brooking institute report: investing in young children

8/8/2019 Brooking Institute Report: Investing in Young Children

1/104

Investing inYoung ChildrenNew Directions in Federal Preschool andEarly Childhood Policy

Edited by Ron Haskins and W. Steven Barnett


2/104

The Brookings Center on Children and Families works to promote policies which advanceopportunity in America. Under the leadership of Brookings Senior Fellows Isabel Sawhill and

Ron Haskins, the Centers team of experts employs simulation modeling capabilities to create the

most effective strategies for reducing poverty, mitigating inequality, and promoting opportunity.They leverage Brookings convening power to bring policymakers, the media, and key

stakeholders to collaborate on sustainable reforms at both the federal and state level.

The Center on Children and Families (CCF) responds quickly to the pressing issues of the day

with timely and innovative policy recommendations, independent analysis, and outreach

activities to connect our work to the public and policy makers. CCF focuses on core issues that

will remain at the heart of domestic social policy debates for at least the next decade:

The effects of the recession on disadvantaged families Increasing economic mobility and opportunity for low-income families The key role of education in increasing economic mobility Reducing the growth of single-parent families

Investments in children to increase their access to opportunity The enormous fiscal challenge of a growing elderly population

The National Institute for Early Education Research (NIEER) conducts, archives, and

communicates multidisciplinary research on the education of children from birth to age 8. TheInstitute is a source of independent research-based information and technical assistance

regarding early childhood policy and practice. The Institute's research program informs the

development of effective policy and practice and to brings transparency to early childhood policyby monitoring national and state policies with particular attention to access, quality, and

expenditures. Annually NIEER publishes the State of Preschool Yearbook which describes eachstates early education policies in detail and ranks state efforts in educating children at ages threeand four. The Institute also conducts evaluations of early childhood programs, promotes the use

of continuous improvement processes in early childhood programs, and develops instruments for

the assessment of learning and teaching. The Institutes Directors are Dr. Steven Barnett and Dr.Ellen Frede.

NIEER is a unit of the Graduate School of Education at Rutgers University in New Brunswick,NJ. The Institutes web sitenieer.orgis a source of daily news about early childhood

education, a weekly blog, an award winning magazine, facts and figures, policy briefs and

research reports, videos on early childhood policy and practice, peer-reviewed research articles,

and state policy data from the annual Preschool Yearbooks dating back to 2002. Although amajor aspect of NIEERs mission is to translate research for policy makers and the general

public, the Institute conducts rigorous scientific studies including benefit-cost analysis and

randomized trials in the US and internationally. NIEERs research is widely published inacademic journals. Recent publications include Young English Language Learners: Current

Research and Emerging Directions for Practice and Policy (Teachers College Press, 2010) edited

by Eugene Garcia and Ellen Frede, and Preschool Debates (Brookes Publishing, in press) editedby Edward Zigler, Steven Barnett, and Walter Gilliam.


3/104

Investing in Young Children

New Directions in Federal Preschool and Early Childhood Policy

Edited by

Ron Haskins and W. Steven Barnett

September 2010


4/104

This research was funded by The Pew Charitable Trusts. Any opinions, estimates or conclusions

expressed herein are those of the authors and do not necessarily reflect the position of the

Trusts, the Brookings Institution or the National Institute for Early Education Research.


5/104

CONTENTS

1 Introduction: New Directions for Americas Early Childhood Policies


EARLY HEAD START

29 Getting the Most out of Early Head Start: What Has Been Accomplished and

What Needs to be Done

John M. Love and Jeanne Brooks-Gunn

39 Ten Ideas for Improving Early Head Startand Why the Program Needs Them

Nicholas Zill

HEAD START

49 Leave No (Young) Child Behind: Prioritizing Access in Early Education

Jens Ludwig and Deborah A. Phillips

59 Head Start: Strategies to Improve Outcomes for Children Living in Poverty

Craig T. Ramey and Sharon Landesman Ramey

HOME VISITING

69 The Nurse-Family Partnership

David Olds

79 Strengthening Home-Visiting Intervention Policy: Expanding Reach, BuildingKnowledge

Deborah Daro and Kenneth A. Dodge

COORDINATING PROGRAMS

89 Coordinating Americas Highly Diversified Early Childhood Investment Portfolio

Walter S. Gilliam


6/104


7/104

This volume explores whether the nations early childhoodprograms are boosting child development and preparing childrenfor schooling and proposes reforms that would improve theprograms. The volume contains contrasting papers on the successof Head Start, Early Head Start, and home-visiting programs andon policies that would improve these three programs. In thisoverview paper, we detail government spending on earlychildhood programs, review the number of children enrolled ineach type of program, review the papers on the three programs

and an additional paper on program coordination, andrecommend policies that would increase the returns produced byearly childhood programs.

NEW DIRECTIONS FOR AMERICAS EARLYCHILDHOOD POLICIES1


Ron Haskins is a Senior Fellow in the Economic Studies program and Co-Director of theCenter on Children and Families at the Brookings Institution and Senior Consultant atthe Annie E. Casey Foundation.

W. Steven Barnett is a Board of Governors Professor and Co-Director of the NationalInstitute for Early Education Research (NIEER) at Rutgers University.


8/104

2 HASKINS AND BARNETT

The Obama administration has shown agreat commitment to expanding andimproving early childhood programs anda willingness to make difficult budgetarydecisions based on hard evidence about

program effectiveness. Thus, withgenerous support from The PewCharitable Trusts, the Center onChildren and Families at the BrookingsInstitution and the National Institute forEarly Education Research at Rutgersasked several of the nations leadingscholars to examine current policies andevidence regarding the impacts of earlychildhood programs supported by federalpolicy and, in light of the best available

evidence, to propose reforms theythought would improve current programsindividually and collectively. Theirreviews cover Head Start, Early HeadStart, home visiting, and somediscussion of child care in the context ofcoordinating early childhood policy.There are at least four reasons webelieve now is the right time to conductthis review and to propose newdirections for federal early childhoodpolicy.

First, most of the major federalpolicies have been in effect for at least adecade, providing adequate time to judgetheir value after full implementation.The field of early education now hasconsiderable information on howfederally funded programs operate andtheir effectiveness. Second, despite somecommon aims, each federal program hasa unique history characterized by its own

goals, administrative structures,financing, and interest groups. As oftenhappens when policymakers createseveral policies with similar oroverlapping purposes, some trimming orcoordination between the programsmight be in order. Third, the federalbudget is on an unsustainable path

creating the virtual certainty that federaltaxes will have to be raised and spendingwill have to be cut soon.

2Similarly,

states collectively face one of their worstfiscal situations ever and can be

expected to continue their search forboth new revenues and program cuts.3Thus, although early childhoodprograms account for a small percentageof federal and state governmentspending, the years of steady ifsomewhat bumpy increases in fundingfor such programsincluding childcaremay soon come to an end and thepotential is great for cuts in the nearfuture. These fiscal considerations make

efficient and effective use of resourcesfor early childhood programs imperative.Finally, the Obama administration hasmade a commitment to improving andeven, at least temporarily, expandingfederal support for some early childhoodprograms despite the fiscal constraints.Congress may be expected to approachsuch proposals warily, at best. For allthese reasons, we believe the time hasarrived to provide a frank assessment ofearly childhood policy and to makerecommendations based on programperformance and fiscal realities.

We begin by reviewing the mostimportant numbers needed to analyzecurrent policy; namely, governmentexpenditures on the major programs andthe number of children receiving varioustypes of services. We then turn to a briefoverview of the major programs, withspecial attention to the evaluations of

program effectiveness, the details ofwhich are provided by the papers thatfollow. Having reviewed demand,spending, and program effectiveness, weconclude with a set of policyrecommendations for the Obamaadministration and Congress.


9/104

NEW DIRECTIONS FOR AMERICAS EARLY CHILDHOOD POLICIES 3

EARLY CHILDHOOD EXPENDITURESAND ENROLLMENTS

Table 1 presents estimated andprojected expenditures for the majorfederal early childhood programs.Although our projections are roughestimates, it is clear that there has been asubstantial infusion of public fundsbetween 2008 and 2011, some of itthrough the American Recovery andReinvestment Act of 2009 (ARRA).

4

Although much of the ARRA fundinghas been spent, the CongressionalBudget Office indicates that some childcare funds will be spent after 2011. Ourprojections for 2011 also includeincreases in the presidents 2011

proposed budget, though we do notknow as of this writing that Congresswill approve the continued funding (wedid not include the Early LearningChallenge Grants which appear unlikely

to be funded). Early Head Start, HeadStart, child care, and home visitation arelikely to have received substantialpermanent increases by 2011. Preschoolspecial education and early interventionare not expected to receive a permanentincrease. It appears that total federalspending across all these programs couldrise from just over $17 billion to over$21 billion, an increase of $4.6 billionfrom 2008 to 2011, not including the

increase in tax credits which could addanother $2.2 billion.5 State and local

Program4

2008 2010 2011 Federal Budget

Head Start (excluding Early Head Start)4a

$6.2 billion $6.7 billion* $7.2 billion*

Early Head Starta

688 million 1 billion* 1.7 billion*

Child Care Subsidies4b

5.2 billion 5.7 billion* 6.5 billion*

Child Care Food Programc

1.3 billion 1.4 billion 1.5 billion

Tax Credits (CDCTC and DCAP)4d

2.2 billion 2.2 billion

DoD Child Caree

300 million 750 million 800 million

Title I Preschool4f

400 million 500 million* 550 million*

Preschool Special Education (IDEA Part B,

Sect. 619)4g

374 million 574 million* 374 millionEarly intervention for infants and toddlers

with disabilities (IDEA Part C)4h

436 million 632 million* 649 million*

Home Visiting4 i

0 100 million 250 million

Total** $17.1 billion $19.5 billion $21.7 billion***

*Totals include additional ARRA funds for FY 2010 and FY 2011.

Sources: See endnote 4.

Table 1: Spending on Children Under Age 5 by Major Federal Early Childhood Programs

***This total includes tax credits continuing at 2010 levels. President Obama has announced his intention to increase

the credit amount, "nearly doubling" its value; see Eric Karolak, FY 2011 Federal Budget Process Begins with Bold

Proposal, Child Care Exchange (March/April 2010), p. 40.

**Note that part of the reason for the significant increase from 2008 is ARRA funding, which accounts for $1.4 billion

of the 2010 funding and $1.6 billion of the 2011 funding.

Expected increase since

Obama proposal raisesCTCDC eligibility limits


10/104


governments provide additional supportfor early childhood programs, much of itfor children with special needs as federalexpenditures for preschool specialeducation and early intervention cover

only a small fraction of the costs of theseprograms.6 State and local expendituresfor pre-K passed $5.7 billion by 2009,up $500 million from 2008, but are notexpected to increase appreciably through2011.7 State expenditure on child carefor children under age 5 is estimated toadd at least another $2 billion annually.

8

Thus, total federal and state spending onpreschool programs and child care islikely to exceed $31 billion in 2011. A

major purpose of this volume is topropose reforms that would increase thereturns on this substantial investment inchildrens development.

Exactly how many children areserved by public programs is less easilydetermined. Table 2 presentsparticipation rates in center-basedprograms in the two years prior tokindergarten by family income, asreported by parents in 2005. For childrenwho are in more than one arrangement,Head Start is always reported if this isone of the arrangements; otherwise thearrangement providing the most hours isreported. Very few parents reportedregular multiple center-basedarrangements (supplementary home-based arrangements are more common).9As can be seen, roughly 75 percent ofchildren are enrolled in a center duringtheir 4-year-old year and 50 percent

during their 3-year-old year. Childrenalso receive care by relatives and non-relatives outside their homes, but rarelyat age 4 and somewhat more frequentlyat age 3. About half of children underage 3 receive nonparental care oreducation, and this is more likely to behome-based than center care.10 Infants

and toddlers in low-income families aremore likely to be at home with theirparents; about 40 percent receive regularnonparental care.11

The number of children enrolledin Head Start was somewhere betweennearly 800,000 and just over 900,000 in2009, depending on how enrollment isdefined (continuously enrolled orenrolled at some time during the year)and reported. However, no matter whatfigure is used, it is clear that Head Startenrollment declined slightly from 2006to 2009, and there was no substantiveenrollment increase in 2010 (an increaseof about 14,500 children is projected for2011). Most of the increased expenditurefor Head Start (not including Early HeadStart) has been devoted to cost of livingincreases for staff, programimprovements, staff training, and similarpurposes.12

From the data in table 2, itappears that there are numerous income-eligible children who are not in HeadStart. Since 2005, some of these children

may have instead been taken up by theexpansion of state pre-K programs,which have not received any targetedfunding from the federal government.Early Head Start has been quite small,but is expected to increase enrollment bymore than 50,000 children by 2011, alarge increase over the 66,000 served in2009, but a tiny fraction of poor childrenunder age 3.13

With an enrollment of even900,000 Head Start would servesomewhat less than half the nationsover two million 3- and 4-year-olds inpoverty (10 percent of Head Startenrollees are permitted to be over thepoverty line and additional children maybe enrolled up to 130 percent of poverty


11/104


if all children under the poverty line are

served in a given area). However,poverty is a moving target, and perfecttargeting is not possible. As familyincomes rise and fall, some children whoare poor at entry to Head Start will notbe poor six, twelve, or eighteen monthslater, while others who did not qualify atthe beginning of the school year willbecome poor later in the year. Thisensures that Head Start will alwaysinclude some families who are not poor

when income is measured later in theyear (as in table 2) and will always misssome families who are poor. In addition,a recent Government AccountabilityOffice (GAO) study reported that, at theurging of local Head Start staff, someineligible families misrepresent theirincome in order to enroll their children

in the program.14 How much this

misrepresentation (as opposed tochanging family circumstances)contributes to over-income enrollment isunknown, but the bottom line is thatwhen measured in the spring, about halfthe children enrolled in Head Start arenot poor, and about 18 percent of thechildren served by Head Startare noteven in the bottom 40 percent of familiesby income considering all childrenacross both age groups. Obviously, some

error is possible in the self-report, butthe overall numbers are consistent withHead Starts own enrollment figures.Head Start appears to serve manychildren who are not poor. While mostof those are low-income, of the childrenin the bottom income quintile, all ornearly all of whom are in poverty, Head

Type of center

Percent of

all children

1st

($100,000)

3-year-old cohort

Head Start 8 20 9 7 1 1

Special Education 4 1 3 3 3 10

Privatefee paid 32 15 18 27 51 66

Privateno fee 4 7 4 2 1 2

Other Public 3 3 3 4 3 3

Total 51 45 38 43 59 82

4-year-old cohort

Head Start 13 29 23 4 5 0

Special Education 6 2 5 10 7 4

Privatefee paid 36 12 18 37 52 72

Privateno fee 6 10 6 7 5 3

Other Public 13 11 13 11 16 12

Total 74 64 64 69 85 90

Percentage within each family income quintile

Table 2: Ages 3 and 4 Preschool Program Participation, Spring 2005

Source: Estimates from the 2005 National Household Education Survey. Data are reports in the spring of 2005 for school age

cohorts so that children age 4 are those expected to attend kindergarten in the fall of 2005 (some of whom have already turned

5 by the time of the interview) and children age 3 a re those expected to attend kindergarten in fall of 2006. Head Start is listed

as the participant's primary enrollment whenever it is rep orted; otherwise primary enrollment is the one i n which the child spends

the most time. Few children are reported to participate in multiple types of centers.


12/104


Start serves only about 30 percent ofchildren at age 4 and 20 percent ofchildren at age 3.

Of course, not every poor familywith a preschooler wants to enroll theirchild in Head Start. Most Head Startprograms are open only on school days.Some offer only half-day services, whilemost others are open six hours, and someare only open four days per week. Theseschedules may not provide sufficientchild care for some parents. Unlesswrap-around care is available foradditional days and hours parents maychoose other services. Many Head Starteligible families are also eligible forstate and local pre-K programs, themajority of which also operate for six orfewer hours per day. Nevertheless, in2005 about half of 3-year-olds and overone-third of 4-year-olds from families inthe bottom fifth of the incomedistribution (poor families) were notenrolled in any center-based programand most of these were at home withtheir parents.

In sum, there is a substantialpopulation of children in poverty whocould be served by Head Start or someother program, but are not. Head Startcould serve many of these children if itwere better targeted. However, we donot know how much targeting can beimproved at a reasonable cost. Familiesincomes fluctuate over time so the targetis always moving. Some unservedchildren in poverty might not enroll if

offered the opportunity, though therecent GAO report suggests that manywould enroll. Another potentialdifficulty for Head Start is that itscenters (unlike public schools) arelocated in only some communities, andwhen families in poverty move to othercommunities Head Start may find it

difficult to recruit and transport childrento one of its programs.

Federal child care subsidies (notincluding tax credits) served about 1.2million children under five per month(550,000 at ages 3 and 4; 660,000 underage 3) in 2008.15 Not all familiesreceiving these subsidies are below thefederal poverty line. The numberreceiving subsidies is about 10 percentof all children under age 5 in childcare.16 It was anticipated that the ARRAfunds for child care subsidies wouldincrease the number of subsidizedchildren by about 150,000 per month,and 82,500 of these would be expectedto be under age 5.17 Many of the childrenand families receiving child caresubsidies and tax credits appear in thetotals in table 2. Children who onlyreceive family home day care or relativecare at ages 3 and 4 (and children underage 3 regardless of type of care) do notappearin table 2. As noted earlier, morethan half the children under age 5receiving subsidies are actually underage 3 where informal care is even morecommon.18 We note that the number ofchildren receiving a subsidy at sometime during the year is considerablyhigher than the monthly average becauseof turnover. Eligibility for subsidies iscontingent on periodic redeterminationof parental work status and income. Thisalso means that many children receivefar less than a year of continuouslysubsidized care. The quality of muchsubsidized care is not high, and it

provides little support for childdevelopment.19 As a result, most of thesubsidized child care children from poorand low-income families receive is notan adequate substitute for a regular earlyeducation program.


13/104


Federal child care tax credits,which reimburse between 20 percent and35 percent of the amount parents pay forcare up to a maximum, helped by far thelargest number of children and families.

The credit was claimed by nearly 6.6million families of children under age 13in 2007.20 About $2.2 billion of thealmost $4 billion in credits were forchildren under age 5.21 In contrast toother policies and programs, tax creditsfavor higher-income families. None ofthe child care tax credits is estimated tobenefit families in poverty, while over70 percent is estimated to be received byfamilies in the top 40 percent of the

income distribution.

22

State and locally funded pre-Kprograms serve a substantial portion ofthe population at age 4. Most targetlower-income families, but they aremuch less targeted by design than HeadStart, and in practice participation ratesin state and local pre-K are similar for

lower-, middle-, and higher-incomefamilies. In 2005, preschool specialeducation and other public programsfacilities served 19 percent of 4-year-olds. Special education and other publicprograms are primarily administered andfunded by state and local educationagencies. Enrollment in these state and

local education programs at age 4 rose to28 percent by 2009, but much of thisincrease appears to have beenaccompanied by a reduction in privateprogram enrollment or, more precisely,

by the incorporation of private programsinto state pre-K, usually with higherstandards.23 State and local publicprograms enroll few children at age 3,and this has changed little for nearly adecade.

OVERVIEW OF MAJOR PROGRAMS

In this section we summarize thefollowing papers on Head Start, EarlyHead Start, and home visiting as well asthe conflicting views expressed on boththe purposes of these programs and theireffects on childrens development. Wealso comment briefly on the paperoutlining ways these programs might bebetter coordinated.

Head Start

Initiated in 1965, Head Start aimsto improve childrens school readiness

by enhancing their social, emotional, andintellectual growth. The programprovides educational, health, nutritional,social, and other services to low-incomefamilies and children, primarily thoseages 3 to 5 in the two years prior tokindergarten. Annually, Head Start mayserve around 900,000 children at a costexceeding $7 billion.24 As noted earlier,programs vary in the number of hoursper day, days per week, and days per

year. Less than half of the childrenenrolled received a full school day, fivedays a week during the school year in2009.25 The Obama administrationexpanded Head Start funding with nearly$1 billion in the ARRA and hasproposed a permanent increase of nearly

Most of the subsidizedchild care childrenfrom low-incomefamilies receive is notan adequate substitute

for a regular earlyeducation program.


14/104


$1 billion for Head Start and Early HeadStart (see below) in its 2011 budget.26

Following a 1997 GovernmentAccountability Office report tellingCongress that the information thenavailable was not adequate to evaluateHead Start effectiveness,27 in 1998Congress ordered the U.S. Departmentof Health and Human Services (HHS) toconduct a large-scale random assignmentstudy of Head Start impacts. Theresulting studythe Head Start ImpactStudysubsequently conducted inresponse to the congressionalrequirement provides longitudinal dataon about 5,000 3- and 4-year-oldchildren who applied for Head Start inthe fall of 2002 and were randomlyassigned to treatment and control groupsfor one year (3-year-olds in the controlgroup could attend a year later at age 4).Participating children represented 383centers in eighty-four agencies aroundthe country, making this the biggeststudy of a representative sample of HeadStart children and control children everconducted. The study collectedinformation from in-person andtelephone interviews, child assessments,direct observations of program quality,and teacher ratings of individualchildren.

Two detailed reports have nowbeen published about the outcomes ofthe study. The first report, published in2005, found some modest impacts onlanguage development and parents

behavior at the end of the Head Startyear.28 The most recent report, publishedin 2010, found virtually no overallcognitive, social, or emotional impacts atthe end of first grade.29 In assessingthese results, it should be recognizedthat: (a) there were significantcrossovers between treatment andcontrol groups (no-shows and controlchildren who wound up in Head Start);(b) there were many control children

who attended child care or otherpreschool programs; and, (c) the publicschools children attended after HeadStart may have helped control childrencatch up with children who had attendedHead Start. In addition, the studyestimated the effects of only one year ofHead Start at a time and not thecombined effects of two years.

Even taking all of the mitigatingconsiderations into account, we concludethat these two reports show that HeadStart is not fulfilling its promise. Eventhe most favorable estimates of HeadStarts initial impacts are far smallerthan those demonstrated by the programsthat inspired Head Start, including theChild Parent Centers which weredelivered to disadvantaged children on alarge scale by the Chicago PublicSchools.30 Head Start learning gains alsoare distinctly smaller than the estimated

gains in some state-funded preschoolprograms for similar populations.31 TheImpact study itself indicates that thelearning trajectories of Head Start (andcontrol children) in literacy and mathwere quite slow at ages 3 and 4compared to rates of growth for the samechildren in kindergarten and first grade.

Even taking all ofthe mitigating

considerations intoaccount, we concludethat these reports showthat Head Start is notul illin its promise.


15/104


This is consistent with results of the2003 longitudinal studythe Head StartFamily and Child Experiences Study(FACES)of a random sample of HeadStart children that found minimal

increases in standard scores for literacyand math from fall to spring.32

These findings are troubling.There is evidence of long-term impactsof Head Start from othernonexperimental studies, as discussed inthe chapter by Ludwig and Phillips.33However, the experimental evidenceindicates that Head Start produced atbest minimal improvements in short-term educational, social, emotional, andhealth outcomes in the last decade.There is no plausible mechanism bywhich substantial long-term impactscould be produced without even strongershort-term results.

34While some health-

related benefits may have beenimportant in the past, advances in accessto health care for children in low-incomefamilies have obviated much of thereason for Head Start to assist inobtaining health care, as the Impactstudy suggests.35 Given that the majorpurpose of Head Start is to improvelearning and development with a view toincreasing school readiness, the HeadStart program needs major reform.

Early Head Start

Established in 1994, Early HeadStart (EHS) began operation with sixty-eight programs in 1995. In 2009, EHS

funded more than 650 programsenrolling over 66,000 children under age3 at a cost of $709 million.36 Additionalfunds were made available through theARRA for 2009 and beyond, and theadministration has proposed to fundEHS at $1.3 billion in 2011, exclusive of

ARRA funding.37 This figure is nearlydouble the level of a few years ago.

In 2008-2009, Early Head Startserved about 27,000 children in full-timecenter-based programs and 23,500 inhome-visiting programs.38 Additionalchildren were served through othermodes of service including family homeday care, and some children receivedboth home visiting and child care. TheEHS program was established at a timeof increased attention to the prenatal-to-toddler period as a period of rapidgrowth and development by the end ofwhich large gaps have opened upbetween poor children and middle-classchildren.39 Earlier intervention toprevent development of the gap wasbelieved to be important and EHS wasdeveloped to fulfill the need for suchearly intervention.

A rigorous evaluation of EHSproduced findings that were consistentwith the results of other randomizedtrials of home-visiting andcomprehensive two-generational

models.40 Although there were somemodest effects on parenting and fewersubsequent births in the two years afterprogram entry, effects on childdevelopment by the end of the programat age 3 were small in both absoluteterms and relative to the gap with moreadvantaged children. To put themagnitude of the effects in perspective,the percentage of parents who reportedreading to their child every day at age

three increased from 52 percent to 57percent, and childrens vocabularyscores increased from the 10th percentileto the 13th percentile. In follow-up atage 5, no lasting effects were found forlanguage or cognitive development,while small effects were found forparent-reported behavior problems and


16/104


approaches to learning, but not for otherparent-reported measures of social andemotional development. Although it hasbeen reported that some effects arelarger for some types of programs or for

some subgroups of children, thesefindings are not the same at age 3 andage 5, suggesting to us that they resultfrom capitalizing on chance rather thanreal differences in programs orsubgroups. It appears that EHS is nomore effective at improving childrensschool readiness than is one year ofHead Start. The large increase in EHSfunding presents an opportunity forserious evaluation to develop more

effective models.

Home Visiting

Among its many provisions, thehistoric health care legislation enactedby Congress in March 2010 contained aprovision establishing a federal home-visiting program with $1.5 billion infunding over five years, starting at $100million in 2010 and gradually increasingto $400 million in 2013 and 2014.41

From beginning to end, the story ofhome-visiting programs is steeped in thesocial science communitys call forpolicy based on evidence.

Although home visiting has beenstudied since the 1960s and has beenpublicly provided on a limited scale formany years, it began to receive greaterattention from researchers andpolicymakers due to findings of positive

outcomes for the Nurse-FamilyPartnership program developed by DavidOlds in Elmira, New York, beginning in1978.42 The fundamental justification forthe Olds programand most otherhome-visiting programsis the age-oldnotion that the best way to reach the veryyoung child is through the mother. The

Olds program sends trained nurses intothe homes of disadvantaged mothersstarting before the third trimester ofpregnancy with their first child, andprovides guidance to the mother about

prenatal care, breast feeding, nutrition,smoking, alcohol consumption,employment, and many other topics vitalto child development and the mothersresponsibilities and opportunities. Thevisits continue throughout the pregnancyand, with decreasing frequency, throughthe first and second years of the babyslife. A major characteristic of theprogram is that the mother develops aclose relationship with the nursea

respected authority figurewhose majorgoal is to help the mother make gooddecisions in her personal life and for herbaby.

In two carefully controlledrandomized trials Olds replicated hisElmira program, with some plannedvariations, in Memphis and Denver.Meanwhile, his program developed asubstantial national following and Olds,eschewing rapid expansion of his modelprogram, tried to control expansion inorder to maintain quality. Even so, by2009 the Nurse-Family Partnershipprogram was serving mothers andchildren in twenty-eight states, and theprogram had gained international noticeand was being implemented in England,Australia, and other countries.

During the 2008 presidentialcampaign, Barack Obama stated on

several occasions that, if elected, hewould fund a national network of home-visitation programs based on the Oldsmodel.43 True to promise, once electedPresident Obama included funding for anurse-visiting program, modeled on theOlds program, in his 2010 budget.Although child advocates supported the


17/104


presidents proposal to establish a home-visiting program, many of them believedthat focusing on the Olds program wastoo narrow and left out other worthyhome-visiting programs.44 Advocates

were particularly concerned aboutpotential disruptions if states alreadyoperating home-visiting programs thatdid not follow the Olds model wereforced to change their programs in orderto receive federal funds. Theadministration and Congress, with childadvocates playing an important role,worked out a compromise in which theObama home-visiting initiative wouldprovide two pools of funding for states.

The first pool would reserve 75 percentof the funds to pay only for programsshown by random assignment or quasi-experimental designs to have significantimpacts. In the case of home visiting,several programs besides the Oldsprogram have shown significant impacts,though not always consistently.45Programs that have some but lessevidence of success and thus showpromise would be eligible for fundingfrom the second pool that has 25 percentof the funds.

This two-tier structure andfunding arrangement for home-visitingprograms has been passed by Congressand federal guidelines are now in theprocess of development.46 Meanwhile,

the Obama administration followedalmost the same procedure in fundingteen pregnancy prevention programs aspart of the 2010 budget. It would appearthat the administration intends to fund

some social programs based on evidenceof successas indeed both the presidentand his cabinet secretaries have publiclystated on many occasions.47 Thepresident has also indicated that theadministration intends to improve orreduce funding for programs that do notwork.

However, even Olds Nurse-Family Partnership has not consistentlyproduced overall effects on the cognitiveor language development of all childrenserved by the program, although it hasshown consistent impacts on thedevelopment of children born to poormothers with fewer psychologicalresources to manage the care of theirchildren. It has also produced sustainedeffects on childrens math and readingachievement for this vulnerable group inthe Memphis trial.48 The Nurse-FamilyPartnership has yet to be evaluated bysomeone other than its originator or as afull-scale routine public program, but theObama home-visiting initiative willapparently provide just such a broad trialof the program and perhaps other home-visiting programs.

Child Care Subsidies

Federal and state governmentsprovide various subsidies for child care,

the largest of which is the Child Careand Development Fund (CCDF) andrelated state expenditures, includingfunds from the Temporary Assistancefor Needy Families program (TANF). Inits budget for 2011 the administration isproposing an increase of $1.6 billion infederal funds for the CCDF (over

The story of home-visiting programs issteeped in the socialscience communityscall for policy based

on evidence.


18/104


baseline funding, not including ARRAfunds), which would create funding foraround 235,000 additional children, butCongress has not enacted the increase asof this writing. We estimate federal

funding (including ARRA funds) onchild care subsidies will reach $6.5billion for children under 5 under currentproposals for 2011 (see table 1). TheCCDF is relatively free of federalrequirements, but states do have tosubmit an annual report, set aside asmuch as 7 percent of their funds toimprove the quality of care (in partthrough rating systems and professionaldevelopment), have minimal health and

safety regulations, and offer parents avoucher to purchase care.

Despite the quality improvementset-aside, the quality of care supportedby the CCDF has been shown to bemediocre or worse.49 A major ingredientin quality is well-trained and well-supervised teachers.

50But many child

care facilities, including both centers andfamily day care offered in the homes ofwomen who run a small business, do notprovide an enriching educationalexperience. The evidence shows thatchildren may receive a slight boost incognitive development from the averagechild care center while their social andemotional development is notsignificantly harmed.51 Thus, from theperspective of preparing low-income andminority children to better succeed inkindergarten and beyond, routinesubsidized child care is at best modestly

effective and often does not reach eventhat level. There is even some evidencethat it can be harmful.52 Under currentpolicies, subsidized child carecontributes less to improving childdevelopment than does Early Head Startand Head Start. On the other hand,subsidized child care provides a

relatively inexpensive and usually safeenvironment for families to place theirchildren while they work.

53

Significantly increasing thequality of the many thousands ofsubsidized child care facilities aroundthe country would cost billions of dollarsand would therefore substantially raisethe cost of care. Despite its increases infunding over the years, even now the

CCDF does not serve all eligiblefamilies. Research finds that child caresubsidies reduced the out-of-pocketburden of child care costs by 14percentage points among low-incomesingle parents who received subsidies.54Given constraints on governmentspending on child care, federal and statepolicy is encountering a quantity-qualitytradeoff in which increased spending onquality improvement would increase the

already high number of strugglingfamilies paying out-of-pocket costs orshifting to informal, unregulated carethat could be of even lower quality.55Some parents could even findthemselves unable to work because oflack of affordable care. While themagnitude of these negative effects is

Whether theadministration willactually defundprograms that are notsucceeding remains anopen question withgreat implications forHead Start, Early HeadStart, and home visiting.


19/104


uncertain, it remains clear that if publicfunds are not available to pay forincreased quality while maintaining oreven increasing the amount spent onchild care subsidies, there would be

negative as well as positiveconsequences. On the other hand, ifquality is not sufficiently high, increasesin child care subsidies could havemodest negative effects on childcognitive development and socialbehavior.56

ANALYSIS OF INDIVIDUALPROGRAMS

Given the significant federal andstate resources invested in the variousprograms to promote child developmentor provide routine care for children, andthe problems with quality and coveragejust reviewed, we turn now to a briefoverview of the papers that analyzeHead Start, Early Head Start, and home-visiting programs and the cross-cuttingpaper on how early childhood programscan be better coordinated. Authors ofeach chapter were asked to assess

evidence on the effects of the respectiveprograms and to make policyrecommendations that would enable thenation to promote the development andschool readiness of poor children in themost effective and efficient waypossible. We asked two authors or setsof authors to examine Head Start, two toexamine Early Head Start, and two toexamine home-visiting programs. Theauthors were selected to provide

contrasting views of the evidence andthe best policies to improve theprograms.

Chapters on Head Start

of the chapters that providedetailed scrutiny of Head Start, Early

Head Start, home visiting, and thepossibility of coordinating the efforts ofthese and other preschool programs.

The Head Start chapters by JensLudwig and Deborah Phillips and byCraig Ramey and Sharon LandesmanRamey provide a sharp contrast. Themajor recommendation made by Ludwigand Phillips, based on four decades ofresearch, is that the impacts of HeadStart, though modest, are nonethelessworthwhile and provide children with aboost that can be detected well into theteenage years and after.

57Thus, they

make the interesting argument that thepriority for policy should be to ensurethat as many children as possible receivea preschool program that is at least asgood as Head Start.

By contrast, Ramey and Rameybelieve that far too many Head Startcenters are ineffective because they areof such low quality. Indeed, they claimthat the poor quality and lack of impactof Head Start have been widely knownfor many years and that a culture of

silence about these shortcomings hasgrown up around the program to protectit from being cut. In fact, Head Startsbudget has grown every year from 1970to the present under both Republican andDemocratic administrations with theexception of level funding in 1975 andsmall declines in 1986 and 2008.

Consistent with the Ramey andRamey claim, there has been little

reaction by policymakers to the 2010longitudinal study report showing thatvirtually no impacts of the Head Startprogram on academic performance werefound at the end of the first year ofschooling.

58Until this year, there has

been no plan for improving the averagequality of Head Start, other than to


20/104


implement the important changesCongress mandated in 2007. The 2007legislation required that 50 percent ofteachers and all education coordinatorsin centers have at least a bachelors

degree by September 30, 2013. It alsoestablished new procedures for granteerenewal and recompetition.59 In January2010, the Obama administration releaseda document outlining a series of specificchanges they were taking to improveHead Start quality.60 In addition, basedon discussions with administrationofficials, it is clear that further changesaddressed to quality are beingcontemplated. We are encouraged by

this information and offer specificsuggestions in a later section of thispaper for what we think theadministration should do about HeadStart.

Chapters on Early Head Start

Like the two Head Start chapters,the chapters on Early Head Start (EHS)by John Love and Jeanne Brooks-Gunnand by Nicholas Zill offer a remarkable

contrast. Both EHS chapters base theirconclusions about effectiveness on alarge-scale random assignment study ofseventeen program sites. Initiated in1996, the study followed the same groupof children from 14 months to 63 monthsof age. The children were also followedup for a last time when they were in fifthgrade (not yet published). Data werecollected on cognitive, language, andsocial-emotional measures based on

direct assessment of the child or onparent report.

Love and Brooks-Gunnemphasize impacts during andimmediately following the program at 14through 36 months of age. The resultsshowed modest impacts on a range of

child and parent outcomes includingcognitive, language, and social-emotional development as well asattention and engagement. The authorsalso emphasize that some of these

effectsparticularly increased attentionand reduced behavioral problemswereobserved even two years after the end ofthe program (at age 5), although theeffects on vocabulary (except for a fewsubgroups) and school-related cognitiveabilities in literacy and mathematics didnot continue. The authors alsoemphasize the importance of impacts onparenting and the home environmentfrom ages 14 months through age 5

years. Finally, Love and Brooks-Gunnnote that these impacts were producedwhen the program was first created anddespite the fact that many children in thecontrol group were in center-basedprograms other than EHS, thus making itmore difficult to show differencesbetween experimental and controlchildren.

Zill on the other hand argues thatthe EHS impacts are mostly quite small,that many expected outcomes measuredat the various assessment points did notmaterialize (no difference between theprogram and control groups), and thatmany of the positive impacts were foundonly on parent reports while directassessments of the same or relatedmeasures showed no differences. Zillconcludes that the lack of sustainedimpacts in critical areas of childrenscognitive and language development

tells us that the program is notsucceeding. We note that when expertsdisagree to the degree shown by our twopapers on EHS (or for that matter thetwo papers on Head Start), policymakersare tempted to throw up their handsbecause theyre not sure what to believe.In such a situation, it is useful to focus


21/104


on specific quantitative results, as we dolater in our recommendations section.

Chapters on Home Visiting

Although the differencesbetween the paper by David Olds andthe paper by Deborah Daro and KennethDodge are not as stark as the differencesbetween the two Head Start or the twoEarly Head Start papers, there areinteresting differences nonetheless. Olds,not surprisingly, is fully supportive ofthe current federal policy approach thatfocuses on low-income, first-timemothers and concentrates most of itsspending on programs with the strongestempirical evidence of effectiveness.Daro and Dodge would like to changefederal policy in several major ways.They strongly object to adopting Oldsapproach as the prevailing modelbecause they believe that many familieswill be left out, that other programs havegood evidence of effectiveness, and thatmore effort should be placed onscreening to determine who needs moreextensive home visiting.

More specifically, Daro andDodge want a universal program ofhome visiting that would serve as ascreening program to determine whichmothers and babies need additional help.Under the Olds program, only first-timemothers with income below poverty (orsome other low-income cutoff) would beeligible, potentially leaving out morethan 90 percent of newborns.61

Moreover, families winding up in thechild protection system may not useprenatal care and could therefore bemissed by the Olds program.62 Daro andDodge cite evidence that the per childcost of the type of screening they favorwould be around $200. Families thatwere found to be experiencing

difficulties in child rearing or relatedissues would be eligible for a moreextensive home-visiting program. Theirarguments are primarily based onproblems that are or could be

preventable. Whether the proposeduniversal screening and referral to moreintensive home visitation could prevent asignificant amount of abuse and neglectand contribute to improved childdevelopment is largely untested.

Chapter on Coordinating Programs

The early childhood field ispopulated by a diverse array of publicprograms. Not only does the fieldinclude Head Start, Early Head Start,and home-visiting programs, it alsoincludes major federal spending on childcare programs and significant statespending on pre-K, child care, and otherprograms. In addition, the federalgovernment makes funds available forpreschool special education, for thedisadvantaged through Title I (thoughlittle of Title Is $14.5 billion is spent onchildren under 5),

63and for an

assortment of other special-purposeprograms. These programs areadministered by different funding

Programs areadministered by different

funding agencies,pursue different goals,have different rulesand regulations, andoften have differentlicensing agencies.


22/104


agencies at the federal and state level,pursue different goals, have differentrules and regulations, and often havedifferent licensing or accreditationagencies.

Walter Gilliams chapterproposes to bring some order out of thisunruly mix of programs. His biggestrecommendation is to let state pre-Kprograms continue to expand andeventually serve all 4-year-olds or atleast all poor 4-year-olds. Head Startcould then focus on what Gilliam says itdoes best; namely, providecomprehensive services, work withparents, and conduct home visits. Healso raises the possibility that Head Startcould focus its attention on 3-year-oldsand children under age 3. Implicit in thisrecommendation is the necessity forHead Start and Early Head Start tocooperate even more closely than theydo now and extend the cooperation toevery state in the nation. He also callsfor closer cooperation between childcare programs and high-qualityprograms that provide services for onlypart of each day. To facilitate full-timework or education by parents, state pre-K programs would work with child careproviders to establish wrap-aroundservices for children who attend theirpart-day program, something theyalready do on a limited basis. Similararrangements would also have to bemade for children under age 4 withworking parents served by Head Startand Early Head Start (about 10 percent

of Head Start children currently receivemore than eight hours per day).Although Gilliam does not discuss thespecifics in any detail, he raises thepossibility that both state pre-Kprograms and Head Start could helpchild care programs increase the quality

of their services through coordinated in-service training and coaching.

An important problem with anycoordination effort would be figuring outhow to consolidate the many licensingand certification agencies that nowoperate in the early care and educationfield. Gilliam points out that someprograms must now be licensed oraccredited by as many as four differentagencies, each with its own standardsand requirements. Again, Gilliam doesnot explore the details of a solution, buthe implies that a way should be found tohave programs licensed or certified byonly one agency and set of standards.This outcome would probably mean thatsome agencies relinquish control andcould be put out of business.

RECOMMENDATIONS

The Obama administration, asthe president promised during thecampaign, began with more ideas andmore specific proposals for earlychildhood programs than any previous

administration. The president put morethan $4 billion in the ARRA law forHead Start, Early Head Start, and childcare; initiated a home-visiting programthrough health care legislation; andproposed expanded funding for the childcare block grant, the child care taxcredit, and Head Start and Early HeadStart in his 2011 budget. In addition, heproposed to spend about $10 billion overten years on a program called the Early

Learning Challenge Fund, the majorpurpose of which would be to help statesimprove the quality of early care andeducation programs in their state.64 Mostof the money would be reserved forstates that have a plan for coordinatingprograms and improving quality,although some money would be reserved


23/104


for states that are just getting started inplanning and conducting a broad effortto increase the number of children inhigh quality care. The Challenge Fundpassed the House, but has not been

considered by the full Senate, althoughthe Senate Labor, Health, and HumanServices Subcommittee of theCommittee on Appropriations approved$300 million for the new program ininitial action on the 2011 spending bill.Budget realities are now rising inimportance, which could spell difficultyfor the Challenge Fund and perhapsother Obama administration earlychildhood initiatives.

Like many others, we applaudthe administrations zeal in supportingpolicies to improve the development ofpoor children during the preschool years.We also applaud the administrationsemphasis on data-based decision makingand willingness to fund or defund publicprograms based on their effectiveness. Inthe spirit of friendly advice, we havefour sets of proposalsnot all of whichimpose new costs on the already-stressedfederal budgetthat we bring to theattention of the administration and otherpolicymakers interested in getting themost out of what we spend as a nationon the care and education of childrenbefore they reach kindergarten.

Head Start

First, we support greatlyincreased attention to and bold action in

improving Head Start. Judged strictly onthe basis of impacts on childdevelopment at age 5, Head Start cannotbe judged more than modestly effective.Head Start does not accomplish its mostimportant goal because it has only smalleffects on learning and development atthe beginning of school and, at least

according to the recent nationalrandomized trial, few discernable effectson academic abilities at the end of eitherkindergarten or first grade.65

Although it is not reasonable toexpect Head Start to close the entire

achievement gap by itself, it isreasonable to expect effects severaltimes larger for both cognitive and socialdevelopment.66 If this is to happen, HeadStart needs a new approach to encourageinnovation, improve averageperformance, and eliminate persistentlypoor performance. The current processfor ensuring compliance with HeadStarts existing performance standards isnot producing these desired results.

Indeed, it appears that programs withfewer regulations and standards areoutperforming Head Start on a regularbasis, though the standards they do haveare often higher.67 Head Start should bejudged primarily by its outcomes forchildren based on actual performance,not by its ability to produce a paper orelectronic document of compliance. Theprimary focus should be on learning andteaching in the classroom. However, asothers may doubt this view, werecommend that the federal governmentdevelop a competitive research programwhich uses randomized trials tosystematically test new models of HeadStart services. These trials would alsopermit generous waivers of the HeadStart Performance Standards when

We support greatlyincreased attention toand bold action inimproving Head Start.


24/104


needed to implement new models andintegrate with other programs.

In addition, the average qualityof Head Start would be improved byterminating defective programs.Evidently Congress agrees with thisapproach. In 2007, Congress directed theSecretary of HHS to appoint aCommittee on Re-Designation of HeadStart Grantees. The Committee wascharged with recommending a system toevaluate every local Head Start programon a periodic basis for the purpose ofimproving the program, ending theprogram, or allowing other programs tocompete for the Head Start money. TheCommittee was duly appointed in 2008and reported back to the Secretary in2009 at the end of the Bushadministration. There are indications thatthe Obama administration may now beblowing some of the dust off the report.Recently, for example, HHS SecretaryKathleen Sebelius stated that we aregoing to fund only high-performingprograms, and her staff has indicated tous that they are reviewing the Re-Designation Committee report.68

In any case, what we needconsistent with the Re-DesignationCommittees recommendationsis aprogram of continuous evaluation ofevery Head Start program. We are awarethat the National Reporting System,implemented early in the Bushadministration and then terminated byCongress, was an attempt to move in the

direction we recommend.69 However, thefate of the National Reporting Systemshould not prevent the administrationfrom creating a continuous improvementsystem for Head Start that includes moreappropriate assessments of the learningand development of children enrolled inHead Start.70 Such a system would

provide feedback to teachers, staff, andadministrators on learning, teaching, andother program activities. Teachercoaches are likely to play an importantrole in such a system.71 Most programs

will improve with feedback, but thosethat continue to be identified as failingbecause they teach children poorly andproduce little or no gains in learning anddevelopment should be replaced.

Any discerning reader can tellthat we admire the emphasis the Obamaadministration has placed on earlychildhood programs and the capablepersonnel planning and administering itsvarious early childhood initiatives. Itwould be surprising if an administrationthat is rapidly gaining a reputation forthe most data- and evaluation-orientedadministration ever should take thepowerful indications that too many HeadStart programs are floundering withoutliving up to its word to do somethingrigorous about failing programs. Ourplan would at least offer a relativelylow-cost way to generate innovation,improve the quality of many Head Startprograms, and terminate failing programs.

Early Head Start

Our second recommendationaddresses Early Head Start. EHS costsmore per child than Head Start andproduces results for children that may beeven weaker than one year of Head Startdespite the fact that many EHS childrenand families receive three years of

services. Combined with the fact that thebest evaluation is more than a decadeold, the evidence of modest to poorresults leads us to recommend that theadministration take steps to innovate andsystematically evaluate promising newapproaches as well as the status quo ofEHS on program quality and outcomes.


25/104


The administration is emphasizingrigorous program evaluations as thebasis for funding decisions, but in thecase of EHS, there is really no currentevidence on which to make decisions.

Thus, the administration should create abetter system of continuous datacollection on child and family progressand on program implementation togetherwith random assignment evaluations ofinnovative EHS programs. Such anapproach is similar to what we suggestfor Head Start, though with an evenlarger scope for innovation andexperimentation and with fullrecognition of the difficulties of

adequately assessing the development ofvery young children.

Home Visiting

Not enough is yet known about

how the new home-visiting initiativewill be conducted for anyone torecommend major changes. It is difficultnot to like the administrations emphasison home-visiting program models thathave been shown by evidence fromrigorous evaluation to produce impactson mothers and children, a decision that

should result in this network of programsstarting with local projects that are moreeffective than the initial projects in manyother federal grant programs. The trick,of course, will be to figure out ways to

ensure that projects are implemented in amanner consistent with the programmodel they are following, although someroom for reasonable local adjustmentsmust be allowed. The administration willhave its hands full figuring out how tomaintain project fidelity and produceimpacts without being subjected toexternal criticisms at this early point.Advocates, researchers, and members ofCongress should be patient for three or

four years and see what these projectsproduce. Rigorous evaluation will berequired to determine whether even theOlds program produces the promisedoutcomes when implemented as regularpublic programs with a predefinedpopulation. While we agree thatprograms with the strongest evidencecurrently should be the presumptivefavorites, we also believe that it is muchtoo soon to close the door on alternativesand innovations. Our only recommendationis that the administration conduct somerandomized trials to test existing andnew approaches, including low-costscreening and referral to home visiting,and collect reliable information onimplementation fidelity and a set ofstandard outcome measures, includingassessments of childrens learning anddevelopment, for samples of children inall programs. Such studies could berandomized by place rather than byindividual mothers.

Coordination

Our fourth recommendation is themost controversial. As part of its EarlyLearning Challenge Fund or some otherlegislative vehicle, the administration

The administrationshould give some statesthe authority to usefederal funds to try tobuild a coordinated

early care andeducation system withan emphasis onprogram effectiveness.


26/104


should give some states the authority touse federal funds from the child careblock grant, Head Start, Early HeadStart, Title I, and perhaps other federalsources to try to build a coordinated

early care and education system with anemphasis on program effectiveness,particularly for disadvantaged children.72We do not anticipate that fundsnecessarily would be withdrawn fromlocal schools and Head Start agenciesand given to states, but that barriers tothese agencies co-mingling funds andactivities as part of a state plan would bewaived.

These efforts, implemented inperhaps three or four diverse states,should include third-party evaluations.States should agree to a minimum set ofconditions, including substantialcontributions of their own funding topreschool programs, specification of theirplan for increasing quality in differenttypes of services, maximization of parentchoice, adoption of good measures of

learning and development for childrenparticipating in programs, andtermination of funding for programs thatdo not meet quality standards or agreed-upon child outcomes. One or two of thestates should agree to rigorousevaluations in which they followchildren for several years after they enter

the public schools and perhaps all the waythrough their college years. In return foragreement on all these conditions, statesshould receive additional federal subsidiesfrom the Challenge Fund (or other sources)

and the federal government should pay 100percent of evaluation costs. We realizethat what we propose may seem like adaunting challenge and may be perceivedas a threat by agencies currentlyreceiving funds. But we also believe thatmany states and federal agencies wouldbe eager to take up the challenge.

CONCLUSION

The federal governments effortsto deliver on its promise, first made byPresident Lyndon Johnson in 1965, tohelp children from poor and minorityfamilies arrive on the doorstep of thepublic schools ready to learn has, for themost part, become stale and stagnant.Neither Early Head Start nor Head Starthas delivered much, and the nationschild care programs have beenrepeatedly shown to be of mediocrequality or worse. The real bottom line is

that disadvantaged children are notachieving in the public schools as theyshould or could with proper help andpreparation during the preschool years.Efforts to dramatically improve federalearly childhood programs will succeedonly if they are tied into ongoing,systematic, rigorous evaluations ofalternative approaches. Our four-partplan holds promise for breaking us outof the current stagnation by creating

continuous improvement processes andallowing the boldest and most innovativestates to use all the resources at theirdisposal to develop models forcoordinating the money we now spendand producing school readiness outcomeswe all know should be within our reach.

Disadvantagedchildren are notachieving as theyshould or could withproper help andpreparation duringtheir preschool years.


27/104


Notes

1 The authors appreciate the comments they received on earlier versions of this chapter from authorsDeborah Daro, John Love, Jens Ludwig, David Olds, Deborah Phillips, Sharon Ramey, and Nicholas Zill;and from Ben Allen of the National Head Start Association, Ruth Friedman of the House Education andWorkforce Committee, Naomi Goldstein of the Office of Planning, Research, and Evaluation in theAdministration for Children and Families at HHS, Joan Lombardi and Jeffrey Capizzano of theAdministration for Children and Families at HHS, Albert Wat and his colleagues at Pew Charitable Trusts,and Edward Zigler of Yale University. None of them are responsible for any of the content of the chapter.We also thank Julie Clover, Mary Baugh, Alex Gold, and Milagros Noresfor their help with research andpreparing the manuscript. The authors are responsible for any remaining errors.2 Alice Rivlin and Isabel Sawhill,Restoring Fiscal Sanity: How to Balance the Budget(Washington, DC:Brookings Institution Press, 2004); Alan Auerbach and William G. Gale,Dj Vu All Over Again: On the

Dismal Prospects for the Federal Budget(Washington, DC: Urban-Brookings Tax Policy Center, 2010).3 Some states are already beginning to cut funding for pre-kindergarten programs; see Elizabeth McNichol,Phil Oliff and Nicholas Johnson,Recession Continues to Batter State Budgets; State Responses Could Slow

Recovery (Washington, DC: Center on Budget and Policy Priorities, 2010 [Updated May 27]).4 Some figures in this chapter refer to amounts for fiscal years, not calendar years.Some of the numbers in Table 1 are not originally with respect only to children under 5. We scaled these

numbers to reflect either the percentage of children under age 5 or the percentage of federal money thatwould go to children under age 5; see Matthews, Child Care and Development Block Grant Participationin 2008; Administration for Children and Families, FFY 2007 CCDF Data Tables: Table 15, Child Careand Development Fund, Average Monthly Payment to Provider (Including Family CoPay) by Age Groupand Care Type (FFY 2007), (Washington, DC: U.S. Department of Health and Human Services,Administration for Children and Families, 2007). Available athttp://www.acf.hhs.gov/programs/ccb/data/ccdf_data/07acf800/table15.htm

4a Head Start and Early Head Start:For 2008 data, see Administration for Children and Families,Head Start Program Fact Sheet 2009:Fiscal Year 2009 (Washington, DC: U.S. Department of Health and Human Services, Administrationfor Children and Families, 2009). Available athttp://www.acf.hhs.gov/programs/ohs/about/fy2009.html. For 2010 and 2011 data, we used acombination of baseline data and ARRA figures. For baseline data, see Administration for Children

and Families,Justification of Estimates for Appropriations Committees, Children and FamiliesServices Programs, FY2011 Budget(Washington, DC: U.S. Department of Health and HumanServices, Administration for Children and Families, 2010): 91. For ARRA figures, see Chris Kelly,Childrens Budget 2010 (Washington, DC: First Focus, 2010): 129. Available athttp://www.firstfocus.net/sites/default/files/ChildrensBudget2010.pdf.

4b Child Care Subsidies:(CCDF)Child Care Development Fund(CCDBG) Child Care Development Block GrantFor 2008 data, for total level of subsidy, see Matthews, Child Care Assistance in 2007. Center forLaw and Social Policy (September 2009). Available athttp://www.clasp.org/admin/site/publications/files/childcareassistance2007.pdf. For proportion ofchildren under age 5, see Ibid., Child Care and Development Block Grant Participation in 2008. Tocalculate expenditure by age, see Author analysis of Administration for Children and Families, FFY

2007 CCDF Data Tables: Table 15, Child Care and Development Fund, Average Monthly Payment toProvider (Including Family CoPay) by Age Group and Care Type (FFY 2007), (Washington, DC:U.S. Department of Health and Human Services, Administration for Children and Families, 2007).Available at http://www.acf.hhs.gov/programs/ccb/data/ccdf_data/07acf800/table15.htm. For 2010 and2011 data, both baseline and ARRA (adjusted by percentage of spending on children under age 5), seeKelly, Childrens Budget 2010, pp. 30, 128.

4c Child Care Food Program:Data was adjusted by percent of children under age 5. For 2008 data, see Office of Management andBudget,Budget of the United States Government, Fiscal Year 2009, Analytical Perspectives, Table 8-10: Child and Adult Care Food Program (Washington, DC: U.S. GovernmentPrinting Office, 2008).


28/104


For 2010 and 2011 data, see Office of Management and Budget,Budget of the United StatesGovernment, Fiscal Year 2011, Analytical Perspectives, Special Topics, Table 17-10: Child and AdultCare Food Program (Washington, DC: U.S. GovernmentPrinting Office, 2010).

4d Tax Credits:(CDCTC)Child and Dependent Care Tax Credit

(DCAP) Dependent Care Assistance ProgramData was adjusted by percent of children under age 5. Adam Kent and others,Data Appendix toFederal Expenditures on Pre-Kindergarteners and Kindergarteners in 2008 and Federal Expenditureson Elementary-Age Children in 2008 (Washington, DC: Urban Institute and The Brookings Institution,2010).

4e Department of Defense (DoD) Child Care:Data was adjusted by percent of spending on children under age 5. For 2008 data, see Gail L. Zellmanand others, Options for Improving the Military Child Care System, prepared for the Office of theSecretary of Defense by the RAND National Defense Research Institute, 2008 (p. 35). Available athttp://www.rand.org/pubs/occasional_papers/2008/RAND_OP217.pdf. For 2010 and 2011 data, seeOffice of Management and Budget,Budget of the United States Government, Fiscal Year 2011,FactSheets, Department Fact Sheets, Department of Defense (Washington, DC: U.S. GovernmentPrintingOffice, 2010). The Department of Defense received an additional $240 million in ARRA fundingspecifically for the purpose of constructing new childcare centers; see U.S. Department of Defense,American Recovery and Reinvestment Act 2009: Department of Defense Expenditure Plans(Washington, DC: U.S. Department of Defense, 2009): 2. Available athttp://www.defense.gov/recovery/plans_reports/2009/march/ARRA_DoD_Expenditure_Plans.pdf.

4fTitle I Preschool:For 2008 data, see Sara Mead, FY2008 Budget Cuts Early Education Funding, New AmericaFoundation, The Early Ed Watch Blog (January 8, 2008). Available athttp://oti.newamerica.net/blog/early-ed-watch/2008/fy2008-budget-cuts-early-education-funding-353.For 2010 and 2011 data, see Laura Bornfreund, A Look at Proposed Federal FY 2011 Funding forEarly Education, New America Foundation, Early Ed Watch Blog (July 19, 2010). Available athttp://earlyed.newamerica.net/blogposts/2010/a_look_at_proposed_federal_fy_2011_funding_for_early_education-34524. To determine the percent of funding for preschool, see Daniel Ewen and HannahMatthews, Title I and Early Childhood Programs: A Look at Investments in the NCLB Era(Washington, DC: Center for Law and Social Policy, 2007). Available at

http://www.clasp.org/admin/site/publications/files/0379.pdf4g Preschool Special Education (IDEA Part B, Sect. 619):

(IDEA)Individuals with Disabilities Education ActKelly, Childrens Budget 2010: p. 36. Available athttp://www.firstfocus.net/sites/default/files/ChildrensBudget2010.pdf. For ARRA data, see U.S.Department of Education, American Recovery and Reinvestment Act of 2009: IDEA Recovery Fundsfor Services to Children and Youths with Disabilities, (April 4, 2009). Available athttp://www2.ed.gov/policy/gen/leg/recovery/factsheet/idea.html.

4h Early intervention for infants and toddlers with disabilities (IDEA part C):Kelly, Childrens Budget 2010, pp. 34, 128.

4i Home Visiting:Nurse-Family Partnership, House of Representatives Passes Historic Health Care ReformLegislation, public policy update, Nurse-Family Partnership (March 22, 2010). Available at

http://www.nursefamilypartnership.org/assets/PDF/Policy/NFP_PP_Update_March_22_20105 Eric Karolak, FY 2011 Federal Budget Process Begins with Bold Proposal, Child Care Exchange(March/April 2010), p. 40.6 W. Steven Barnett and Leonard Masse, Funding Issues for Early Childhood Care and EducationPrograms, inEarly Childhood Education and Care in the USA, ed. Debby Cryer (Baltimore, MD: Paul H.Brookes Publishing, 2003).7 W. Steven Barnett and others, The State of Preschool 2009: State Preschool Yearbook(New Brunswick,NJ: Rutgers University, National Institute for Early Education Research, 2009); Dale Epstein and W.Steven Barnett,Brief Report: Funding Cuts to State-Funded Pre-Kindergarten Programs in FY 10 &11(New Brunswick, NJ: Rutgers University, National Institute for Early Education Research, 2010).


29/104


Available at http://nieer.org/pdf/Funding_Cuts_to_State-Funded_Prekindergarten_Programs_in_FY10_FY11.pdf. These figures do not include much of the cost ofpreschool special education that is borne by states.8 Matthews, Child Care Assistance in 2007. States are not required to report amounts spent in excess offederal requirements.9

These enrollment statistics differ slightly from others in that they estimate enrollment for schoolcohorts4-year-olds are children who will start kindergarten the next year (half of whom have alreadyturned 5 by the spring before kindergarten) and 3-year-olds are those who will start kindergarten in twoyears (many of whom are 4 by the spring).10 Iheoma Iruka and Priscilla Carver,Initial Results From the 2005 NHES Early Childhood ProgramParticipation Survey, NCES 2006-075 (Washington DC: U.S. Department of Education, National Centerfor Education Statistics, 2006).11 Kristin Flanagan and Jerry West, Children Born in 2001: First Results From the Base Year of the EarlyChildhood Longitudinal Study, Birth Cohort (ECLSB),NCES 2005036 (Washington DC: U.S.Department of Education, National Center for Education Statistics, 2004); Gail Mulligan and KristinFlanagan,Age 2: Findings From the 2-Year-Old Follow-up of the Early Childhood Longitudinal Study,

Birth Cohort (ECLS-B),NCES 2006-043 (Washington DC: U.S. Department of Education, National Centerfor Education Statistics, 2006).12 Administration for Children and Families,American Recovery and Reinvestment Act: ImprovingChildren and Community Service: Administration for Children and Families: Head Start(Washington DC:U.S. Department of Health and Human Services, Administration for Children and Families). Available athttp://www.hhs.gov/recovery/reports/plans/headstart.pdf[August 13, 2010].13 Administration for Children and Families,Head Start Program Fact Sheet: Fiscal Year 2010 (Washington DC: U.S. Department of Health and Human Services, Administration for Children andFamilies, 2010). Available at http://www.acf.hhs.gov/programs/ohs/about/fy2010.html; U.S. Departmentsof Education and Health and Human Services, The Federal Budget: Fiscal Year 2011: Promoting Early

Learning for Success in School and in Life (author, July 2010). Available athttp://www.ed.gov/sites/default/files/2011-budget-promoting-early-learning.pdf.14 Gregory D. Kutz,Head Start: Undercover Testing Finds Fraud and Abuse at Selected Head StartCenters, GAO-10-733T (Washington, DC: U.S. Government Accountability Office, 2010). Available athttp://www.gao.gov/new.items/d10733t.pdf. The U.S. Department of Health and Human Services has takenseveral actions in response the GAO study; see U.S. Department of Health and Human Services, HHS

Secretary Kathleen Sebelius Outlines Plans and Steadfast Commitment to Upholding the Integrity of HeadStart Program, news release, May 17, 2010, available athttp://www.hhs.gov/news/press/2010pres/05/20100517d.html; and Kathleen Sebelius, Letter to Head StartGrantee, news release, May 17, 2010, available athttp://www.hhs.gov/news/press/2010pres/05/head_start_letter.html.15Hannah Matthews, Child Care Assistance in 2007: Spending Update. (Washington DC: Center for Lawand Social Policy, 2009); Matthews, Child Care and Development Block Grant Participation in 2008. Anestimated 2.2 million children were subsidized by Child Care Development Block Grants (CCDBG) andTemporary Assistance to Needy Families (TANF) in 2007, 55 percent of whom were under age 5.16 Matthews, Child Care and Development Block Grant Participation in 2008; National Child CareInformation and Technical Assistance Center, United States Child Care Statistics, an NCCIC ResourceGuide (author, 2010). Available at http://nccic.acf.hhs.gov/poptopics/statistics.pdf.17 Matthews, Child Care and Development Block Grant Participation in 2008.18

National Center for Education Statistics,Digest of Education Statistics: Table 46: PercentageDistribution of Children at About 2 and 4 Years of Age, by Type of Child Care Arrangement and SelectedChild and Family Characteristics: 2003-04 and 2005-06(Washington, DC: U.S. Department of Education,2006).19 Chris M. Herbst and Erdal Tekin, Child Care Subsidies and Child Development,Economics of

Education Review 29, no. 4 (2010): 618-638; Edward Zigler, Katherine Marsland, and Heather Lord, TheTragedy of Child Care in America (Ann Arbor, MI: Yale University, 2009); Deborah Lowe Vandell andBarbara Wolfe, Child Care Quality: Does It Matter and Does It Need to Be Improved? special report no.78, Institute for Research on Poverty (November 2000).


30/104


20 Internal Revenue Service, Statistics of Income2007 Individual Income Tax Returns: Table 1.3, AllReturns: Sources of Income, Adjustments, Deductions, Credits, and Tax Items, by Marital Status, Tax Year2007, Publication 1304, Rev. 07-2009 (Washington, DC: IRS, 2009).21 It is not possible for us to identify those children who have late birthdays and should be counted with theunder 5s but will not enter kindergarten until the following year. Thus, we exclude one-third of the

estimated credits for children ages 3 to 5 from our estimate. (A similar problem afflicts our estimates forchild care subsidies which exclude all children reported to be age 5); see Kent et al.,Data Appendix toFederal Expenditures on Pre-Kindergarteners and Kindergarteners in 2008 and Federal Expenditures on

Elementary-Age Children in 2008. We estimate that the total value of the tax credits was $3.94 billion in2008. Based on the percentage of eligible children under age 5, we estimate that about $2.2 billion went tochildren under age 5.22 Tax credits allow families to deduct a given amount directly from their tax payment. Thus, if families donot pay any federal income taxes, they cannot benefit from a tax creditunless the credit is refundable,which the child care credits are not; see Elaine Maag, Taxation and the Family: How Does the Tax SystemSubsidize Child Care Expenses? in The Tax Policy Briefing Bookby the staff and affiliates of the TaxPolicy Center (The Urban-Brookings Tax Policy Center, updated 2010).23 Barnett and others, The State of Preschool 2009: State Preschool Yearbook.24 Administration for Children and Families,Head Start Program Fact Sheet: Fiscal Year 2010 .25 Some Head Start programs provide full-day child care coverage using other state or federal funds.Statistic derived from Early Head Start enrollment data by the Office of Head Start in the Administrationfor Children and Families, U.S. Department of Health and Human Services.26 Estimates of ARRA spending by year for Head Start are from Kelly, Childrens Budget2010; also seeLisa Guernsey,A Closer Look at Obamas FY11 Budget: Head Start, Early Ed Watch Blog (WashingtonDC: New America Foundation, Early Education Initiative, 2010). Available athttp://earlyed.newamerica.net/blogposts/2010/a_closer_look_at_obama_s_fy11_budget_head_start-27490.27 U.S. Government Accounting Office,Head Start: Research Provides Little Information on Impact ofCurrent Program, GAO/HEHS-97-59 (Washington, DC: U.S. Government Accounting Office, 1997).Note: In 2004, the Government Accounting Office changed its name to the Government AccountabilityOffice.28 Michael Puma and others,Head Start Impact Study: First Year Findings (Washington, DC: U.S.Department of Health and Human Services, Administration for Children and Families, 2005).29 Michael Puma and others,Head Start Impact Study: Final Report(Washington, DC: U.S. Department of

Health and Human Services, Administration for Children and Families, 2010).30 Arthur J. Reynolds, Success in Early Intervention: The Chicago Child-Parent Centers (Lincoln, NE:University of Nebraska Press, 2000).31 W. Steven Barnett and others,Effects of Five State Pre-Kindergarten Programs on Early Learning(Rutgers University, National Institute for Early Education Research, 2007); William T. Gormley, Jr.,Deborah Phillips, and Ted Gayer, Preschool Programs Can Boost School Readiness, Science 27 (June2008): 1723-1724.32 Nicholas Zill, Alberto Sorongon, and Kwang Kim, Childrens Outcomes and Program Quality in HeadStart(Washington, DC: Administration for Children and Families, December 2006).33 Craig Ramey, Frances Campbell, and Clancy Blair, Enhancing the Life Course for High-Risk Children:Results from the Abecedarian Project, in Social Programs that Work, ed. Jonathan Crane (New York:Russell Sage Foundation, 1998); Lawrence J. Schweinhart and others,Lifetime Effects: The High/ScopePerry Preschool Study Through Age 40 (Ypsilanti, MI: High/Scope Press, 2005); Reynolds, Success in

Early Intervention .34 The research literature as a whole is remarkably consistent in finding that long-term outcomes are muchreduced from short-term outcomes, even for programs with the strongest long-term outcomes; see GregoryCamilli and others, Meta-Analysis of the Effects of Early Education Interventions on Cognitive and SocialDevelopment, Teachers College Record112, no. 3 (2010): 579-620. Available athttp://www.tcrecord.org/content.asp?contentid=15440.35 The study failed to find significant impacts on health using a conventional cutoff of p


31/104


36 Administration for Children and Families,Head Start Program Fact Sheet: Fiscal Year 2010.37 Administration for Children and Families Children and Families Services Programs: FY 2011 Budget(Washington DC: U.S. Department of Health and Human Services, Administration for Children andFamilies, 2010). Available at http://www.acf.hhs.gov/programs/olab/budget/2011/CFS.pdf.38 Statistic derived from Early Head Start enrollment data by the Office of Head Start (Washington DC:

U.S. Department of Health and Human Services, Administration for Children and Families, 2009).39 Christopher Jencks and Meredith Phillips, The Black-White Test Score Gap: An Introduction, in TheBlack-White Test Score Gap , ed. Christopher Jencks and Meredith Phillips (Brookings Institution Press,1998), pp. 1-54; Tamara Halle and others,Disparities in Early Learning and Development: Lessons fromthe Early Childhood Longitudinal StudyBirth Cohort (ECLS-B) (Washington, DC: Child Trends, 2009).40 Administration for Children and Families,Early Head Start Research and Evaluation Project (EHSRE),1996Current(Washington, DC: U.S. Department of Health and Human Services, Administration forChildren and Families). Available athttp://www.acf.hhs.gov/programs/opre/ehs/ehs_resrch/index.html#reports [August 13, 2010].41 Nurse-Family Partnership, Public Policy Update, policy update, March 22, 2010. Available athttp://www.nursefamilypartnership.org/assets/PDF/Policy/NFP_PP_Update_March_22_2010.42 David Olds and others, Long-term Effects of Home Visitation on Maternal Life Course and ChildAbuse and Neglect: A 15-Year Follow-up of a Randomized Trail,Journal of the American Medical

Association 278, no. 8 (1997): 637-643; Deanna S. Gomby, Home Visitation in 2005: Outcomes forChildren and Parents Invest in Kids Working Paper No. 7 (Washington, DC: Committee for EconomicDevelopment, 2005); Monica A. Sweet and Mark I. Appelbaum, Is Home Visiting an Effective Strategy?A Meta-Analytic Review of Home Visiting Programs for Families With Young Children, Child

Development75, no. 5 (2004): 1435-1456; Kimberly S. Howard and Jeanne Brooks-Gunn, The Role ofHome-Visiting Programs in Preventing Child Abuse and Neglect, Future of Children 19, no. 2 (2009):119-146.43 Barack Obama, Remarks of Senator Barack Obama to the Hampton University Annual Ministers'Conference (speech, Hampton, VA, June 5, 2007). Available athttp://www.barackobama.com/2007/06/05/remarks_of_senator_barack_obam_14.php.44 Ron Haskins, Christina Paxson, and Jeanne Brooks-Gunn, Social Science Rising: A Tale of EvidenceShaping Public Policy, Future of Children Policy Brief, Fall 2009.45 Kimberly S. Howard and Jeanne Brooks-Gunn, The Role of Home-Visiting Programs in PreventingChild Abuse and Neglect, Future of Children 19, no. 2 (2009): 119-146.46 The Department of Health and Human Services has released a fu

brooking institute report: investing in young children

Documents