The durability of that pattern makes a recent finding by economists at Harvard and Stanford universities all the more puzzling: Among the poor, the opposite is now true. Girls who grow up in poor families are more likely than the boys who grow up with them to work as adults.
It's an exception that holds up in national data. And in segregated, heavily minority communities like Baltimore — places where rates of incarceration, poverty and single-parent families run high — the gender divide is especially wide. Poor children struggle there. But boys are left even further behind.
And in Baltimore, for example, about 71 percent of girls born in the early 1980s to poor families were working at age 30. Among boys of similar backgrounds, 58 percent were.
In Washington, the difference is even more pronounced: 72 percent of poor girls, but just 56 percent of poor boys, would be employed at age 30. That divide — a kind of reverse gender gap — suggests that boys may be more susceptible than girls to the harmful effects of growing up in unstable, high-poverty environments, the researchers say.
"There's quite consistent evidence throughout life that boys seem more sensitive than girls," says Raj Chetty, a Stanford economist and one of the study's lead researchers. "What we're documenting here is that this shows up even when you're 30 years old."
It also shows up in ways that vary dramatically by geography. The kind of employment gender gap that exists in Baltimore for low-income children doesn't show up in more rural Harford County, Md., 30 miles away. Across the Potomac from Washington in suburban Fairfax County, poor boys are slightly more likely than poor girls to work as adults — a pattern in line with the traditional gender gap in employment.
The research — to be published as a working paper Monday by the National Bureau of Economic Research — measures and maps gender gaps for nearly every county in the nation. The authors contend that the differences that emerge aren't mere coincidence, or an example of people self-sorting into communities.
Rather, they suggest that something about the environment in Baltimore and Washington is altering employment outcomes for the poor children who grow up there, an argument they support by looking at how families fare when they move away.
The findings raise questions about fundamental differences between boys and girls, and between places where children thrive and where they don't. The results also challenge the idea that unemployed men particularly in inner cities have become jobless through laziness or that they simply need to take more responsibility.
"If you could put yourself in their shoes, if you grew up where they grew up," says Harvard economist Nathaniel Hendren, another co-author of the study, "we think you'd basically have the same chance of doing the same thing."
'When they were little ... '
Economists seldom think about the effect of childhood on gender gaps in employment. But the connection is not new to community advocates in the kinds of neighborhoods the researchers describe.
Joe Jones, the president of the Center for Urban Families in Baltimore, works with men with high school diplomas who score at the sixth- and seventh-grade levels on reading and math tests.
"That means somewhere along the line, a combination of family, schools and policy failed this particular population," Jones says. "This didn't happen when they turned 21. This happened when they were little people."
"If a little girl falls down, we want to grab her, we want to coddle her, we want to heal her," he says. "And when it comes to men and boys, it's a little different philosophy. We believe they should pull themselves up by their bootstraps."
The reasons boys and girls experience adversity differently are complex, though.
"Is it actually something physiologically, psychologically different about boys?" says Melissa Kearney, an economist at the University of Maryland who is familiar with the new research but did not contribute to it.
She cautions that it may not simply be that boys are harmed more than girls by poverty — but that they're harmed in ways that are easier to see. Their test scores go down. They're more likely to have behavioral problems and school suspensions. As Chetty and Hendren's work has shown, their employment rates and earnings are measurably lower as adults.
Maybe, Kearney suggests, girls are more likely to experience problems like quiet depression, unhealthy relationships or early parenthood. If that's true, it raises another troubling prospect: Perhaps poor boys are affected by the environments where they grow up in ways that particularly show up — unlike for girls — in their ability to get a job.
"If boys are more likely to respond by acting out, getting in trouble in school, engaging in crime," Kearney says, "those are going to have consequences in the labor market."
"It's not that easy to just say these guys are making poor choices," she adds, "or that they need to be shamed in some way into stepping up, or they need to be on the hook for sending the moms a check."
Common risk factors?
Looking across the country, Chetty, Hendren and the other co-authors Frina Lin, Jeremy Majerovitz and Benjamin Scuderi, tried to identify what connects the places where poor boys seem to fare the worst.
Baltimore and Washington jump out. So do St. Louis, Atlanta, Memphis and Richmond. Nationally, the counties and regions with the widest gender gaps among the poor are also places where there are relatively large black populations, where racial and economic segregation are high and where there are greater concentrations of single-parent families.
The researchers can’t explain why each of those factors might matter, or what segregation has to do with a child’s future job prospects. The reversed gender gap also appears only among poor children with unmarried parents, and those households are more common in these places. It seems to make sense that poor, segregated places would not be ideal environments for children. And perhaps boys in such places, who don’t see stable jobs awaiting them, are more likely to drop out of the labor force and turn to crime.
Chetty does suggest that the situation in high-poverty communities may be related to the plight of men more broadly as they are increasingly left behind by the American economy. Their labor-force participation has been dropping, and economists have pointed to such contributing factors as the rise of competition from China and the changing nature of post-industrial work.
"From a completely different perspective," Chetty adds, however, "it may be that more boys are growing up in really segregated cities with more inequality. And now 20 to 30 years later, we're seeing the consequences of that."
Claudia Goldin, another Harvard economist who studies gender-related issues, offered another idea. Going back to the 19th century, she said, girls have attended high school and graduated at rates that surpass those of boys. Differences have long been apparent between girls’ and boys’ maturity levels and skills such as the ability to concentrate.
"We haven't changed the development of boys and girls," she says. "It's just that the behavior of boys mattered less."
A boy who grew up in deep poverty a century ago didn’t encounter today’s criminal justice system or an economy that values high-skilled labor as much as today’s economy does.
"The consequences today of not graduating, of not going to college, are just much greater. That's what they're picking up," Goldin says of the new research. "It always existed. But the boy who didn't graduate from high school in 1924 just went into the steel mill and did okay."