Testosterone, Race, and Crime

Science you will never hear about.

Thumbnail credit: © Vladimir Chicas/DPA via ZUMA Press
This video is available on Rumble, BitChute, Odysee, and X.

Testosterone is associated with masculine traits. A typical article at PubMed called “Correlation Between Personality Traits and Testosterone Concentrations” finds that high testosterone levels “have been associated with aggression, sexual behaviour and social status.”

High testosterone — or high T — is associated with big muscles and denser bones, and high T men are likely to have low-pitched voices.

Some men like testosterone so much they have tattoos of its chemical structure.

Criminals, especially violent criminals, tend to have high T.

People of different races have different average levels of testosterone, and those levels correspond closely to race differences in crime rates. More T, more crime.

This is not a welcome finding for people who pretend that race doesn’t even exist. However, T level is important in medicine, especially for treating prostate cancer. Here are cancer death rates in the US from 1975 to 2010.

The red line at the top is black men, who are more likely to die of cancer than white men, who are the yellow line below them. There are big differences for prostate cancer. Blacks are 60 percent more likely to get it and nearly 2-1/2 times more likely to die from it. Prostate cancer is influenced by T levels, and if men of different races have different levels, doctors need to know, despite fashionable resistance — even among doctors — to recognizing racial distinctions.

And that’s why there’s a recent article in The Journal of Urology called “Ethnicity and Race Specific Total Testosterone Norms — Time for Change?”

Why time for a change? Because T levels by race are “an area of significant controversy and are critically important in clinical care.”

This is the most careful study of race and testosterone I have ever seen. Some studies measure T levels from saliva samples, but these authors drew blood and used the most accurate method: “isotope dilution liquid chromatography-mass spectrometry certified by CDC Hormone Standardization Program.”

They had a sample of 4,772 men — very large for a study like this — and controlled for body mass index, diabetes, hypertension, coronary artery disease, high cholesterol, and smoking, all of which affect T levels.

Here is a plot of T levels for men of different ages for — from the top — blacks, Hispanics, whites, and Asians.

As you can see, from ages 18 to 49, blacks have the highest levels, followed by Hispanics, then whites, then Asians. The general trend of T levels is down as men age, but to my surprise, this study found that for whites, testosterone increases from their 40s to their 50s. Likewise for Hispanics, and to some extent for Asians — but not blacks. In whites, T levels keep rising into their 60s, when testosterone is even higher than in their 20s. I have never seen such findings before. It doesn’t seem plausible that whites in their 60s have so much testosterone or that Hispanic should have their highest T levels in their 50s. Unfortunately, the authors of this study do not comment on this, and I can’t find any papers that do.

This was not a longitudinal study that followed the same people into old age. Is it possible that older Americans grew up in an environment with far fewer environmental pollutants so that although their T levels did decline with age, they started from a higher level? There are a lot of questions here.

This is a more usual plot of just black and white T levels over time, with blacks in red and whites in blue.

Until the ages of 12 to 15, blacks have lower levels, but then peak dramatically, declining from the early 20s, while whites increase until their late 20s. Then, it’s a steady decline for both races. This study measured T with an electrochemiluminescence immunoassay, which is less accurate than the method of the first study. The sample was smaller — 986 participants — but they were blacks and whites only, with no Asians. The fact remains that from the late teens through the 20s — the most violence-prone ages — blacks have higher T than whites.

What race differences do we find in crime rates? Almost all police reports track blacks and Asians separately and accurately, but many lump Hispanics in with whites. Hispanics have higher crime rates than whites, so this artificially raises what are reported as white crime rates. California, however, tracks whites and Hispanics in entirely separate categories, so it is possible to get clean data.

Table 31 shows felony arrests for 2023 by race.

Arrest rates are not the same as crime rates; some criminals are not arrested, and some people who are arrested are later released or found not guilty. However, in extensive surveys that ask victims of violent crime the race of the perpetrator, the racial proportions reflected in their answers closely match the racial proportions of arrests.

Here are total felony arrests for whites, Hispanics, blacks, and “others.”

The number of Asians included in the “other” category can be calculated from Table 30 of the report. The population of California is approximately 35 percent white, 40 percent Hispanic, 15 percent Asian. With these figures we can calculate what are called odds ratios for other races compared to whites. Hispanics are 44 percent more likely than whites to be arrested for a felony, and blacks are 5.3 times more likely. Asians have only a 27 percent chance of being arrested for a felony, compared to whites.

What about the top three violent felonies: Homicide, rape, and robbery? Again, we can calculate odds ratios. Hispanics are arrested for homicide at 2.8 times the white rate and blacks are arrested at 12.4 times the white rate. For rape, the figures are 2.4 times for Hispanics and 5.4 times for blacks. For robbery, the multiples are two times for Hispanics and 13 times for blacks. All these figures are very much in line with data we have from other states. Asians commit these crimes so seldom they aren’t even included in this more detailed report.

There is not a one-to-one connection between testosterone and violent crime. This article from the NIH Library of Medicine explains:

“Testosterone levels are influenced by conditions that are partly controlled or initiated by the hormone itself, but also by circumstances beyond hormonal or individual control. Different kinds of behaviour are not only subject to influence by environment, but also androgens [that is, male-associated hormones including testosterone] can reinforce the particular kind of conduct and the behavioural impact can wield negative or positive feedback on testosterone secretion.”

In other words, it’s complicated. The article also warns not to draw conclusions about individuals. And there are other associations with crime, such as low IQ and poor impulse control. For example, in the graph we saw earlier, in the late teens and early 20s, whites and Asians have similar levels of testosterone, but Asians are considerably less crime prone. So, it’s not just testosterone.

But what if it were clear that juvenile offenders with high T were overwhelmingly likely to go on to commit serious crimes? I can imagine several responses. “Well, they can’t help it, so we need to give them more social support.” Or, “They are prone to crime, so punish them more severely as a deterrent.”

Another option would be to give sex-criminals hormones that reduce testosterone. There are six US states that permit or even require some form of chemical castration: Alabama, California, Florida, Louisiana, Montana, and Texas. It is usually used as a condition for probation, especially for pedophiles.

Here is an article from the NIH library reporting that chemical castration “reduc[es] recidivism rates to 2% to 5% compared with expected rates of 50%.”

So far as I know, there is no state or country that chemically castrates violent non-sexual offenders.

Of course, in the United States, there is hardly any public debate even on the association between testosterone and crime, much less the possible connection to race differences in crime rates.

This is just one more example of how taboos on race close off useful discussion.