urban institute nonprofit social and economic policy research

Five Questions for Austin Nichols


Austin Nichols of the Urban Institute's Income and Benefits Policy Center delves into child poverty, social insurance, taxes, retirement, and family structure. His most recent paper, "Understanding Changes in Child Poverty Over the Past Decade," explores reasons behind two sharp changes in child poverty.

Five Questions Archives

June 7, 2006

1. How has child poverty changed over the past decade?

Child poverty rates dropped dramatically in the 1990s, especially for blacks and Hispanics. Some driving forces included welfare reform and expansions in the EITC [Earned Income Tax Credit]. Economic changes played an even more major role. The '90s were a big boon for everyone, so it's not surprising that children were better off. But the size of the improvement did take people by surprise, especially the change in black child poverty.

More recently, child poverty has climbed since the economic downturn in 2000-2001, though not to the level of the early 1990s. So for seven or eight years, child poverty fell dramatically. As the economy cooled in the past three to four years—the most recent data are from 2004—child poverty inched back up. We don't know where it's going to go in the next few years.

Interestingly, my research shows that historical explanations for child poverty, which target family structure, don't say as much about recent changes in child poverty as job market conditions do.

A lot of the Urban Institute research on low-income working families focuses on job conditions and the job market. It's not that family isn't important—a single-parent family is still exposed to a lot more risk, and is much more likely to be poor—but it's no longer the case that shifts in poverty are about shifts in marriage or childbearing patterns.

I haven't looked at all possible causes for child poverty. For instance, age at first birth and ages of parents have generally increased over the past two decades, and older parents are less likely to be poor, which is another potential explanation. Also, the teen birth rate has dropped, and kids having kids are more likely to be poor. Family size has decreased. But these and other demographic trends can't explain the big drop in child poverty since the timing is wrong. For one thing, those changes in family structure have continued since 2000, while the child poverty rate has increased.

There's the EITC and welfare reform, which are clearly important. Carolyn Ratcliffe and Signe-Mary McKernan have looked at them. But it's often hard to identify how a single policy change leads to fluctuations in poverty. And, again, the EITC and welfare reform did not disappear after 2000, while child poverty has risen. The strongest evidence in my work links changes in child poverty to changes in the job market driven by the broader economy.

2. How do changes in family composition factor into this?

Family structure seems to be much less of a factor in child poverty than it has in past decades. Robert Lerman found that changes in family accounted for changes in poverty in prior decades. Lots of other people have found that the change from a two-parent to one-parent family is a large determinant of poverty.

Recent huge changes in child poverty show little evidence of being caused by family composition, which was a major predictor of child poverty before 1993. Family has been changing since the 1930s, but in the past 10 years there hasn't been a big and abrupt shift to correspond to the big and abrupt shift in child poverty. What changed dramatically were the job market, welfare reform, and the EITC.

All these things tie together: education, work, and family structure can be mutually dependent. Someone earning a low wage because of a low-quality high school education is going to be a less-stable participant in a family. If a parent loses a job, that could lead to a family breakup. Causation runs in every direction—teenagers who have kids might never finish their educations, and then their job market outlook is pretty bleak. All those factors working together are what lead to child poverty. The Urban Institute had a roundtable discussion recently on how much risk many families and children face, and a lot of these causal links were discussed there.

3. Why is a black child twice as likely to be poor as a white child?

The change in the black child poverty rate inspired my research. The white child poverty rate has bounced around for 50 years, but the black child poverty rate dramatically declined from 1993 to 2001, and no one seemed to have a very good explanation for that.

A black child is actually more than three times as likely to be poor as a white child, 33 percent versus 10 percent, over the last five years. A black child is often twice as likely to be poor as a white child conditional on work, family structure, and education: for example, in families with at least one full-time worker and both parents present, 7 percent of black children are poor compared with 3 percent of white children.

But as bad as that sounds, it still represents an improvement: a black child was four times as likely to be poor as a white child in the late 1980’s, before the big decline in child poverty during the 1990’s. At the beginning of the civil rights era, the mid-60s, the majority of black children were poor. Throughout, the white child poverty rate has stayed relatively steady, in the 10 to 15 percent range. 

Despite all the strides being made, black children are still substantially worse off than children on average. Different characteristics can explain some of this. The schools and neighborhoods for black kids are lower quality by many measures. But an important part of the difference is due to labor market discrimination. It's not just that black workers may earn less, it's also that they have a harder time finding jobs. This fact can discourage people from seeking jobs and even seeking the education to get the jobs, triggering a vicious cycle.

Yet, we also have some recent immigrant groups doing quite well but facing similarly poor conditions. It's hard to say how much the cycle of poverty is self-perpetuating and how much is due to conditions on the ground, like schools and neighborhoods, and then how much parents can work their way out of poverty. No one really knows for sure, but it's clear that a great deal of that black-white differential is due to the schools and labor market factors.

Still, though black kids are more likely to be poor, poor kids are not more likely to be black. Only about a quarter of poor kids are black—the biggest group among poor kids is non-Hispanic whites.

4. How do the economy and the job market figure into changes in child poverty?

I've identified it as being huge. The economy fuels the job market. Whether someone is working is important, as is the level of state unemployment. These fluctuations matter more for the disadvantaged. They matter more for people with a high school education or less. They matter more for black families. The more marginalized in society, the more any economic downturn or upturn matters. Black child poverty improved in the 1990s because black families made greater gains in the labor force than white families. So it all connects.

We can't just make economic growth happen, but we can shape policy to serve the needs of low-income workers and children. For instance, the EITC is cited as the main reason single parents and other low-wage workers joined the workforce in droves. The "Working to Make Ends Meet" paper I wrote with Greg Acs shows that many working families struggle to cover their expenses; the EITC can mean the difference between paying rent or not. A lot of these families require supports. For instance, the low-income families that are working are more likely to get subsidized child care.

The minimum wage is a fraught issue. I've found mixed evidence. It helps some people and hurts others. The minimum wage discourages employment among some parts of the population. It's been found to increase unemployment in some studies. Yet, it will raise the wages of some workers. It's not clear whether the benefits outweigh the costs.

5. What are your concerns for the future?

My research doesn't target specific policy options directly, but it does evoke policy ruminations. Child-friendly policies in this country increasingly focus on middle-income families instead of those at the bottom rung. For example, the child tax credit has been expanded recently and that mostly helps the middle class. Yet, Congress hasn't expanded the EITC recently and the IRS keeps it under constant scrutiny and pressure.

Should we be shifting policies to benefit the middle class if we care about investing in children? Reasonably well-off families can do a good job for their children no matter what the policies. But they often get the largest subsidies—for example the tax subsidy for child care, which is more valuable for people in higher tax brackets. We could put that money into quality child care for low-income families while their parents work. Spending on better early childhood education could be more equitable. As it is, some areas get really intensive investments and others get none at all. These are the kinds of investments that would really pay off.

In terms of concerns for future research, Jane Hannaway, Laura LoGerfo, Rob Olsen, and I are looking at strategies to improve educational outcomes and opportunity. Usually such improvement would mean greater equity, too.

And later this year I'll be looking at risks faced by low-income families with Greg Acs and Pam Loprest. Risk is defined pretty broadly to capture the economic impacts of a range of events families might face, from deteriorations in health or job losses to childbirth or death of a family member. We want to see how families deal with certain life events, what kind of strategies might help families navigate setbacks, and what kind of policies might help families ride out economic shocks.