Preschool Pays

High-Quality Early Education Would Save Billions

The youngest children suffer the highest poverty rates of any age group in the United States. Nearly one in five children under age 6 lives in poverty, and the number is rising.

Poor children often have inadequate food, safety, shelter, and healthcare. In school, poor children too often fall far short of achieving their academic potential, making them more likely to enter adulthood lacking the skills to compete in the global labor market. As adults, they are more likely to suffer from poor health and participate in crime and other antisocial behavior; they are also less likely to be gainfully employed and contributing to economic growth and community well-being.

There is a strong consensus among the experts who have studied high-quality early childhood development (ECD) programs that these programs have substantial payoffs. Although the programs vary in whom they serve and in the services they provide, most high-quality ECD programs have the following characteristics in common: well-educated and trained staff; a low child-to-teacher ratio and small classes; a rich curriculum that emphasizes language, pre-literacy, and pre-numeracy activities, as well as motor, emotional, and social development; health and nutritional services; and lots of structured and unstructured play. Good programs also typically include parental involvement and education.

What benefits have such programs produced? We can answer this question thanks largely to carefully conducted, long-term studies that have compared the school and life outcomes of participants in four high-quality ECD programs—the Perry Preschool Project, the Prenatal Early Infancy Project, the Abecedarian Early Childhood Intervention, and the Chicago Child-Parent Center Program—with a control group of children who attended no such program.1 (Two of these, the Perry Preschool Project and the Chicago Child-Parent Center Program, are described in more detail in the sidebar "The Benefits of High-Quality Early Education.")

These studies have established that participating children are more successful in school and in life than children who were not enrolled in high-quality programs. In particular, children who have participated in high-quality ECD programs tend to have higher scores on math and reading achievement tests, have greater language abilities, are better prepared to enter elementary school, are more likely to pursue secondary education, have less grade retention, have less need for special education and other remedial coursework, have lower dropout rates, have higher high school graduation rates, higher levels of schooling attainment, improved nutrition, better access to healthcare services, higher rates of immunization, better health, and experience less child abuse and neglect. These children are also less likely to be teenage parents and more likely to have higher employment rates as adults, lower welfare dependency, lower rates of drug use, show less-frequent and less-severe delinquent behavior, engage in fewer criminal acts both as juveniles and as adults, have fewer interactions with the criminal justice system, and lower incarceration rates. The benefits of ECD programs to participating children enable them to enter school "ready to learn," helping them achieve better outcomes in school and throughout their lives (Barnett, 1993; Karoly et al., 1998; Masse and Barnett, 2002; Schweinhart, 1993).

Parents and families of children who participate in high-quality ECD programs also benefit. For example, mothers have fewer additional births, have better nutrition and smoke less during pregnancy, are less likely to abuse or neglect their children, complete more years of schooling, have higher high-school graduation rates, are more likely to be employed, have higher earnings, engage in fewer criminal acts, have lower drug and alcohol abuse, and are less likely to use welfare (Karoly et al., 1998).

Because of these positive results, there is now a consensus among experts of all political persuasions that investments in high-quality ECD programs have huge potential long-term payoffs. Investments in high-quality ECD programs consistently generate benefit-cost ratios exceeding 3-to-1—or more than a $3 return for every $1 invested. While participants and their families get part of the total benefits, the benefits to the rest of the public and government are even larger and, on their own, tend to far outweigh the costs of these programs. Several prominent economists and business leaders (many of whom are skeptical about government programs generally) have recently issued well-documented reviews of the literature that find very high economic payoffs from ECD programs. For example, Nobel Prize winning economist James Heckman of the University of Chicago has concluded:

Recent studies of early childhood investments have shown remarkable success and indicate that the early years are important for early learning and can be enriched through external channels. Early childhood investments of high quality have lasting effects.... In the long run, significant improvements in the skill levels of American workers, especially workers not attending college, are unlikely without substantial improvements in the arrangements that foster early learning. We cannot afford to postpone investing in children until they become adults, nor can we wait until they reach school age—a time when it may be too late to intervene. Learning is a dynamic process and is most effective when it begins at a young age and continues through adulthood. The role of the family is crucial to the formation of learning skills, and government interventions at an early age that mend the harm done by dysfunctional families have proven to be highly effective.2

The director of research and a regional economic analyst at the Federal Reserve Bank of Minneapolis, Arthur Rolnick and Rob Grunewald, have come to similar conclusions:

... recent studies suggest that one critical form of education, early childhood development, or ECD, is grossly under-funded. However, if properly funded and managed, investment in ECD yields an extraordinary return, far exceeding the return on most investments, private or public.... In the future, any proposed economic development list should have early childhood development at the top.3

This Federal Reserve Bank of Minneapolis study (Rolnick and Grunewald, 2003) further determined that annual real rates of return on public investments in the Perry Preschool project were 12 percent for the non-participating public and government and 4 percent for participants, so that total returns exceeded 16 percent. Thus, again it is advantageous even for non-participating taxpayers to pay for these programs. To comprehend how extraordinarily high these rates of return on ECD investments are, consider that the highly touted real rate of return on the stock market that prevailed between 1871 and 1998 was just 6.3 percent.4

Likewise, after reviewing the evidence, The Committee for Economic Development (CED), a nonpartisan research and policy organization of some 250 business leaders and educators, concluded that:

Society pays in many ways for failing to take full advantage of the learning potential of all of its children, from lost economic productivity and tax revenues to higher crime rates to diminished participation in the civic and cultural life of the nation.... Over a decade ago, CED urged the nation to view education as an investment, not an expense, and to develop a comprehensive and coordinated strategy of human investment. Such a strategy should redefine education as a process that begins at birth and encompasses all aspects of children's early development, including their physical, social, emotional, and cognitive growth. In the intervening years, the evidence has grown even stronger that investments in early education can have long-term benefits for both children and society.5

What if we provided high-quality early childhood development to all poor children?

How much would it really cost the government to provide such an experience to all poor children? And how much would it actually save the government in terms of crimes not committed, welfare payments no longer needed, reduced remedial education costs, more taxes collected, and so forth?

In a new study published by the Economic Policy Institute and summarized here, I calculate how much taxpayers would save, how much the economy would grow, and how much crime would be reduced over the next 45 years if high-quality programs were provided for all poor children. To create these estimates, I've extrapolated from research on the Perry Preschool Project.6 Perry was not chosen because it is an ideal program (or even better than the three other programs named above). It is simply the only program with data suitable for these extrapolations. (For a fuller explanation of the merits of such extrapolations, see the sidebar "Extrapolating from One Program...".) These estimates assume the launch of an ECD program for all of the nation's three- and four-year-olds who live in poverty in 2005, with full phase-in by 2006. (For practical purposes, such as finding appropriate staff and locations, a large-scale ECD program would have to be phased-in over a longer period.) The costs set forth in these estimates may understate the start-up costs of such an ambitious program, especially the costs of recruiting and training teachers and staff and of establishing appropriate sites. On the other hand, the total benefits of ECD investment are also understated in these estimates (see the sidebar "The Benefits of High-Quality Early Education" for a discussion of some of the benefits of the Perry Preschool Project that are unaccounted for). Thus, although the benefit-cost ratio of a national ECD program could be somewhat higher or lower than that which is found in the pilot programs, it is implausible that the ratio would be less than the 1-to-1 ratio necessary to justify launching the program.

In the next two sections we'll look at the results of these extrapolations and specifically the effects of ECD investments on 1) government budgets, and 2) on the economy and crime.

What is the effect on government budgets?

We can expect, based on long-term research on children who participated in high-quality ECD programs and similar non-participating children, that these ECD investments would benefit taxpayers and generate government budget benefits in at least four ways.7 First, subsequent public education expenses would be lower because participants spend less time in school (as they fail fewer grades) and require expensive special education less often. Second, criminal justice costs would come down because participants—and their families—would have markedly lower crime and delinquency rates. Third, both participants and their parents would have higher incomes and pay more taxes than non-participants. Fourth, the ECD investment would reduce public welfare expenditures because participants and their families would have lower rates of welfare usage. Against these four types of budget benefits, we must consider two types of budget costs: the expenses of the ECD program itself and the increased expenditure due to greater use of higher education by ECD participants.

The ECD programs do not perform miracles on poor children. Substantial numbers of ECD participants go on to do poorly in school, commit crimes, have poor health outcomes, and receive welfare payments. The key point is that ECD participants as a group have far lower rates of these negative outcomes than do non-participants.

Given all of this, what effect would such ECD investments have on government budgets?8 In the second year of the program, 2006, when the program would be fully phased-in, government outlays would exceed offsetting budget benefits by $19.4 billion (in 2004 dollars). The annual deficit due to the ECD program would shrink for the next 14 years. By the 17th year of the program, in 2021, the deficit would turn into a surplus that would grow every year thereafter. Within 25 years, by 2030 if a nationwide program were started in 2005, the annual budget benefits would exceed costs by $31 billion (in 2004 dollars). By 2050, the net annual budget savings would total $61 billion (in 2004 dollars). In short, for the first 16 years, additional costs exceed offsetting budget benefits, but by a declining margin. Thereafter, offsetting budget benefits exceed costs by a growing margin each year. This pattern is illustrated in the figure below, which shows annual revenue impacts and costs in constant 2004 dollars.

The reason for this fiscal pattern is fairly obvious. The costs of the program will grow fairly steadily for the first decade and a half, in tandem with modest growth in the population of three- and four-year-old participants. Thereafter, costs will grow at a somewhat faster pace for a few years as, in addition to the costs of educating three- and four-year-olds, the first and subsequent cohorts of participant children begin to use public higher education services. After the first two years, when the first cohort of children starts entering the public school system, public education expenditures will begin to diminish due to less grade retention and remedial education. After a decade and a half, the first cohort of children will be entering the workforce, resulting in increased earnings and thus higher tax revenues and lower welfare expenditures. In addition, governments will experience lower judicial system costs.

The timing of these fiscal benefits resulting from a nationwide ECD program should appeal to those concerned about the fiscal difficulties posed by the impending surge of retiring baby boomers. The substantial fiscal payoffs from investing in young children would become available to governments just as the wave of new retirements puts the greatest pressure on government resources. For example, the government-wide budget savings in 2030 and in 2050 from ECD investments begun next year would be enough to offset about one-fifth of the deficits in the Social Security trust fund projected for those years. This potential contribution to the solvency of the Social Security system would be achieved without raising social security taxes or cutting benefits.

[[{"type":"media","view_mode":"wysiwyg","fid":"938","attributes":{"alt":"Annual Budgetary Benefits and Outlays","title":"","class":"media-image","typeof":"foaf:Image","wysiwyg":1,"width":"248","height":"193"}}]]

What is the effect of ECD on crime reduction, earnings, and the economy?

It is important to keep in mind that savings to government is not the only benefit of ECD investments. These other benefits come in many forms. Investments in high-quality ECD programs are likely to substantially reduce crime rates and the extraordinary costs to society of criminality. Some of these reduced costs are savings to government in the form of lower criminal justice system costs. These savings to government would total nearly $28 billion (in 2004 dollars) in 2050, and were included in the earlier discussion of the fiscal effects of ECD investments. But there are other savings to society from reduced crime. These include the value of material losses and the pain and suffering that would otherwise be experienced by the victims of crime. By 2050, these savings to individuals from less crime would amount to $127 billion (in 2004 dollars).

Another major benefit of ECD investments is their impact on the future earnings of participants.9 The initial increase in earnings occurs in 2020 when the first cohort of participating children turns 18 and enters the labor market. By 2050, the increase in earnings due to ECD investments is estimated to amount to 0.43 percent of GDP, or some $107 billion (in 2004 dollars).

The increased earnings of children who participate in a high-quality ECD program not only allow the U.S. to compete more effectively in a global economy, but also aid both earlier and future generations of children. These increased earnings will benefit earlier generations when they reach retirement age because these earnings will contribute to the solvency of Social Security and other public retirement benefit programs. Future generations will benefit because they will be less likely to grow up in families living in poverty.

A nationwide commitment to high-quality early childhood development would cost a significant amount of money up front, but it would have a substantial payoff in the future. The United States' political system, with its two- and four-year cycles, tends to under-invest in programs with such long lags between when investment costs are incurred and when the benefits are enjoyed. The fact that lower levels of government cannot capture all the benefits of ECD investment may also discourage them from assuming all the costs of ECD programs. Yet, the economic case for ECD investment is compelling.

To recapitulate, I estimate that providing poor three- and four-year-old children—20 percent of all children in this age range—with a high-quality program would initially cost about $19 billion a year. Such a program would ultimately reduce costs for remedial and special education, criminal justice, and welfare benefits, and it would increase income earned and taxes paid. Within about 17 years, the net effect on the budget would turn positive (for all levels of government combined). Within 30 years, the offsetting budget benefits would be more than double the costs of the ECD program (and the cost of the additional youth going to college).

In addition, investing in our poor young children is likely to have an enormous positive effect on the U.S. economy by raising GDP, improving the skills of the workforce, reducing poverty, and strengthening U.S. global competitiveness. Crime rates and the heavy costs of criminality to society are likely to be substantially reduced as well

Robert G. Lynch is associated professor and chairman of the department of economics at Washington College. His most recent book is Rethinking Growth Strategies—How State and Local Taxes and Services Affect Economic Development. This article is adapted with permission from "Exceptional Returns: Economic, Fiscal, and Social Benefits of "Investment in Early Childhood Development," published by the Economic Policy Institute, © 2004,


1. All but the Chicago Child-Parent Center Program had random assignment of potentially eligible children into the intervention program or the control group. The Chicago Child-Parent Center Program did not use randomized assignment but the control group did match the intervention group on age, eligibility for intervention, and family socioeconomic status.

2. Heckman (1999), pp. 22 and 41.

3. Rolnick and Grunewald (2003), pp. 3 and 16.

4. Burtless (1999).

5. Committee for Economic Development (2002).

6. The annual average impact for various types of costs and benefits per Perry Preschool Project participant, estimated by Rolnick and Grunewald (2003) of the Federal Reserve Bank of Minneapolis, was used as the baseline for the analysis. (Rolnick and Grunewald used the costs and benefits as described by Schweinhart [1993] and Barnett [1993].) The annual costs and benefits per program participant of the preschool program were adjusted for inflation and/or wage increases every year through 2050 in line with projections made by the Congressional Budget Office (June 2004).

The numbers of three- and four-year-olds entered in the estimating model were taken from recent population projections made by the U.S. Census Bureau (2004). The total costs and benefits of the preschool program were determined by multiplying the number of participants of a particular age by the average value of the cost or benefit for each year that the cost or benefit was produced by participants of that age as determined by Rolnick and Grunewald (2003). Thus, for example, the reductions in the cost of providing public education per participant were assumed to kick in when that participant entered the public school system at age 5 and were assumed to cease when that participant turned 18 and left the school system.

7. Other savings to taxpayers and boons to government budgets, such as reductions in public healthcare expenditures, are likely to exist. But, we lack the data to quantify all these other potential savings.

8. This analysis considers budget effects on all levels of government—federal, state, and local—as a unified whole. As a practical matter, the source estimates have not made such a distinction, nor should they. All levels of government share in the costs of education, criminal justice, and income support. Responsibilities have shifted in the last half-century and will continue to do so over the nearly half-century timeframe used in this analysis. Although a case can be made that ECD investments should be the responsibility of the federal government to address educational inequalities before children enter the school system, these investments could be made at any or all levels of government. This analysis focuses on capturing national effects of ECD investments.

9. The guardians of participants are also likely to experience increases in earnings since they will have more time for employment as a consequence of the day- care provided to their children by the ECD program. These earnings benefits have not been calculated for our nationwide ECD program.


Barnett, W. Steven (1993). "Benefit-Cost Analysis of Preschool Education: Findings From a 25-year Follow-up." American Journal of Orthopsychiatry, Vol. 63, No. 4, pp. 500–508.

Burtless, Gary (1999). "Risk and Returns of Stock Market Investments Held in Individual Retirement Accounts." Testimony before the House Budget Committee, Task Force on Social Security Reform, May 11.

Committee for Economic Development (2002). Preschool for All: Investing in a Productive and Just Society. New York, N.Y.: CED.

Congressional Budget Office (2004). The Outlook for Social Security. Congress of the United States, Washington, D.C., June.

Fuerst, J.S. and Fuerst, D. (1993). "Chicago Experince with an Early Childhood Program: The Special Case of the Child Parent Center Program." Urban Education, Vol. 28, pp. 69–96.

Heckman, James (1999). "Policies to Foster Human Development." Working paper 7288. Cambridge, Mass.: National Bureau of Economic Research.

Karoly, L., Greenwood, P., Everingham, S., Hourbe, J., Kilburn, R., Rydell, C.P., Sanders, M., and Chiesa, J. (1998). Investing in Our Children: What We Know and Don't Know About the Costs and Benefits of Early Childhood Interventions. Washington, D.C.: Rand Corporation.

Karoly, Lynn (2001). "Investing in the Future: Reducing Poverty Through Human Capital Investments," in Understanding Poverty, Danziger, S. and Robert Haveman (eds.), Cambridge, Mass.: Harvard University Press.

Masse, L. and Barnett, W.S. (2002). A Benefit Cost Analysis of the Abecedarian Early Childhood Intervention. New Brunswick, N.J.: National Institute for Early Education Research, Rutgers University.

Reynolds, Arthur (1994). "Effects of a Preschool Plus Follow-on Intervention for Children at Risk." Developmental Psychology, Vol. 30, pp. 787–804.

Reynolds, A., Temple, J., Robertson, D. and Mann, E. (2001). "Age 21 Cost-Benefit Analysis of the Title I Chicago Child-Parent Center Program: Executive Summary." Institute for Research on Poverty (

Reynolds, A., Temple, J., Robertson, D. and Mann, E. (2002). "Age 21 Cost-Benefit Analysis of the Title I Chicago Child-Parent Centers." Discussion Paper no. 1245-02, Institute for Research on Poverty (

Rolnick, A. and Grunewald, R. (2003). "Early Childhood Development: Economic Development with a High Public Return." Fedgazette, Federal Reserve Bank of Minneapolis, March.

Schweinhart, Lawrence (1993). Significant Benefits: The High/Scope Perry Preschool Study Through Age 27. Ypsilanti, Mich: High/Scope Press.

Schweinhart, Lawrence (2004). "The High/Scope Perry Preschool Study through Age 40: Summary, Conclusions, and Frequently Asked Questions." Ypsilanti, Mich.: High/Scope Press.

U.S. Census Bureau (2004). Projected Population of the United States, by Age and Sex: 2000 to 2050. Washington, D.C.: Population Division, Population Projections Branch, May 18.

Related Articles

Preschool Pays
High-Quality Early Education Would Save Billions
By Robert G. Lynch

The Benefits of High-Quality Early Education

Extrapolating from One Program...

American Educator, Winter 2004-2005