Headstart at sixty — what worked, what didn't

The Head Start design committee, chaired by pediatrician Robert Cooke, included Zigler, Urie Bronfenbrenner, Mamie Phipps Clark, and others. The committee's recommendations were ambitious: comprehensive services including health, nutrition, parental involvement, and education, delivered by community-based grantees with substantial local autonomy. The implementation was rushed. The first summer enrolled 561,000 children with a workforce largely untrained in early childhood. The architectural decisions made in this rush — community-based grantees rather than school district delivery, comprehensive services rather than narrowly educational, federal direct funding bypassing states — shaped the program permanently and explain both its political durability and its quality variability.

The Westinghouse Learning Corporation's 1969 evaluation of Head Start was sharply negative, finding that initial cognitive gains faded by first or second grade. The report nearly killed the program politically. Zigler and others mounted a defense partly methodological and partly philosophical: that fade-out reflected the absence of follow-through in the K-3 system rather than the failure of Head Start itself, and that non-cognitive benefits — parental engagement, health, nutrition — were being undercounted. The defense largely held. The fade-out finding has recurred in essentially every subsequent rigorous evaluation, and the explanation has remained roughly the same.

Janet Currie and Duncan Thomas, in a series of papers beginning in 1995, used sibling comparison designs to estimate Head Start's long-run effects. By comparing children who attended Head Start to their own siblings who did not, the analyses controlled for family-level confounders. They found persistent gains in educational attainment, particularly for white children, and reductions in grade repetition and special education placement. Garces, Thomas, and Currie's 2002 paper found long-run effects on high school completion, college attendance, and reduced criminal justice involvement among African American men. These designs are not as clean as randomization, but they are far cleaner than naive comparisons, and they consistently find effects that fade-out narratives miss.

The HSIS randomized roughly 5,000 children to Head Start or to a control group that could access other available care. End-of-program effects were small to moderate on cognitive outcomes — roughly 0.1 to 0.2 standard deviations — and somewhat larger on parenting and social-emotional outcomes. By third grade, most cognitive effects had converged to near-zero, with persistent effects only on parenting and on certain subgroups. The interpretation challenge is the counterfactual: about 60 percent of the control group accessed some other form of center-based care. The study estimates Head Start's effect against a mixed alternative, not against no care. The fade-out is partly real, partly a measurement artifact of the convergence of treatment and control conditions over time.

Head Start has always been more than preschool. The program provides dental and vision screening, immunizations, nutrition support, developmental screening, family case management, and parent engagement. Evaluations focused only on cognitive outcomes miss these benefits. Sarah Anzia and Christopher Berry's work and others have documented Head Start's health effects, including reductions in child mortality from causes amenable to public health intervention. These effects are large in magnitude relative to the cost and have been chronically underweighted in the political narrative around the program. A more accurate framing would treat Head Start as a public health intervention with an educational component, rather than the reverse.

In 1995, recognizing the 0-3 evidence base, Congress authorized Early Head Start to extend services to pregnant women and children under three. Evaluations have shown modest but consistent positive effects, particularly when the program is delivered with fidelity and at higher dosages. EHS serves a small fraction of eligible families — under five percent — and the program's funding has not kept pace with the demonstrated value of the 0-3 window. The mismatch between what the science says about timing and what the funding allows is one of the cleanest examples of the policy-evidence gap in American social policy.

The single largest constraint on Head Start quality is the workforce. As of the mid-2020s, the median Head Start teacher earned roughly $35,000 per year, with limited benefits and high turnover. Comparable K-12 teachers earned roughly 50-70 percent more. The wage gap produces predictable consequences: difficulty recruiting qualified candidates, high turnover that disrupts child-teacher relationships, and chronic quality variance across sites. Raising Head Start teacher qualifications without raising compensation, as has periodically been attempted, simply shrinks the candidate pool. The workforce problem is the quality problem, and it is fundamentally a compensation problem.

Aggregate Head Start effects mask large variation across sites. The best Head Start centers produce effects comparable to the Abecedarian Project. The worst produce essentially nothing. The factors that distinguish high-performing from low-performing sites have been studied extensively: teacher qualifications, curriculum fidelity, child-staff ratios, leadership stability, integration with K-12. None of these are mysteries. The Office of Head Start's Designation Renewal System, introduced in 2011, was an attempt to address quality through accountability — requiring under-performing grantees to recompete for funding. Its effects have been mixed. The deeper issue is that the federal per-child allocation does not allow most sites to operate at the quality level the evidence requires.

Recent revisions of Head Start have emphasized two-generation models that pair child services with parental education, employment, and mental health support. The CAP Tulsa program, Aspen Institute's Ascend portfolio, and others have demonstrated that addressing parental human capital alongside child development produces additive effects. Head Start's parental engagement requirements, while long-standing, have historically been thin in execution. Two-generation models represent a more serious attempt to act on the bioecological reality that child outcomes depend on family-level conditions, not just child-level inputs.

A persistent finding across evaluations is that Head Start's effects fade partly because the K-3 system does not sustain them. Head Start children often enter kindergartens that are themselves under-resourced, that group them with children who did not have similar preparation, and that operate on curriculum and pacing that does not build on what Head Start provided. The integration problem is structural: Head Start is federally funded and community-grantee operated; K-12 is state-funded and district-operated. The seam between them is jagged. Models that bridge the seam — district-operated Head Start, Head Start-to-K transition programs, aligned curricula — show better persistence of effects. Most Head Start sites do not have these arrangements.

Head Start has, throughout its history, served disproportionately Black and Hispanic children, and has been one of the relatively few federal programs that operated meaningfully in deeply segregated and rural communities. The community-based grantee model, while it has produced quality variance, also produced cultural responsiveness that a more centralized model might not have. The program's role in African American educational history is substantial — Garces, Thomas, and Currie's findings on long-run effects for Black men are some of the strongest in the literature. The equity case for Head Start has been strong; the quality case has been mixed; the case for expansion-with-redesign is, on the evidence, overwhelming.

A serious sixtieth-anniversary revision of Head Start would: raise teacher wages to K-12 parity; require and fund bachelor's-level qualifications for lead teachers; extend duration to full-day, full-year matching working parents' actual schedules; expand Early Head Start to reach the majority of eligible families; integrate with K-3 through aligned curricula and shared assessment; implement two-generation services as standard rather than optional; and increase per-child funding to the level the evidence base requires, roughly double the current level. The total cost would be substantial — perhaps $30-40 billion annually at full implementation, against a current federal investment of roughly $12 billion. The ROI, by Heckman's framework, would still be strongly positive. The question is whether the polity will fund what it already knows works.

Head Start at sixty is a national experiment in what happens when a society commits to an idea at a level of funding below what the idea requires. The result is a program that has produced real benefits for tens of millions of children, that has demonstrated the principle that public investment in early childhood is both morally required and economically productive, and that has fallen chronically short of its own design intent. The next sixty years should not be a continuation of this pattern. The evidence base has caught up with the original ambition. The policy architecture has not. Closing that gap is the revision the founders, Zigler in particular, repeatedly called for. We are still owing it to them, and to the children.

1. Zigler, Edward, and Sally J. Styfco. The Hidden History of Head Start. New York: Oxford University Press, 2010.

2. Vinovskis, Maris A. The Birth of Head Start: Preschool Education Policies in the Kennedy and Johnson Administrations. Chicago: University of Chicago Press, 2005.

3. Garces, Eliana, Duncan Thomas, and Janet Currie. "Longer-Term Effects of Head Start." American Economic Review 92, no. 4 (2002): 999-1012.

4. Currie, Janet, and Duncan Thomas. "Does Head Start Make a Difference?" American Economic Review 85, no. 3 (1995): 341-364.

5. Puma, Michael, Stephen Bell, Ronna Cook, Camilla Heid, Pam Broene, Frank Jenkins, Andrew Mashburn, and Jason Downer. Third Grade Follow-up to the Head Start Impact Study: Final Report. Washington, DC: Office of Planning, Research and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services, 2012.

6. Deming, David. "Early Childhood Intervention and Life-Cycle Skill Development: Evidence from Head Start." American Economic Journal: Applied Economics 1, no. 3 (2009): 111-134.

7. Ludwig, Jens, and Douglas L. Miller. "Does Head Start Improve Children's Life Chances? Evidence from a Regression Discontinuity Design." Quarterly Journal of Economics 122, no. 1 (2007): 159-208.

8. Phillips, Deborah A., Mark W. Lipsey, Kenneth A. Dodge, Ron Haskins, Daphna Bassok, Margaret R. Burchinal, Greg J. Duncan, et al. Puzzling It Out: The Current State of Scientific Knowledge on Pre-Kindergarten Effects. Washington, DC: Brookings Institution, 2017.

9. Bailey, Martha J., Shuqiao Sun, and Brenden Timpe. "Prep School for Poor Kids: The Long-Run Impacts of Head Start on Human Capital and Economic Self-Sufficiency." American Economic Review 111, no. 12 (2021): 3963-4001.

10. Heckman, James J. "Skill Formation and the Economics of Investing in Disadvantaged Children." Science 312, no. 5782 (2006): 1900-1902.

11. Love, John M., Ellen Eliason Kisker, Christine Ross, Helen Raikes, Jill Constantine, Kimberly Boller, Jeanne Brooks-Gunn, et al. "The Effectiveness of Early Head Start for 3-Year-Old Children and Their Parents: Lessons for Policy and Programs." Developmental Psychology 41, no. 6 (2005): 885-901.

12. Yoshikawa, Hirokazu, Christina Weiland, Jeanne Brooks-Gunn, Margaret R. Burchinal, Linda M. Espinosa, William T. Gormley, Jens Ludwig, Katherine A. Magnuson, Deborah Phillips, and Martha J. Zaslow. Investing in Our Future: The Evidence Base on Preschool Education. Ann Arbor, MI: Society for Research in Child Development and Foundation for Child Development, 2013.