How much can early childhood education do to reduce income inequality?

New York Times columnist Nicholas Kristof wrote in a recent column  that the Occupy Wall Street movement “is shining a useful spotlight on one of America’s central challenges, the inequality that leaves the richest 1 percent of Americans with a greater net worth than the entire bottom 90 percent.”  Kristof then goes on and uses most of his column to argue that “the single step that would do the most to reduce inequality [is]… an expansion of early childhood education.”

Is this true? What is the magnitude of the reduction in income inequality brought about by early childhood education? How does it compare with the magnitude of America’s income inequality problem? I want to address those issues in this blog post.

The evidence is strong that early childhood education can significantly increase the future earnings and income of low-income children. For example, in chapter 8 of my book Investing in Kids, I estimate that even a preschool program that lasts only for one school-year at age 4, and is a half-day program, can, if it is a high-quality program, raise the future income from earnings of the lowest income quintile by over 6% of their expected future income.

(The lowest income quintile is based on first ranking all households by their income. We then divide the households into five groups or quintiles based on this income ranking. This lowest income quintile has a little over 3% of total U.S. household income, or in other words, their average income is only 1/6th of average income.)

Similar results come from my recent study, with Bill Gormley and Shirley Adelstein of Georgetown University, of Tulsa’s near-universal pre-K program. For children whose families have low enough income that they are eligible for a free-lunch, a one-half-day program for one school year at age 4 is estimated to raise their future earnings by about 7%.

More intensive and more expensive programs can do more. In the Tulsa study, we estimate that for children eligible for a free lunch, a full-day program for one school year at age 4 will raise their expected future earnings by over 10%.

In chapter 8 of Investing in Kids, I also consider the long-run earnings and income impacts of the Abecedarian program. This program was a full-time high quality child care and preschool program from shortly after birth up to age 5.  The program is similar to the Educare Centers that have been sponsored around the U.S. by a number of major foundations, including the Buffett Early Childhood Fund and the Ounce of Prevention Fund.

I estimate that such an intensive program can raise the future income as adults of children from the lowest income quintile by 35%. (Of course, this larger percentage effect comes at a price. An Educare/Abecedarian program can increase earnings and income by about 6 times as much as one-half-day of preschool for one year, but it costs over 10 times as much per child assisted. However, both programs are estimated in my book to clearly have economic benefits greater than costs. )

What about other income groups? In our recent study of Tulsa, we estimate that more middle-class children, from families that have enough income that they must pay full price for lunch, get about the same test score impact and same DOLLAR impact on future income as is true for low-income children. However, as their expected future income is larger, the percentage impact is somewhat lower. For middle-class children, one school year of half-day preschool raises expected future earnings by 4%, and a full-day raises expected future earnings by over 5%. The dollar benefits still exceed the costs for this middle-class group, but the percentage impacts don’t loom quite as large compared to their future prospects.

All of these estimates are for changing early childhood education, and nothing else. Improvements in K-12 education for low-income and middle-class children may add to these economic benefits. Synergies may mean that with better K-12 schools, the rates of return to these early childhood programs will be even greater.

Are these percentage effects large? They seem large to my intuition, but can that be backed up by some relevant comparison?  According to the Congressional Budget Office, from 1979 to 2007, prior to the Great Recession,  average incomes of the lowest-income quintile of households grew by about 40 percentage points less than overall “average incomes” in the U.S. , where that average includes the incomes of all Americans, including the wealthiest. For the middle-income quintile, their average incomes grew about 30% less than the overall average for all Americans from 1979 to 2007. (How can middle-income Americans have slower income growth than the overall average? Because the growing income of the wealthiest Americans skews the overall average up. )

(Technical note to those interested: I am using CBO’s tables on after-tax and benefit income for households, adjusted for household size. These data include in-kind income such as health benefits as well as the impact of federal taxes, welfare, and food stamps. The percentage gap is the difference between the percentage growth of average household income over all households for the 1979 to 2007 period, versus the percentage growth for the lowest-income quintile and the middle-income quintile. For example, adjusted for household size, CBO calculates that average after-tax and benefit household income grew from $49,300 in 1979, measured in 2007 year prices, to $76,400 in 2007, a growth of about 55 %.)

Therefore, growing income inequality, even prior to the Great Recession, has meant that the lowest income quintile has about 40% lower income than it would have if its income growth had matched the growth of the overall American economy.  If we did really intensive programs such as Educare for ALL lower-income children, we could close almost all this income gap by that program alone. Even one year of half-day or full-day preschool would make a significant dent in this gap.

For the middle class, early childhood education by itself could significantly help, but it doesn’t loom quite as large. The middle class has about 30% lower income than it would have if its income growth over the last 30 years had matched overall economic growth.  One year of half-day or full-day preschool could eliminate perhaps one-tenth to one-sixth of this gap. That’s a significant effect, but obviously there would still be a considerable remaining gap.

What about the rich? If we make pre-K education universal, won’t they also gain, offsetting any reductions in income inequality from helping the poor and middle-class?

On this topic, we don’t really have direct evidence for the benefits of universal pre-K education for the rich.  However, we would expect benefits to be lower because many of these families already have their children enrolled in high-quality private preschools. Even if universal pre-K education provided the rich with similar dollar increases in earnings for their children, the percentage effects on their expected future income would be much lower. Therefore, even with a universal pre-K program, gains for the poor and middle class would serve to reduce income inequality.

Furthermore, for the most intensive programs, such as Educare, these programs by their design and cost are intended to be targeted at the most needy families.

Therefore, I would conclude that it is true that early childhood education can do a large amount to reduce income inequality. Realizing this potential requires expanding programs that are high-quality to full-scale.

Early childhood programs may not be the only way to reduce income inequality. But they are a way of reducing income inequality that has been proven by rigorous studies to have a high bang for the buck.  From an overall economy perspective, the national benefits significantly exceed costs. Furthermore, some types of early childhood education, such as universal pre-K education, have the potential for delivering sizable benefits to the middle-class as well as the poor, which attracts political support. It is hard to think of any other policy option for significantly reducing income inequality that simultaneously does all of the following:  provides national benefits greater than costs; has rigorous research evidence; and, potentially helps a majority of voters.

(Thanks to the Smart Start program for their mass e-mail that drew Kristof’s column to my attention.)

About timbartik

Tim Bartik is a senior economist at the Upjohn Institute for Employment Research, a non-profit and non-partisan research organization in Kalamazoo, Michigan. His research specializes in state and local economic development policies and local labor markets.
This entry was posted in Distribution of benefits, Early childhood programs. Bookmark the permalink.