Lane Kenworthy has an excellent recent essay on what the U.S. can do to increase equality of opportunity. (Kenworthy is a well-known sociologist at the University of Arizona, who has written several insightful books on issues of poverty and income inequality.)

However, I want to focus attention on one statement in his recent essay with which I disagree:

“Suppose the United States increased tax rates for all households and used the revenue to fund universal early education…That step would do little to counter income inequality, but it could substantially expand opportunity.”

It is true that tax-supported universal early education would not immediately have large effects in reducing income inequality. Tax-supported universal early education would also not directly have government take money away from the rich and transfer that money to the poor.

However, under the range of plausible assumptions about how early education affects different income groups, universal early education would significantly reduce income inequality. Some researchers, such as Nobel prize-winning economist James Heckman, argue that early childhood education has much larger effects on children from low-income families than for children from higher-income families.

If this is so, then universal early childhood education would have much larger dollar effects on the future adult income of children from lower income families. This would have significant effects in redistributing economic opportunity and income.

In my book, Investing in Kids, I estimate the income distribution effects of universal pre-K, under the assumption that the benefits of universal pre-K fall off quite rapidly with higher incomes. For example, I assume that dollar effects on average adult earnings for the middle income quintile are less than a third of the dollar effects on average adult earnings for the lowest income quintile. (Each quintile comes from ordering all households by income, and then dividing that distribution into five equally sized groups of households.)

This results in huge income redistribution effects of universal pre-K.  I assume universal pre-K is financed by taxes that are modestly “regressive”, that is the taxes are a larger percentage of income for low income groups. But even under that assumption, the net impact on income of different groups is highly progressive.

For example, I estimate that after adjusting for both tax costs and future earnings benefits, universal pre-K will increase the net present value of the income of the lowest income quintile by 6%. The middle-income quintile has the net present value of their income increased by 0.4%. The highest income quintile has a slight loss from universal early education – this group’s taxes exceed the future earnings benefits, so the net present value of this group’s income declines by a small amount, about 0.1%.  (All these numbers, and more details on how they are calculated, can be found in chapter 8 of my book. This is one of the chapters that can be downloaded for free.)

But universal early education can have large effects on reducing income inequality even if the benefits don’t flow so disproportionately to the poor. In my recent paper on Tulsa’s universal pre-K program, co-authored with William Gormley and Shirley Adelstein of Georgetown, our estimates imply that universal pre-K will increase the future average adult earnings of all income groups by about the same annual dollar amount.

However, the same dollar annual earnings boost for all income groups is a much larger percentage boost in the fortunes of low-income groups. Common sense as well as mainstream economics thought would say that adding the same dollar amount to the incomes of the poor has much greater social welfare benefits than adding the same dollar amount to the incomes of higher-income groups.

For example, we estimated that the full-day version of Tulsa’s pre-K program for 4-year-olds would be likely to increase the present value of adult earnings for children eligible for a free lunch by about $27,000, compared to about $25,000 for children whose family income was high enough that they had to pay full-price for lunch.  Not much of a difference.  But we also estimated that this adult income boost would be sufficient to increase the present value of adult earnings of the free-lunch children, over their entire working career, by 10.4%, compared to 5.5% for children from families ineligible for a lunch subsidy.

By itself, universal early education does not “solve” the problem of income inequality.  But if public policy keeps on finding and adopting public policies that help a wide variety of families, but increase the income of disadvantaged families by a higher percentage, over time this mix of policies would have profound effects on income inequality and the nature of U.S. society.