When children go on to kindergarten andhigher grades, those who have had ECErepeat grades less often and need lessspecial education assistance. Both ofthese reductions translate into cost sav-ings for schools. Children who have hadECE also commit fewer crimes thanchildren in the same socioeconomic cir-cumstances who have not had ECE. Theresulting cost savings may be significant,given the high expense incurred for eachcrime and the strong impact ECE pro-grams have been shown to have on crim-inal activity.

Schools will reap cost savings from theless direct effects of ECE, too. Annually,total education spending on K–12 schooling in Ohio is over $12 billion(Ohio Office of Budget and Management,

February 15, 2005

Federal Reserve Bank of Cleveland

Should Ohio Invest in Universal Preschooling?by Clive R. Belfield

ISSN 0428-1276

A growing body of evidence showsthat quality preschooling is associatedwith a host of benefits to students,schools, and society at large. Thesebenefits can have a positive impact ona state’s budget, too: Children whoattend preschool are not only moresuccessful throughout their schoolcareers, they also graduate and go onto college more often, commit fewercrimes, and earn higher wages oncethey enter the workforce. Researchsuggests that making public pre-schooling available to all childrenwould be a cost-effective way toimprove academic performance. ThisCommentary explains why makingpublic preschooling available to allchildren in the state would be a wiseinvestment for Ohio.

Ohio has almost 150,000 three-year-old children; however, fewer than 41,000are covered by publicly supportedpreschool programs, and the majority ofthese are in special education or HeadStart programs. Preschooling is associ-ated with a number of positive outcomes,from higher test scores, graduation rates,and college progression to reductions inspecial education, grade repetition, andcrime. One might wonder why it isn’tmade available to every child.

At issue is whether the benefits wouldoutweigh the costs of providingpreschool programs to a larger propor-tion of children. Currently two states,Georgia and Oklahoma, offer universalpreschooling, but many others are con-sidering expanding their coverage.Should Ohio make publicly fundedpreschool programs available to all chil-dren across the state?

■ The Economics of Preschooling

Economists think of education as aninvestment that enhances children’sprospects and yields economic payoffsfor society (see Carneiro and Heckman,2003). Early childhood education (ECE)is no exception to this rule. In fact, mostrecent research finds that investing earlyis a very powerful investment and thatwhat hinders children from learning inschool (and going on to college) is thelack of adequate preparation at an earlier age.

But would an investment in universalECE make economic sense for Ohio? Toanswer that question, the investmentshould be viewed within a balance sheetframework, relating the costs of provid-ing the investment to its anticipatedfuture benefits.

On one side of the balance sheet is thecost of providing ECE programs. Thecost ingredients include facilities, materi-als, and salaries for administrative direc-tors, teachers, and assistants. The actualcosts could vary considerably, dependingon how the programs are designed andthe supplemental services provided.High-quality programs cost more to pro-vide. Class sizes are smaller, and teachersare more likely to be certified and experi-enced. Meals are provided, and childrenare screened for health and learning prob-lems. One of the most successful ECEprograms, the High/Scope PerryPreschool program, cost approximately$7,000 per child annually. For the sake of comparison, Head Start programs inOhio average $5,500 per child, and K–12education costs an average of $8,441 perchild per year.

On the other side of the balance sheet arethe benefits of wider ECE participation.These are numerous. Some benefitsaccrue in the short term, while the childis enrolled and immediately after leavinga program. Some accrue in the mediumterm, as the child progresses throughschool. Some accrue over the longerterm and through adulthood, with entryinto the labor force.

The short-term benefits are obtained pri-marily by the children who participateand include the obvious—improved academic achievement—but also higher-quality health (because they receiveimmunizations and early diagnosis oftreatable conditions) and better nutrition.Children enrolled in ECE are also lesslikely to suffer neglect or abuse. Stateand federal governments may also bene-fit at this stage, because some parentswill be able to enter the workforce andgenerate tax revenue while their childrenattend preschool.

2004). Any improvement in the profi-ciency of a cohort of children enteringschool may yield cost savings, andexpanding ECE programs will generatesuch improvements in proficiency. Academic advantages for children whoparticipate in ECE programs are well-established, but in a generalized programthere will be spillover impacts for otherstudents as well. These spillovers comethrough peer effects: More able studentsenhance the learning of their classmatesby not disrupting class discussions, forexample, or by influencing their aspira-tions and values. These effects are alsowell-established. A second source oflearning productivity gains are the moregeneral behavioral advantages that arisewhen students are better prepared forschool. Good student behavior relievespressure on school resources, both forteaching (teacher turnover is reduced andfewer substitutes are needed, for exam-ple) and the noninstructional aspects ofeducation (fewer disciplinary, security,and custodial services are required).These productivity gains, and the costsavings associated with them, will accruewhere participation in ECE programs ismore widespread.

Over the long term, gains from ECE arerealized as the children enter the laborforce themselves, earn higher salaries,and contribute larger tax payments.

These benefits have been established in anumber of high-quality, peer-reviewedresearch studies, although not every ben-efit was obtained in each study (see table1 and Gilliam and Zigler, 2000, in therecommended reading for details forindividual ECE programs). High-qualityprograms, such as the High/Scope PerryPreschool program, the AbecedarianEarly Childhood Intervention project,and the Chicago Child-Parent Center andExpansion program, have shown thestrongest effects, clearly demonstratingthat they improve outcomes in earlychildhood and have long-lasting impacts.

Prior economic analyses have found thatthe public benefits easily outweigh thecosts. However, these analyses havebeen small scale and focused mainly ontargeted programs for at-risk children.These are children for whom preschoolprograms might be expected to have thehighest payoff. (They are at the highestrisk of being involved in crime and rely-ing on welfare). What is needed is infor-mation on what would happen if oppor-tunities were expanded for all children,

taking into account Ohio’s labor market,school system, and crime rates.

■ Economic Benefits for OhioWhat kind of economic benefits couldOhio expect to gain if it decided toexpand its publicly funded early child-hood education programs to more chil-dren? It is possible to answer that ques-tion by simulating the economicconsequences of the proposed expan-sion. (This simulation is detailed in fullin Belfield, 2004.)

We simulate the consequences ofexpanding public preschooling to thosewho receive no public provisionpresently. To simplify the analysis, wecalculate the effects of making thischange for a single cohort of three-year-olds—coverage would be available to allOhio children aged three in 2004 andwould extend for the two years beforethey enter kindergarten.While the pro-gram would be offered to all three-year-olds, we assume only an additional 40percent will choose to enroll. We basethis on the take-up rates that have beenobserved in states that offer universalprovision. In Ohio, this policy wouldmean that an extra 43,000 three-year-olds would attend public preschooling(from the initial base of 5,000 children).We also assume that the preschoolingoffered would be of a quality sufficientto generate the outcomes identified inprior research. Specifically, it wouldneed to meet a standard defined by therating it achieves on the revised EarlyChildhood Environment Rating Scale.We assume a rating of at least 5, whichcorresponds to programs such as theHigh/Scope Perry Preschool program, ahighly successful program.

What would this investment cost?Assuming per-child, per-year expendi-tures of $5,900, the expanded programwould cost $482 million over the twoyears the child is aged three and four.This is a generous amount (above whatis currently spent per child on HeadStart); it compares reasonably well withamounts spent in Ohio public schools onK–12 education; and over two years itcomes close to the resource commitmentfor exemplary programs (which are oftenshorter durations). Importantly, it shouldguarantee beneficial outcomes for partic-ipants and for the state.

The economic benefits of this investmentin ECE can be estimated using state-level data from the Ohio Children’s Budget, evidence from large-scale

national datasets, and the results fromfield trials. Only the economic benefitsthat accrue to society are estimated,leaving out those that accrue to individ-ual students and their families. Eachbenefit is calculated using conservativeassumptions. We also calculate thesebenefits in terms of their present value,that is, we take into account the fact thatbenefits that accrue years after theinvestment are worth less now.

The results of the simulation show thatOhio would gain across four domainsfrom investing in universal preschooling.

Educational Cost Savings. School sys-tems will save primarily because theywill be able to reduce both special edu-cation expenditures (fewer children willneed it) and the total cost of educatingeach child to graduation (fewer childrenwill repeat a grade). Schools could alsosave on overall spending and achieve thesame outcomes as before (because stu-dents will be more proficient learnersand less disruptive). To appreciate whythese factors should make such a differ-ence, consider the costs of special edu-cation and grade repetition. In fiscalyear 2003, the average per-pupil spend-ing for each year of regular educationwas $8,441. Per-pupil spending on eachyear of special education was propor-tionately higher, $16,038. Children threeyears old in 2004, depending on whichtrack they follow over the course of theirK–12 education, will receive presentvalue expenditures over the next 12years of $69,199 if they do not repeat agrade or receive special educational ser-vices, $135,491 if they receive specialeducational services, or $74,097 if theydo repeat a grade but do not receive special educational services. Assumingthe impacts of preschool are only one-quarter as strong as those found in pub-lished studies, the overall saving to theschool system would be $242 million.

Higher Tax Revenues. Tax revenues willincrease immediately because parentswill be able to work while their childrenare in preschool and also later, becausechildren who had ECE will work andearn more than they would have other-wise. Using models of expected earningsfrom the census and average tax rates,ECE as proposed here would generate anadditional $19 million in parental taxcontributions and $120 million in addi-tional taxes paid by the participants asthey grow up and enter the labor force.

Lower Expenditures by the CriminalJustice System. Perhaps the largesteffect of prekindergarten programs is oncrime: Participants in ECE programsreport lower rates of juvenile crime,adult crime, and less time spent on pro-bation or in prison, all of which reducethe pressure on criminal justice systembudgets. Three separate methods areused to calculate the effect of having

43,000 more children progress throughprekindergarten; taking the average ofthese three methods, savings to thecriminal justice system would amountto $375 million.

Lower Health and Welfare Expendi-tures. Preschooling has been found toreduce the prevalence of risk factors asso-ciated with problem health conditions;there are also health gains associated with

screening, immunization, and nutrition.Other studies find very strong impacts onindicators of child welfare, such as courtpetitions of child maltreatment. In itsChildren’s Budget, Ohio commitsresources for extensive services thataddress the health and welfare needs ofchildren. With wider participation inECE, these resources can be reduced, aschildren who have had ECE are lesslikely to require the state’s welfare pro-grams and health support services. Onvery conservative assumptions, expand-ing ECE programs would save around$25 million on these services.

■ The Net Returns fromInvesting in Universal Preschooling

Table 2 brings the costs and benefitstogether. The costs of $482 million areeasily offset by the total cost savingsfrom expanding early childhood educa-tion of $782 million. The net savings are$299 million. This yields a benefit–costratio of 1.62, which means that for every$1 invested, returns to the state are$1.62. This strongly suggests that theopportunity of universal preschoolingfor all three- and four-year-olds is worththe investment for the state of Ohio,without accounting for the benefits tothe children and their families.

This result is based on an economicmodel. Inevitably, such models are onlyas good as their assumptions, in this caseabout the impacts of early childhoodprograms and their economic conse-quences. Given the high quality of theresearch evidence and the availability of new data, it is possible to substantiatemany of the assumptions about impacts.For costs data, state-specific informationis applied. (Although budgetary infor-mation is far from perfect, Ohio’s Chil-dren’s Budget is particularly compre-hensive in detailing the investmentsmade in children.) Most importantly, ahighly cautious set of assumptions isapplied. Sensitivity analysis—varyingthe required investment and the cost sav-ings—shows that there are no plausiblescenarios in which the costs exceed thebenefits. The overall conclusion is there-fore robust to alternative assumptions.

■ SummaryShould Ohio offer all children theopportunity to attend high-qualitypreschool for two years before theyenter kindergarten? The analysis weconducted here is aimed at answering asimple question, namely, whether there

TABLE 1 BENEFITS/COST SAVINGS FROM EARLY CHILDHOOD EDUCATION PROGRAMS

n Enhanced academic achievementn Improved health and nutritionn Increased well-being and less abuse

Short term

n Higher tax revenue from parents’ freetime

Medium term

n Greater school system efficiency(through reduction in special educationand grade repetition, higher learning productivity, and reduced pressure onschool resources)

n Reduced abuse and neglectn Lower reliance on public health care

Long term

n Higher likelihood of graduation andcollege enrollment

n Higher probability of higher wages and employment

n Lower teen pregnancy and delinquency

n Increased income tax revenuesn Lower welfare dependencen Reduced delinquency and crimen Reduced educational subsidies for

college

For the child (and family) For society and the economy

TABLE 2 ECONOMIC IMPACT ANALYSIS OF EXPANDED EARLY CHILDHOOD EDUCATION (ECE) PROGRAMS IN OHIO

Cohort entering kindergartenin 2006 with expanded ECE

Present value ($ million) provision from 2004–06

Pre-K ECE investment cost $482.40Savings

School system cost savings $241.89Tax revenues from earnings $139.52Criminal justice system cost savings $375.41Health and welfare cost savings $ 24.76

Total fiscal benefits $781.58Net fiscal cost savings $299.19Benefit–cost ratio 1.62

NOTE: Present value figures are discounted over the child’s educational span, from K–12, at a discountrate of 5 percent. Economic values are in 2003 dollars.

Federal Reserve Bank of ClevelandResearch DepartmentP.O. Box 6387Cleveland, OH 44101

Return Service Requested:Please send corrected mailing label tothe above address.

Material may be reprinted if the source iscredited. Please send copies of reprintedmaterial to the editor.

Clive R. Belfield is an assistant professor inthe Department of Economics at QueensCollege, City University of New York.

The views expressed here are those of theauthor and not necessarily those of the FederalReserve Bank of Cleveland, the Board of Governors of the Federal Reserve System, orits staff.

Economic Commentary is published by theResearch Department of the Federal ReserveBank of Cleveland. To receive copies or to beplaced on the mailing list, e-mail your requestto 4d.subscriptions@clev.frb.org or fax it to216-579-3050. Economic Commentary is alsoavailable at the Cleveland Fed’s site on theWorld Wide Web: www.clevelandfed.org/research.

We invite comments, questions, and sugges-tions. E-mail us at editor@clev.frb.org.

PRSRT STDU.S. Postage Paid

Cleveland, OHPermit No. 385

is compelling economic evidence infavor of expanding early childhood edu-cation programs in Ohio. On the evi-dence we have seen, there is likely to bea very strong economic payoff.

The proposed policy is significant andambitious, but not infeasible. Certainly,we should not expect universal pro-grams to generate the level of economicreturns that have been found in acade-mic studies: These are based on small-scale programs targeted to at-risk populations. Expanded programs willgenerate smaller impacts per child. Nev-ertheless, based on the economic evi-dence currently available, and applyingthat evidence to Ohio, the case for pub-lic investments in universal preschool-ing is strong.

■ Recommended Reading Clive R. Belfield. 2004. “Investing inEarly Childhood Education in Ohio: AnEconomic Appraisal.” Center for Ameri-can Progress, Working Paper. Availableat <www.americanprogress.org/atf/cf/{E9245FE4-9A2B-43C7-A521-5D6FF2E06E03}/belfield_report.pdf>.

Pedro Carneiro and James J. Heckman.2003. “Human Capital Policy.” InInequality in America: What Role forHuman Capital Policies? edited byJames J. Heckman and Alan B. Krueger.Cambridge, Mass.: MIT Press.

Walter S. Gilliam and Edward F. Zigler.2000. “A Critical Meta-analysis of AllEvaluations of State-Funded Preschoolfrom 1977 to 1998: Implications forPolicy, Service Delivery, and ProgramEvaluation.” Early Childhood ResearchQuarterly, 15: 441–73.

Ohio Office of Budget and Manage-ment. 2004. Executive Budget for Fiscal Year 2006 and 2007, Tab E,“Agency Budget Recommendations.”<www.omb.ohio.gov.budget/execu-tive/0607/bb0607_e.pdf>.

Ohio Office of Budget and Management.2004. The Ohio Children’s Budget.<www.obm.ohio.gov/Information/bud-get/bluebook0405/tab_d/child.asp>.