Public interest in and funding for public pre-K programs have significantly risen over the past few decades. In general, pre-K programs serve 4-year-olds the year before they enter kindergarten, and most are targeted toward at-risk children. However, some state and local governments have made public pre-K “universal,” meaning these programs are open to all 4-year-olds. Evaluations of universal pre-K programs in Tulsa (Okla.), Boston, Georgia, and Florida[i] have demonstrated that they increase early academic achievement and, sometimes, social and emotional skills. However, one big question about universal pre-K has not yet been fully answered: do the benefits of these programs outweigh their costs?[ii]

New evidence on the benefits and costs of universal pre-K

A new working paper released by the Upjohn Institute and Georgetown University’s Center for Research on Children in the United States suggests that the benefits of universal pre-K exceed the costs of administering such programs. This study, which I co-authored, examines the Tulsa Public Schools (TPS) universal pre-K program in Oklahoma. We found that the benefits (specifically, increased earnings and reduced crime of TPS pre-K participants as adults) outweigh program costs by almost 2-to-1. That is, for every $1 spent on TPS universal pre-K, there is a societal gain of $1.89.

These findings are significant because they represent one of the most accurate estimates to date of the benefits and costs of universal pre-K. There are a number of benefit-cost analyses that examine other preschool programs, and demonstrate that benefits of these interventions generally exceed the costs. However, most of these preschool programs are different from universal pre-K, in part because they are targeted toward at-risk children and often serve a larger age-range. The results of this study suggest that high-quality universal pre-K programs can also be a good public investment.

Estimating the benefits and costs of Tulsa universal pre-K

A preschool program’s benefits are determined by examining the outcomes of program participants as they grow into adulthood, and then comparing them to those of demographically similar individuals who did not attend the program. In the Tulsa study, TPS pre-K participants were still in high school, so the analysis is based on the estimated effects of participating in TPS pre-K on grade retention by ninth grade, which were then used to predict future earnings and crime. We focus on these benefits in particular because prior research has shown that earnings increases and crime reduction can account for almost 90 percent of social preschool benefits.

Before program benefits can be compared to costs, though, they must have a monetary value. While earnings are already in dollars, crimes committed are not. However, economists have come up with estimates for the social costs of committing various types of crimes, which are often used in benefit-cost analyses. Once all benefits are estimated in dollars, they are compared to the total costs of administering the program to obtain the final benefit-cost ratio.

How do these results compare to other preschool programs?

Benefit-cost research of universal pre-K programs is limited, but good studies do exist for other preschool programs. While not exactly an “apples-to-apples” comparison to pre-K, the most well-known preschool benefit-cost analyses focus on small, targeted, and intensive programs like the Perry Preschool and Chicago Child-Parent Center (CPC) programs. The social net benefits of these programs tend to be much higher than those found for Tulsa pre-K: analyses of the Perry program have shown benefit-cost ratios between 8.5-to-1 and 16-to-1, and a study of the CPC program revealed a ratio of 11-to-1.

These relatively large ratios make sense when you consider that the Perry and CPC programs were small and had much more intensive program components than most pre-K programs, including TPS pre-K. The Perry program included weekly home visits and lasted for up to 2 school years, while CPC included an intensive parent component, comprehensive health services, and teacher coaching, all of which could last up to third grade. In comparison, TPS pre-K was designed to last for 1 school year and did not include these supports. Although these factors make the Perry and CPC programs more costly, they also likely contribute to greater earnings and crime reduction benefits.

Additionally, the Perry and CPC studies were conducted decades ago, and the percentage of children participating in early childhood education programs as 4-year-olds has greatly risen since then. This means that the non-participants of TPS pre-K who were included in the study likely had more educational exposure than non-participants of the Perry and CPC programs, which would partially account for the greater benefits found in those earlier studies.  Another reason why the benefits are larger in the Perry and CPC studies is because of key differences in the analyses. For instance, the Perry and CPC studies included additional social benefits like reduced welfare administration costs.

However, Head Start, a much larger-scale, comprehensive preschool intervention for low-income 3- and 4-year-olds, has also been found to have a higher benefit-cost ratio of 2.5-to-1.  This is partially because the at-risk students targeted by the Head Start, Perry and CPC programs have, on average, lower earnings and  higher levels of crime as adults than students in a typical universal program This would likely amplify any effects of these programs on earnings and crime reduction.

Final thoughts

The results from the TPS study support the idea that high-quality, universal pre-K programs will likely provide net benefits to society. The documented benefits of pre-K are not as great as the ones estimated for intensive preschool programs, but universal pre-K may still be a good investment. The two types of programs have different characteristics and target different populations of children. It is critically important to note program and participant differences when comparing benefit-cost analyses of education programs, as well as any variations in the methods used between studies and in the educational environments in which the studies were conducted. Finally, while benefit-cost analyses of preschool programs should certainly be factored into relevant policy decisions, they should not be the only aspect examined. The specific needs of children and families in each state or district should also be a major consideration.

[i] However, some of these early increases in cognitive skills have been shown to fade out over time: http://www.sreb.org/sites/main/files/file-attachments/15e05_prekfadeoutdebate.pdf.

[ii] A partial benefit-cost analysis of the Tulsa Public Schools’ universal pre-K program used early cognitive test scores to predict future earnings benefits: Bartik, T. J., Gormley, W. T., & Adelstein, S. (2012). Earnings benefits of Tulsa’s pre-K program for different income groups. Economics of Education Review, 31(6), 1143-1161.