All people may be created equal, but their futures are often a combination of luck, hard work, and their parents’ socioeconomic status. Although Horatio Alger tales fuel the American dream for millions, many people know that the playing field is not always as level as those optimistic stories imply. Children born to affluent families are more likely to grow up to be affluent adults while children born to less affluent families are more likely to end up poor themselves.

This intergenerational transmission of economic status arises partly because affluent parents are able to invest more time and money in their children, who will do better economically in adulthood than children whose parents are able to invest less. Many people have assumed that as the income gulf among families in the United States continues to widen, this spending divide would also increase, which could contribute to even greater inequality in the next generation.

To bridge this “investment gap,” government policy—through spending on education, health care, and other social programs—aims to make adult income less dependent on social origins and more dependent on skill and effort. Susan E. Mayer, dean and professor, and Leonard Lopoo, PhD’01 (Syracuse University) in their recent Journal of Public Economics article, “Government Spending and Intergenerational Mobility,” provide convincing evidence that social policy is indeed working to narrow the investment gap.

Study Design

Mayer and Lopoo use data from the Panel Study of Income Dynamics (PSID) to estimate the extent to which parents’ income predicts children’s income in adulthood. The PSID has regularly interviewed the same set of families since 1968, capturing a wide range of information on both parents and their children.

The authors then use data from the United States Census of Governments on state spending to estimate whether the amount that state governments spend predicts the relationship between parents and children’s income. If government spending helps close the investment gap among families, then parental income should matter less to children’s eventual income in states with higher spending than in states with low spending. They find that states spent, on average, $13,310 per child in 2000, ranging from a low of $9,157 per child to a high of $16,807. The authors use differences in state spending over time and across states to test their theory.

Government Spending Narrows the Gap

The key indicator in Mayer and Lopoo’s study is the “elasticity” of the relationship between parents and children’s income, which ranges from 0 to 1. When parental income has no bearing on a child’s economic success later in life, the elasticity is 0. When a child’s economic position in adulthood is fully determined by parental income, the elasticity is 1. If government spending helps close the investment gap, then the elasticity should be smaller in high-spending states. They find that the elasticity is 0.5 in low-spending states and 0.3 in high-spending states. Further analysis reveals that a 10% increase in government spending raises low-income adolescents’ future earnings by 4%—reducing the influence of low parental income. However, for higher-income families an increase in government spending hardly raises adolescents’ future earnings.

Additionally, the authors examined what kinds of spending are most effective at increasing children’s future income and for whom it is most helpful. Funding for elementary and secondary schooling has the largest impact on children’s future income, but other kinds of spending matter as well. Government spending also had a much greater impact on the future income of children raised in low-income compared to high-income families. For example, among low-income children spending on public welfare, hospitals, Medicaid, health, higher education, and housing and community development all increased the future income of children raised in low-income families. No form of government spending, however, benefited children from the highest earning families. Because government spending increased the income of adults from low-income—but not high-income—families, it reduced the influence of family origin on adult economic success and effectively increased equality of opportunity. The results remain when the authors hold constant both family and state characteristics.

Policy Implications

Mayer and Lopoo conclude that government spending can make up some of the difference in families’ ability to invest in their children. The results also suggest that increasing spending in states with low levels of spending goes a longer way than investing in states with higher levels of spending.