Being Poor Doesn't Have the Same Effect as Living in Chaos

The question of poverty vs. instability

A reminder that my debut book Troubled: A Memoir of Foster Care, Family, and Social Class is now available.

This piece titled “Throwaway Kids” reports researchers surveyed “nearly 6,000 inmates in 12 states — representing every region of the country — to determine how many had been in foster care and what effect it had on their lives. Of the inmates who took the survey, 1 in 4 said they were the product of foster care. Some spent the majority of their childhood in strangers’ homes, racking up more placements than birthdays.”

In the Los Angeles county foster care system (my beloved alma mater), only 64.5% of foster kids graduate from high school. Maybe not so surprising.

Here’s something more surprising. According to the same report from LA county, the overall high school graduation rate in LA is 86.6%.

And the graduation rate for students categorized as “socioeconomically disadvantaged” is also 86.6%. Poor kids graduate at the same rate as everyone else.

The gap between poor kids and foster kids exists nationwide, too.

72.4% of kids across the U.S. in the lowest socioeconomic quintile graduate from high school. In contrast, nationwide, only 58% of foster kids graduate from high school.

What about college? 11% of kids in the bottom socioeconomic quintile graduate from college. For foster kids: 3%.

Incarceration rates are similar. About 8% of males from families in the bottom socioeconomic quintile do time in prison or jail. For males who were in foster care: 60%

Consider that to become foster parents, people must meet a minimum economic threshold. They can’t be poor.

Which means kids in foster care are not in as materially impoverished an environment as kids in the bottom income quintile.

So what explains the gap in graduation and incarceration rates between foster kids and poor kids?

One reason worth highlighting comes from this widely-cited paper in Developmental Psychology titled, “Evolution, Stress, and Sensitive Periods: The Influence of Unpredictability in Early Versus Late Childhood on Sex and Risky Behavior.”

The researchers used a longitudinal data set. In the 1970s, women at a public health clinic agreed to respond to questions that tracked both themselves and their then unborn children.

Both the mothers and, later, their children, responded to questionnaires at multiple time points until the children reached early adulthood.

The researchers were interested in how 2 different environmental factors affected 5 key variables.

One environmental factor was the amount of environmental harshness children experienced before age 5.

Researchers measured environmental harshness by the mother’s socioeconomic status, occupational prestige, and household income. How materially comfortable was the kid?

The other environmental factor was the amount of environmental unpredictability children experienced before age 5.

The researchers measured environmental unpredictability by number of changes in residence (e.g., moving to a different house/apartment), changes in cohabitation status (e.g., whether and how often male romantic partners moved in or out of the house/apartment), and changes in employment status.

In short, how often the kid moved, how frequently the adults in the kid’s life appeared and disappeared, and how frequently his mom changed jobs. How chaotic was the kid’s life?

And the researchers wanted to know how these two factors influenced 5 outcome variables:

• Age at first intercourse

• Number of lifetime sexual partners at age 23

• Criminal acts

• Aggression (e.g., “I deliberately try to hurt others,” “I destroy things belonging to others”)

• Delinquency (frequency of lying/cheating, breaking rules, setting fires, stealing, drug use)

Researchers found that childhood poverty (harshness) was not significantly associated with any of the 5 outcome variables.

In contrast, there was a significant correlation between childhood unpredictability 4 of the 5 outcome variables—number of sexual partners, aggressive behavior, delinquent behavior, and criminal behavior. For males, but not females, instability predicted having sex at an earlier age.

The correlation between unpredictability in childhood and criminal behavior in adulthood was particularly striking (r = .40, p < .01). This effect size is equivalent to the correlation between socioeconomic status and SAT scores.

The luxury belief class loves to talk about the effect of wealth on test scores. Few discuss the effect of instability in childhood giving rise to harmful behaviors in adulthood.

The researchers re-analyzed the data while controlling for harshness. The relationship between instability in childhood and harmful behaviors in adulthood remained significant.

They conclude their discussion:

“The findings revealed that the strongest predictor of both sexual and risky behavior at age 23 was exposure to an unpredictable environment during the first 5 years of life. Individuals exposed to more unpredictable, rapidly changing early environments displayed a faster life history strategy at age 23 as indicated by having more sexual partners, having sex at an earlier age (for males only), engaging in more aggression and delinquent behaviors, and being more likely to be associated with criminal activities/behavior. By contrast, exposure to either harsh environments or experiencing unpredictability later in childhood (ages 6–16) did not significantly predict these outcomes at age 23.”

Plainly, being poor doesn't have the same effect as living in chaos.

There are some people who will respond "Bruh, it's all in the genes."

Behavior is modifiable for just about everyone, regardless of their genetic propensities and endowments. Behavior responds to incentives and environmental inputs.

Suppose each of us has a different innate propensity to punch others. In this scenario, some researcher collects data on how many people each person punches per year.

In a completely free environment with no norms or consequences, I would punch 10 people a year. And in this free environment, you would punch 3 people a year.

A difference of 7 people. In this simple hypothetical example, I am "genetically" more prone than you to punch.

Now suppose we both exist in an environment with strong norms against punching. In this environment, people lose status for violence. And violent people experience swift and unfavorable consequences.

In this environment, I now punch only 8 people a year, and you now punch only 1. I am still punching 7 more people than you each year.

The gap between us is the same as it was before.

But—and this is crucial—we are punching fewer people than before. Fewer people overall are being punched.

The average number of punched victims was 6.5 a year in the free environment. And now it is 4.5 in the rigid environment.

Relative differences exist.

But so do absolute differences. Those matter too.

---

I am well aware of the behavior genetics research, twin studies, and so on indicating little effect of home environment on personality, propensity for crime, addiction, and so on. These studies are inapplicable for kids living in extreme dysfunction.

Behavior geneticists investigate the relative role of genetic and environmental variation within the sampled population.

Behavioral genetics studies report findings from within the environmental variation in their samples, not in all conceivable environments.

For example, there are many studies on identical twins separated at birth who are adopted by different families.

Researchers find little difference between these twins when they are adults. Their personalities, IQ, preferences, and so on are very similar.

But twins are usually adopted by intact middle-class families.

They are typically taken in by married parents with the means to jump through the hoops to qualify for adoption. Additionally, adoptive parents are the kind of people who would adopt, which introduces another layer of similarity.

I’ve yet to see any twin studies with one set of identical twins raised in extremely bad environments and another in good ones.

The psychologist and intelligence researcher Russell T. Warne has written:

“A problem with heritability study samples is that they tend to consist of more middle and upper-class individuals than a representative sample would have…results of behavioral genetics studies will indicate genes are important—if a person already lives in an industrialized nation in a home where basic needs are met...it is not clear how well these results apply to individuals in highly unfavorable environments.”

The "parents don't matter" idea from behavior genetics comes from an era where most children were raised in homes with two biological parents (range restriction). Families and values were pretty similar across the board. Even when data were collected from children of single parents, most of those children were raised in neighborhoods where most of the parents were married, and attended schools with classmates whose parents were mostly married. Genes will always play a large role in the trajectory of our lives, but as variation in family type expands, my strong suspicion is that the environment will account for more and more of the variance in outcomes between individuals.

In a chapter titled Genes and the Mind, the psychologist David Lykken states:

“If twins were separated as infants and placed, one with a middle-class Minnesota family and the other with an 18-year-old unmarried mother living on AFDC in the South Bronx, the twins will surely differ 30 years later.”

Yes, genes are responsible for human traits and behavior. But these traits are responsive to social norms and other environmental factors too.

Education and weight are both 70 percent heritable.

In the 1970s, 13 percent of Americans graduated from college. Today it’s 35 percent.

In the 1970s, about 13 percent of Americans were overweight. Today, it’s around 70 percent.

Did our genes make us smarter and fatter over the last 50 years?

No.

The environment (incentives, accessibility, social norms, etc.) made it easier to go to college. So more people went.

The environment (incentives, accessibility, social norms, etc.) made it easier to get fat. So more people did.

The heritability of divorce is around 40% percent. In the 1950s, 11 percent of children born to married parents saw their parents get divorced. By the 1970s, more than 50 percent of children born to married parents saw their parents get divorced.

Did genes change that much in 20 years?

No. It became easier to divorce. So more people did.

Crime skyrocketed in 2020. Did genes transform overnight?

No. Lawlessness became easier to get away with. So more people committed crimes.

Height is 90 percent heritable. But it is still malleable by the environment. Before Korea was divided, Northerners were taller than Southerners. Today, North Koreans are 6 inches shorter, on average, than South Koreans. Did their genes change? No. Their environments did.

Tobacco use is highly heritable (60-80%) but the percentage of Americans who smoke has dropped by half since 1982.

Strong norms against smoking contributed to people changing their behavior by smoking less.

Genes have something to do with behavior. But behavior can be unleashed or constrained depending on the norms of a society.

Some people interpret behavior genetics findings to mean environment is unimportant. I interpret them to mean certain aspects of environment matter even more. Norms and customs constrain differences between individuals. The absence of norms and guardrails magnifies them.

---

My late adoptive grandparents were members of the Silent Generation (born in the 1930s). White working class. My grandfather dropped out of school in eighth grade to work and earn money for his parents. When my grandfather was 21, my grandmother (age 18) accepted his proposal after he agreed to give up smoking, drinking, and gambling. They relocated from the midwest to Oregon in search of work. They were married for 60+ years, created a strong intact family, and had 3 biological children and then adopted my adoptive mother. Their 3 biological children (born in the 1950s and 60s) all grew up and got married. All got divorced. Some remarried. They all had kids (my cousins). Despite the divorces, these boomers still believed in marriage and still remember the old ways. What life was like before the western cultural revolution. This is where things take a sharp turn. My cousins (born in the 1980s) all grew up. None got married. All of them have kids. Some of them have kids with multiple different partners who they no longer speak with. These kids (born in the 2000s) have weird non-natural hair colors, aggressive piercings, and behavioral issues. As young adults, my grandparents were far poorer than their children and grandchildren. They had the same genes. But they lived in a different kind of culture.

---

Of my closest friends in high school, two went to prison and one was shot to death. These guys were never going to go to a fancy college. But they didn’t have to end up incarcerated or murdered at age fourteen. The truth is that most foster kids or poor kids could not get a Ph.D. in mathematics at MIT (though some could), regardless of how they are raised or how many resources are dedicated to them. But we could drastically reduce the number of kids who later land in prison, rehab, on the street, and so on. We might not be able to raise the ceiling. But we could raise the floor. That starts at home.

If you appreciate the work I do here on my Substack or on Twitter/X, I’d be thankful if you ordered my book:

Amazon (US)

Barnes & Noble

Bookshop

Amazon (UK)

Audible (I narrated the audiobook myself)

Other book vendors