Sean Taylor M.Ed. Reading Sage: Advocating for Every Child to Read, Reason, and Thrive: The Price of Shortchanging: Early Childhood Education (ECE)

The Price of Shortchanging: Early Childhood Education (ECE)

One-paragraph takeaway

The United States is not saving a coin by under-investing in the first years of life; it is borrowing against its own future at a very high interest rate. Our best estimate is that the gap costs the economy about $350 billion a year (roughly 1.2% of GDP), with a plausible range of $250 to $550 billion, while closing most of it would take about $100 to $200 billion a year in smart, quality-focused spending. The return is not automatic: it depends on paying and keeping excellent teachers, keeping groups small, and building programs that are coherent from birth through third grade. The choice is not between spending and saving. It is between paying early, on purpose, or paying later, in remediation, lost earnings, and lost potential.

The economic cost of the U.S. gaps in paid parental leave, affordable child care, universal preschool, and full-day, right-sized kindergarten

1. Executive summary

The question. What does the United States actually pay for having no national paid parental leave, an expensive and fragmented child care market, partial preschool coverage, and uneven full-day kindergarten, often in crowded classrooms?

The short answer. The cost is large, recurring, and mostly hidden because it shows up in other budgets: lower parental earnings, weaker employer productivity, remediation and special education, reduced adult earnings and tax revenue, higher health and justice costs, and slower growth. Our synthesis of the published evidence puts the steady-state annual cost of the early-childhood investment gap at roughly $250 billion to $550 billion per year, with a central estimate of about $350 billion (about 1.2% of GDP). Lifetime losses attached to each year's birth cohort are of a similar order of magnitude and are counted once, not twice (see Section 6).

The comparison that matters. Closing most of the gap would cost on the order of $100 billion to $200 billion per year in additional public and shared spending. Most rigorous benefit-cost studies of high-quality programs find returns well above one dollar per dollar spent, with long-run annual returns in the high single digits to low teens in the best-studied programs. The strongest conclusion is therefore not "every program pays for itself." It is that quality, targeting, and implementation determine whether the return is 0x or 7x, and that the current U.S. approach, which is cheap in the budget year and costly over a lifetime, is the worst of both worlds.

Five findings.

1. The U.S. is an outlier. It is the only OECD country without national paid parental leave, and it spends roughly half the OECD average share of GDP on early childhood education and care.

2. The largest measurable near-term cost is parental labor supply. One widely cited industry-research estimate puts the annual cost of the infant-toddler child care crisis alone at about $122 billion in lost earnings, productivity, and tax revenue.

3. The largest long-run cost is human capital. Gaps present at kindergarten entry persist, compound, and show up as lower reading proficiency, higher dropout and special education rates, and lower lifetime earnings.

4. The public pays twice. Taxpayers fund remediation, safety-net, and justice costs later because they under-fund prevention earlier.

5. The evidence is strong but not uniform. Fadeout of test-score gains, and at least one large randomized study with negative later outcomes, mean the case is for high-quality, well-implemented programs, not for any spending whatsoever.

2. Scope, definitions, and method

2.1 What is being priced

The counterfactual is a system comparable to peer-country norms and to the best evidence-based U.S. models. It has four levers:

Lever	Age span	Current U.S. gap
A. Paid parental and family leave	Birth to about 12 months	No national program; FMLA unpaid and covers only part of the workforce; paid leave in a minority of private jobs and a growing set of states
B. Affordable infant-toddler care	0 to 3	Market-based, expensive, thin supply, low wages for staff
C. Universal high-quality preschool	3 to 4	Roughly one-third of 4-year-olds and a small fraction of 3-year-olds in state-funded pre-K; Head Start reaches a minority of eligible children
D. Full-day, right-sized kindergarten	5 to 6	Most public kindergartners are in full-day programs, but state requirements vary and class sizes and teacher supports differ widely

2.2 How we keep the analysis MECE

We organize costs along two axes so that no dollar is counted twice and nothing is left out.

Axis 1: Mechanism (who or what is harmed first).

• M1. Parents' labor market outcomes (earnings, career, retention)

• M2. Children's human capital (cognitive, language, executive function, social-emotional)

• M3. Fiscal and social-system costs triggered by weaker outcomes (remediation, safety net, justice, health)

• M4. Child health and welfare effects (birth outcomes, maltreatment, early detection of delays)

• M5. Macroeconomic and intergenerational effects (growth, inequality, mobility)

Axis 2: Bearer (who pays).

• Households, employers, governments (federal, state, local), and the broader economy.

Rules to prevent double counting. Each effect is assigned to the single mechanism where it first appears. Macro effects (M5) are reported as a memo item, not added to the total, because they are the aggregate of M1 through M4. Fiscal costs are reported as a transfer view (taxpayer cost) and are not added to lost earnings except for the tax-revenue component, which is a government loss rather than a household loss.

2.3 Two time frames

• Annual flow: what the economy loses in a typical year at current coverage.

• Cohort lifetime: the present value of losses over the life of one birth cohort of about 3.6 million children.

The annual flow is the most useful comparison to budget decisions; the cohort view is the one that matches how education investments actually pay back.

2.4 Method and honesty about uncertainty

This is a synthesis model that combines published estimates and transparent assumptions, not a new econometric study. Ranges reflect different discount rates, program quality, take-up, and attribution rules. Every number should be treated as an order-of-magnitude planning estimate, and the verification list in Section 11 names what to confirm before publishing.

3. The current state of play

3.1 International position

• The United States is the only OECD member with no national statutory paid maternity, paternity, or parental leave. Most peers offer many weeks to many months of paid leave, often with job protection and incentives for both parents.

• Public spending on early childhood education and care in the U.S. is roughly half the OECD average as a share of GDP, and far below the Nordic countries and France.

• Enrollment of 3-year-olds in organized early education is well below the OECD average and far below the near-universal coverage in countries such as France, Spain, Sweden, and Norway.

• U.S. maternal labor-force participation, which once led the world, fell behind several peers beginning in the 1990s. Research by Blau and Kahn attributes a meaningful share of that relative decline to the absence of family-friendly policies such as paid leave and part-time protections.

3.2 Household affordability

• Average annual center-based infant care costs in many states exceed in-state public college tuition. National averages in recent years have been in the range of roughly $11,000 to $13,000 per child per year, and far higher in high-cost states.

• Families with low and middle incomes, who benefit most from care, are least able to pay for it; subsidy eligibility is limited, and only a minority of eligible children receive federal child care subsidies.

3.3 Supply and workforce

• Child care and early education staff are among the lowest-paid occupations in the country, with median wages near those of fast-food and retail workers despite credential expectations.

• Turnover is high, programs close or cut slots when they cannot hire, and "child care deserts" cover large parts of rural America and many urban neighborhoods.

• This is the quality constraint: low wages and high turnover directly undermine the teacher-child relationships that drive developmental benefits.

3.4 Kindergarten

• Kindergarten attendance is not compulsory in roughly half of states, and requirements for districts to offer full-day programs vary. Most children now attend full-day kindergarten, but the dosage, class size, and curriculum quality differ widely.

• Crowding matters. Evidence from the Tennessee STAR experiment and follow-up work by Chetty and colleagues shows that early class size and classroom quality in the earliest grades predict later college attendance and adult earnings.

3.5 Early intervention

• Only a small share of infants and toddlers are served under the federal early intervention program (IDEA Part C), well below estimates of the share with developmental delays. Missed early detection is a direct feeder into later special education costs.

4. Why the early years carry outsized economic weight

4.1 Developmental science

The brain's architecture, including language, attention, and emotional-regulation circuits, develops fastest between birth and age five. Early experience, including responsive caregiving, language exposure, stable routines, and stress levels, shapes that architecture. Chronic adversity and under-stimulation raise the later cost of building the same skills.

4.2 The economics

James Heckman's work frames this as a return-on-investment curve that is steepest for the youngest children: skills beget skills, so a dollar spent early compounds, while a dollar spent on remediation later works against a larger accumulated deficit. Heckman and colleagues estimate annual rates of return of roughly 7 to 10 percent for the Perry Preschool program and as high as the low teens for comprehensive birth-to-five programs such as the Abecedarian Project. Even conservative analysts, such as those at the Washington State Institute for Public Policy, find positive benefit-cost ratios for many well-run pre-K programs.

4.3 The reading pipeline

Outcomes at the end of third grade are a hinge. A widely cited longitudinal analysis by Donald Hernandez found that children who were not reading proficiently by third grade were about four times more likely to leave high school without a diploma than proficient readers, with the risk higher still for children who also lived in poverty. National reading results remain stubbornly low: only about one in three fourth-graders reads at or above the NAEP proficient level, and the share has not recovered to its earlier peak.

5. The evidence base, including the counter-evidence

A credible report must show where the evidence is strongest and where it is contested.

Strong long-run positive evidence

• Perry Preschool (Michigan) and Abecedarian (North Carolina): small randomized trials with decades of follow-up showing higher earnings, better health, and lower crime. Abecedarian follow-ups also found improved adult cardiovascular risk markers.

• Chicago Child-Parent Centers: a large quasi-experimental study showing higher educational attainment, lower special education placement, lower maltreatment, and benefit-cost ratios commonly reported in the range of high single digits.

• Boston and Tulsa public pre-K: lottery-based and quasi-experimental studies showing gains in school outcomes, with Boston evidence of higher high-school graduation and college enrollment.

• Head Start long-run studies by Deming and by Bailey and colleagues: positive adult outcomes, including educational attainment and economic self-sufficiency, even where test-score gains faded.

• Paid leave: studies of California's program, and of programs abroad, find more leave-taking, higher post-birth attachment to employers, and in several studies better infant health and breastfeeding outcomes. Findings on long-run maternal earnings are mixed and depend on leave length. Very long leaves can reduce women's career progression, which is why design matters.

Contested or negative evidence

• Test-score fadeout is common: early advantages in test scores often shrink by third grade. The long-run gains seen in some studies suggest that non-cognitive skills, behavior, and attainment persist even when scores do not, but this remains an active debate.

• Tennessee Voluntary Pre-K randomized study found that children randomized into the program did worse on several academic and behavioral measures by sixth grade. It is a warning that low-quality or poorly aligned programs can fail or even backfire.

• Head Start Impact Study: modest early gains that were mostly gone by the end of kindergarten and third grade on measured cognitive outcomes.

• "30 million word gap" (Hart and Risley) is widely cited but its methods and replication are disputed; we do not rely on it for any figure here.

Reading of the evidence. The most defensible position is conditional: programs with trained, well-paid, stable teachers, rich language and play-based instruction, strong curricula, small groups, and family engagement show durable returns; programs that are large, underfunded, and academically narrow show weak or negative returns. The cost of the status quo is therefore real, but the benefit of any fix depends on funding it at a level that preserves quality.

6. The cost stack

All figures are annualized, rounded, in approximately 2026 dollars, and are synthesis estimates with explicit logic.

M1. Parents' labor market losses (households, employers, government)

What happens. Without paid leave and with expensive, scarce child care, parents, disproportionately mothers, reduce hours, leave jobs, decline promotions, or exit the labor force. Employers pay in turnover, absenteeism, and lost productivity. Governments lose income and payroll taxes.

Evidence anchors.

• A widely cited industry-research estimate puts the infant-toddler child care problem alone at about $122 billion per year in lost earnings, productivity, and tax revenue (ReadyNation / Council for a Strong America, 2023 update; a 2021 version estimated about $57 billion on narrower assumptions).

• Child-penalty research (Kleven and colleagues) finds long-run earnings gaps for mothers relative to fathers that are large in the U.S., on the order of roughly 30 percent.

• Only a minority of private-sector workers have access to employer-paid family leave, and the lowest-wage workers have the least access.

Our estimate. Extending the $122 billion infant-toddler estimate to preschool-age children and adding the leave gap yields $150 billion to $250 billion per year, of which roughly 20 to 30 percent falls on governments as lost tax revenue, and the rest falls on households and employers.

M2. Children's human-capital losses (households and the economy)

What happens. Children who lack quality early care enter kindergarten with gaps in language, executive function, and social-emotional skills. These gaps compound through the reading pipeline, producing lower attainment and lower lifetime earnings.

Model logic.

• Cohort size: about 3.6 million births per year.

• Share of children meaningfully under-served relative to a high-quality benchmark: we assume roughly 40 to 50 percent (all children from low-income families plus a large share of middle-income families with no access to quality).

• Present-value earnings and productivity loss per under-served child attributable to the early gap: about $40,000 to $120,000, consistent with the lower half of published program benefit estimates after applying a probability that a given program is fully effective.

• Result: roughly $60 billion to $215 billion per cohort-year, which we round to $70 billion to $200 billion per year.

Cross-check. McKinsey and Company's 2009 analysis of the U.S. achievement gap estimated that the shortfall relative to better-performing countries cost the economy roughly $1.3 to $2.3 trillion in 2008 (9 to 16 percent of GDP), with income-based gaps alone costing $400 to $670 billion. Early childhood is one portion of that K-12 gap, so a slice of this size is consistent with the larger figure. Hanushek and Woessmann's growth models imply cumulative gains in the tens of trillions of present-value dollars if achievement were raised to the level of top-performing countries.

M3. Fiscal and social-system costs (taxpayers)

What happens. Weaker early outcomes lead to more grade retention, special education placement, and remedial services; higher rates of high-school non-completion; and more reliance on safety-net programs and the justice system.

Evidence anchors.

• Studies of Perry, Chicago CPC, and several state pre-K programs show reductions in special education placement, grade retention, and criminal justice involvement.

• Belfield, Levin, and colleagues estimate that "opportunity youth" (young people neither in school nor working) impose taxpayer and social costs on the order of tens of billions of dollars per year for each cohort-age group, with lifetime costs well above $1 trillion for the existing population.

• Earlier work by Levin and colleagues puts the lifetime public cost of each high-school dropout in the hundreds of thousands of dollars when lost taxes, health spending, and justice costs are combined.

Our estimate. After removing the lost-tax portion already included in M1 and M2, the incremental fiscal cost attributable to the early-years gap is about $20 billion to $60 billion per year.

M4. Child health and welfare (households, Medicaid and public health systems)

What happens. Short or no paid leave is associated with lower breastfeeding rates, delayed well-baby care, higher maternal stress and postpartum depression, and in some studies worse infant health. Poor-quality or unsafe care raises injury and neglect risks; missed early detection delays intervention for developmental delays.

Evidence anchors. CDC-based estimates place the total annual economic burden of child maltreatment in the hundreds of billions of dollars. Early childhood programs such as Chicago CPC reduced maltreatment, but only a fraction of that total is plausibly preventable through early education and care.

Our estimate. $10 billion to $40 billion per year attributable to the gap.

M5. Macroeconomic and intergenerational effects (memo item, not added)

Lower workforce participation, lower skills, and lower innovation capacity reduce potential GDP, widen inequality, and reduce mobility; weaker outcomes also carry into the next generation. These effects are the aggregate expression of M1 through M4, so we report them as context and do not add them.

Summary table

Mechanism	Primary bearer	Low	Central	High
M1. Parents' labor market	Households, employers, governments	$150B	$190B	$250B
M2. Children's human capital	Households, economy	$70B	$120B	$200B
M3. Fiscal and system costs	Taxpayers	$20B	$35B	$60B
M4. Child health and welfare	Households, public systems	$10B	$20B	$40B
Total (annual, rounded)		about $250B	about $365B	about $550B
Share of U.S. GDP		about 0.8%	about 1.2%	about 1.8%

The low case sums to less than the simple column total because we assume partial overlap in the low scenario; the central and high cases assume overlap is already removed by the allocation rules in Section 2.2.

Cohort-lifetime view

Using a 3 percent real discount rate, the present value of lifetime losses for one birth cohort is of the order of $300 billion to $800 billion, mostly from M2 and M3. This view is the same cost expressed over a lifetime, not an additional cost, so it should not be added to the annual flow.

7. Who bears the cost

Bearer	Share of central estimate	Main channels
Households	about 45%	Lost earnings, out-of-pocket care costs, lower child lifetime income
Employers	about 15%	Turnover, absenteeism, lost productivity
Governments	about 25%	Lost tax revenue, remediation, safety net, justice, health
Wider economy	about 15%	Lower growth, weaker consumer demand, lower innovation

The key political-economy insight is that the people who would pay for the fix (taxpayers, in the budget year) are not the people who capture most of the benefit (children, parents, and future taxpayers). That mismatch explains much of why the U.S. under-invests.

8. The other side of the ledger: what closing the gap costs

Lever	Rough additional annual cost	Notes
A. National paid family leave (about 12 weeks, wage replacement of about two-thirds, capped)	$20B to $50B	Typically financed through small payroll contributions, often well under 1% of wages
B. Affordable, high-quality infant-toddler care (sliding-scale subsidy and workforce wage floor)	$40B to $90B	The largest single cost; driven by workforce pay and ratios
C. Universal high-quality pre-K for 3- and 4-year-olds	$30B to $60B	Net of existing state and Head Start spending
D. Kindergarten quality (full-day access, class size, aide support)	$5B to $20B	The smallest lever; many costs already in school budgets
Total	about $100B to $200B	About 0.3% to 0.7% of GDP

Planning numbers vary widely by quality standard, wage assumptions, and take-up; these are consistent in scale with published proposals and cost analyses but are not a substitute for a detailed costing.

9. Net return, scenarios, and sensitivity

9.1 Scenarios

Scenario	Program quality	Share of gap closed	Annual cost	Annual benefit	Benefit-cost ratio
Weak implementation	Low wages, big groups, narrow academics	40%	$100B	$50B to $100B	0.5 to 1.0
Base case	Solid quality, partial wage floor	65%	$150B	$230B to $330B	about 1.5 to 2.2
High-quality system	Strong wages, small groups, play-rich curricula	85%	$200B	$310B to $470B	about 1.5 to 2.4 in the first decade, rising as cohorts mature

Note that benefits ramp over time. Leave and parental labor-supply effects arrive within a year or two; child-outcome benefits arrive over 10 to 40 years, so the benefit-cost ratio in the first decade understates the lifetime ratio. Published lifetime benefit-cost ratios for the best-studied programs range from roughly 3:1 to 10:1 or more.

9.2 What moves the answer most

1. Quality and workforce pay. The biggest driver; low-quality scale-up can erase most of the return.

2. Targeting vs universality. Benefits per dollar are highest for low-income children; universal programs gain political durability, broader labor-market effects, and peer benefits, but at lower return per dollar.

3. Discount rate. At 7 percent, lifetime benefits shrink substantially relative to 3 percent.

4. Fadeout and attribution. If early gains fade more than assumed, M2 falls by a third or more.

5. Parental labor supply response. The M1 estimate is sensitive to the share of caregivers who would move into work and to wage assumptions.

10. Strategic implications

For policymakers

• Treat early childhood as infrastructure with a measurable return, not as welfare spending. Use multi-year, life-cycle budgeting so that early spending is scored against later savings.

• Sequence and package. Paid leave (low cost, fast labor-market return) plus affordable infant-toddler care (the largest cost, with the largest parent-labor effect) plus universal pre-K and kindergarten quality create the full pipeline; funding only one segment leaves leaks.

• Fund the workforce. Wage floors and credential pathways for early educators are the quality lever that keeps returns positive.

• Measure what matters. Track kindergarten readiness, third-grade reading, attainment, and adult outcomes, not only test scores in year one.

• Use pay-for-success and state-federal matching to reduce the mismatch between who pays and who benefits.

For employers

• Employer-supported leave and child care show positive returns in retention and productivity, and are a competitive advantage in tight labor markets.

For educators and advocates

• Anchor the case in outcomes the public already values: literacy by third grade, high-school completion, workforce readiness, and fiscal prudence.

• Pair the numbers with the counter-evidence. Acknowledging fadeout and the Tennessee findings builds credibility and sharpens the argument for quality.

11. Limitations and verification list

This white paper synthesizes published research from memory-informed knowledge and was not built from a live literature pull. Before publishing, confirm each item below against its primary source and update to the latest edition.

1. ReadyNation / Council for a Strong America: the $122 billion annual infant-toddler child care cost (2023) and the 2021 comparison figure.

2. McKinsey and Company (2009), The Economic Impact of the Achievement Gap in America's Schools: $1.3 to $2.3 trillion and the income-gap range.

3. Heckman and colleagues: return-on-investment estimates for Perry Preschool and Abecedarian; update to the most recent versions.

4. OECD Education at a Glance and Family Database: U.S. ECEC spending as a share of GDP, enrollment of 3-year-olds, and paid leave comparisons.

5. NIEER State of Preschool yearbook: current enrollment shares for 3- and 4-year-olds and state funding.

6. Child Care Aware of America: current national average child care price.

7. Bureau of Labor Statistics National Compensation Survey: share of private-sector workers with paid family leave.

8. NAEP: current fourth-grade reading proficiency.

9. Hernandez (2011), Annie E. Casey Foundation: third-grade reading and dropout risk.

10. Chetty and colleagues (2011), How Does Your Kindergarten Classroom Affect Your Earnings?, and Tennessee STAR follow-up.

11. Durkin and colleagues, Tennessee Voluntary Pre-K randomized study; Deming (2009); Bailey and colleagues (2021).

12. Belfield and Levin on opportunity youth and dropout costs; CDC maltreatment cost estimates; Kleven and colleagues on child penalties; Blau and Kahn on female labor force participation.

13. Cost figures in Section 8 should be checked against recent Congressional Budget Office, Center for American Progress, and state-level costings.

The modeling assumptions in Section 6 (share of children under-served, present-value loss per child, overlap adjustments) are our own and are the first things to test with sensitivity runs if this paper is used in a formal setting.