 models that investigated the relationship between homicide and various covariates. Covariates related to family structure were found to be more strongly associated with homicides than other measures of SES. Table 1 shows that 100% (8/8) of models found a significant association between homicides and percentage of kids without both parents, and 86% (6/7) of models found a significant association between homicides and percent divorced. For comparison, 86% (38/44) of the models found a significant positive association with percent black, 69% (22/32) of the models found a significant positive association with poverty, 29% (8/28) found a significant positive association with income inequality, 20% (1/5) found a significant negative association with median family income, and 0% (0/9) found a significant positive association with unemployment.
• Kposowa, Breault, and Harrison (1995) found similar results. Table I presents regression results between homicide and other covariates for counties with populations in excess of 100,000 in 1980. The results show that the association between homicides and divorce (Beta=0.187) was stronger than the association between homicide and poverty (-0.119), Gini index (0.177), density (0.126), unemployment (-0.026), and education (-0.027). The only variables with stronger associations with homicides were percent black (0.739) and percent Hispanic (0.265).
• Pratt and Cullen (2005) [archived] meta-analyzed 162 studies that investigated the relationship between a region’s crime rates and 34 variables. 71.53% of studies found a significant positive association between family disruption and crime, with an average effect size of .261 (“family disruption” included “separate categories for percent divorced or separated, single-headed households, and female-headed household” (page 391)). Among the variables with at least 15 studies, only 3 variables had a stronger effect size (incarceration effect, racial heterogeneity, and unemployment with age restrictions had effect sizes of -0.332, 0.293, and 0.283, respectively), and only 2 variables were found to have a significant effect in a larger proportion of studies (racial heterogeneity and social support/altruism were found to have a significant effect in 72.83% and 74.47% of studies). Family disruption had a larger average effect size and was more consistently significantly associated with crime than poverty, inequality, and SES.

Also, Harper and McLanahan (2004) [archived] finds that family structure has a strong impact on incarceration even after controlling for SES:

• Children from mother-only households were still 2 times as likely as children from mother-father households to be incarcerated even after controlling for income, parent education, teen motherhood, minority race/ethnicity, residence in urban areas, regional residence and residence in counties with a high percentage of female-headed households, high unemployment rate, and low median family income (Table 2).
• Children from mother-stepfather and father-stepmother households were actually more likely to be incarcerated than children from single-mother households (Table 2). So clearly the impact of family structure cannot be completely explained by the fact that single-mothers earn less income (since father-stepmother households had comparable incomes to mother-father households, page 381) or the lack of additional parents in the households.
• Interestingly, children from father-only households were not significantly more likely to be incarcerated than children from mother-father households (Table 2).

2. Black people are more likely to have relatively criminogenic family structures

The criminogenic family structures that I mentioned earlier were families headed by single-mothers and families with out-of-wedlock children. These are the family structures in which black children predominantly find themselves:

• According to a 2015 CDC report [archived], 71% of blacks are born to unmarried mothers compared to only 29% of non-Hispanic whites (Table 14).
• A Pew Research Center analysis [archived] of U.S. Census Bureau data showed that 47% of black children lived with a single mother and only 36% lived with married parents. By contrast, only 13% of white children lived with a single mother and 74% lived with married parents.

Much of this is likely due to differences in marital practices between blacks and whites. Compared to whites, blacks are both less likely to marry and are more likely to divorce conditional on marriage:

• A 2017 article [archived] by Pew Research Percentage shows that 54% of white adults ages 18 and older were married compared to only 30% of black adults.
• Raley, Sweeney, and Wondra 2015 [archived] showed that, among women aged 40-44, 34% of black women have never married, compared to only 7% of white women (Table 2). Among women aged 40-44 who have ever married, 53% of black women have experienced an unstable marriage compared to only 41% of white women (Table 2). Lastly, the study also reports “at nearly every age, divorce rates are higher for black than for white women, and they are generally lowest among Asian and foreign-born Hispanic women. Recent demographic projections suggest that these racial and ethnic gaps in marriage and marital dissolution will continue growing.”

The high rate of unintended pregnancies for blacks likely explains the high rate of births to unmarried black mothers. Finer and Zolna (2016) [archived] reported that the unintended pregnancy rate for black women in 2011 (79 per 1000 women) was over twice the rate for white women (33 per 1000 women), and the percentage of black pregnancies that were unintended (64%) was almost twice the percentage of white pregnancies that were intended (38%) (Table 1).

3. Differences in family structure explain much of the black-white poverty gap

Controlling for marital status reduces the majority of the family poverty gap between blacks and whites. Data [archived] published by the US Census in 2019 showed that the poverty rate for all black and white families with children was 25.6% and 9.1%, respectively. However, among married-couple families, the poverty rate black and white families with children was 8.4% and 3.7%, respectively. Thus, the family poverty gap reduces from 16.5 percentage points to 4.6 percentage points after focusing on married couples. In other words, 72% of the poverty gap is eliminated after focusing on married-couple families. This is without controlling for IQ, education, occupation, or even geography.

In a previous post, I argued that the black-white SES gap is a poor explanation of the black-white crime gap. However, it is of course possible that the SES gap may have some effect. Insofar as the SES gap (specifically, the poverty gap) is the cause of the black-white crime gap, this can be attributed largely to the gap in family structure, because the gap is family structure explains a significant degree of the poverty gap.

4. The black-white IQ gap and SES gap does not explain much of the differences in family structure

So controlling for education (one measure of SES) does not reduce much of the disparity in family structure between blacks and whites:

• Raley, Sweeney, and Wondra (2015) [archived] show that blacks were actually more likely to get married in the early 20th century when both the SES gap and the IQ gap were far larger (Figure 1).
• The study also shows that the marital rate for black women aged 40-44 with 16+ years of education was only 71% in 2012, which is lower than the marital rate for any other race-education demographic (Table 3).
• A 2016 article [archived] by Child Trends shows that among births to women aged 20-29 with a bachelor’s degree or higher, 9% of births to white mothers were to unmarried mothers whereas 48% of births to black mothers were to unmarried mothers.

Also, controlling for parental income does not reduce much of the disparity in family structure. Chetty et al. (2018) [archived] find substantial gaps in marital rates for children at every given index of parental income (page 20):

We first document the large intergenerational gaps in marriage rates between black and white children in our sample. Figure IVa plots marriage rates for black and white children in 2015 (between ages 32-37) by parental income percentile. Black children have substantially lower marriage rates across the parental income distribution, with a gap of 32 percentage points (pp) for children with parents at the 25th percentile and 34 pp at the 75th percentile. White children at the bottom of the income distribution are as likely to be married as black children at the 97th percentile of the parental income distribution.

Finally, controlling for IQ also does not reduce much of the disparity in family structure. Murray and Herrnstein (1994) found that:

• Only 20% of the disparity in illegitimacy rates is erased when fixing to persons of average IQ (IQ=100). For 29-year-old mothers, the probability of having a child out of wedlock was 62% for Blacks and 12% for Whites. For persons of average IQ, the number dropped to 51% for Blacks and 10% for Whites (page 331).
• The disparity in marital rates is almost unaffected when fixing to persons of average IQ (IQ=100). For persons over the age of 30, the probability of being married by age 30 is 54% for Blacks and 78% for Whites. For persons of average IQ, the number rises to 58% for Blacks and 79% for Whites (page 329).

So controlling for IQ also does not explain racial disparities in family structure. This suggests that the SES gap does not explain differences in family structure, because (as shown above) controlling for IQ accounts for the vast majority of the SES gap. Likely, differences in family structure are the result of cultural differences on the importance of marriage, particularly marriage before birth.

Cause of differences in IQ and family structure

My post here is agnostic with respect to the cause of racial differences in IQ and family structure. In a previous post, I argued that the majority of the IQ gap seems to be the result of environmental, rather than genetic, factors. However, I don’t know what those environmental factors might be. Also, I’m not sure what explains differences in family structures between blacks and whites (even after controlling for IQ and SES). The differences may be cultural, sure, but it’s a mystery as to why black and white marital practices have diverged substantially since the 1960s. One interesting study by Akerlof et al. (1996) suggests that the rapid rise in out-of-wedlock childrearing is mostly the result of the erosion of shotgun marriages, which was caused by the legalization of abortion and the increased availability of contraception to unmarried women. It is possible that these factors may have been particularly strong for blacks, which caused the rate of out-of-wedlock births for blacks to rise particularly rapidly compared to whites. While this is interesting, an adequate review of the possible causes of modern racial differences in family structure warrants a separate post.

Finally, even if genetics explains a significant portion of differences in the black-white crime gap, it does not follow that the crime-gap cannot be eliminated with sufficient environmental intervention. For example, the heritability of obesity ranges from 40% to 70% (Herrera and Lindgren 2010), suggesting that individual genetic differences explain 40%-70% of individual differences in obesity. This implies that some individuals are predisposed to become obese relative to other individuals. However, despite being predisposed to obesity, these individuals can avoid obesity altogether with proper diet and exercise. We know this to be true because, for example, the obesity rate for non-Hispanic white males in 2012 (35.1%) is nearly 3 times the rate in 1974 (12.5%). For another example, dyslexia is a biological disorder that hinders reading ability. Thus, those with dyslexia are predisposed to have difficulties with reading, i.e. if a dyslexic student and non-dyslexic student are placed in the same educational environment, the dyslexic student will likely have more trouble reading. However, if dyslexic students are placed in specialized educational programs, they can achieve normal levels of reading ability. Likewise, there may be individuals who are predisposed to engage in crime (i.e. they are more likely to commit crimes than an average person if placed in the same environment), and perhaps the frequency of these individuals vary by race, but it may still be true that the individuals can avoid crime with proper environmental intervention.  

Conclusion

In a previous post, I argued that systemic racism and SES were poor explanations of disproportionate levels of black crime. In this post, I have argued that differences in IQ are the primary cause of the black-white crime gap, and that differences in family structure are also a significant cause. These should be treated as two independent causes because black-white differences in family structure are not explained by black-white differences in IQ. I am open to the possibility that some portion of the black-white crime gap may be due to differences in SES (particularly differences in neighborhood SES, as I have explored previously). However, insofar as differences in SES are the cause of black crime, this can ultimately be reduced to differences in IQ and family structure, since differences in IQ explain the vast majority of differences in SES (as I have shown previously) and differences in family structure explain the majority of differences in family poverty rate.

I may come across other data in the future that indicates additional factors (e.g. differences in parenting styles) as a significant cause of disproportionate black crime. If so, I will include the relevant data in a new section of this post.