Insurance by BogdanovaThe Encyclopedia of Insurance and Risk Science

RESEARCH DOSSIER

Child and Family Transfer Take-Up and Redistribution

Abstract. Child and family transfers transform statutory family support into disposable resources through an institutional sequence that begins with eligibility and ends with the distribution, timing, and effective use of payments. This monograph examines that sequence across contributory family benefits, tax-financed allowances, refundable tax credits, conditional and unconditional social assistance, and the administration of those instruments. It establishes an explicitly comparative vocabulary for legal coverage, eligible populations, claimed entitlements, cash receipt, benefit adequacy, horizontal and vertical redistribution, and empirically attributable child outcomes. The research combines international labor standards, national administrative and fiscal sources, harmonized social protection statistics, and peer-reviewed causal evaluations. The global child-benefit coverage estimates in the International Labour Organization's 2024–26 report, the World Bank's ASPIRE incidence definitions, the United Kingdom's 2025 Child Benefit administrative statistics, Brazil's Bolsa Família payment data, and evidence on child tax benefits and long-run human development anchor the analysis. The central finding is that distributive performance is the joint result of benefit law, administrative access, household composition, actual payment, tax interaction, and measurement design. A benefit can be progressive among its recipients while leaving a substantial eligible population unpaid; a widely claimed benefit may coexist with fewer households receiving positive payments; and a transfer that crosses an annual poverty threshold may still be too late or too irregular to prevent immediate deprivation. The monograph develops source-grounded analytical interpretations, clarifies causal and descriptive claims, and identifies empirical tests for future comparative research without proposing a new theory or doctrine.

Keywords: Child benefits; family transfers; social protection; benefit take-up; non-take-up; horizontal redistribution; vertical redistribution; benefit incidence; child poverty; tax credits; administrative burden; family composition; cash transfers; microsimulation; child outcomes; effective coverage; social insurance.

Introduction

The distributive purpose of child and family transfers is realized at the point where a legally defined benefit becomes a usable resource in a household containing a child. The interval between these events contains legal tests, administrative communication, application, verification, payment, possible tax recovery, and household allocation. Each stage changes who receives support, when resources become available, and how children experience the transfer. Consequently, transfer expenditure, formal eligibility, and observed coverage are separate quantities requiring separate evidence.

The ILO World Social Protection Report 2024–26 estimated that 23.9 percent of children aged 0–18 globally received a child or family benefit, leaving approximately 1.8 billion children without such coverage. For ages 0–15, the report described 28.2 percent coverage in 2023, compared with 22.1 percent in 2015. These are different age universes and must remain different statistical series. Their importance is institutional: national legal provisions cannot, by themselves, establish effective delivery. A parallel World Bank assessment of social protection reports that substantial numbers of people in low- and middle-income economies lack adequate protection. Its total-population estimates are not child-benefit counts and are used here solely to situate the wider protection system.

This research is located within Social Insurance and Public Protection. Its subject extends across social insurance, tax-financed family allowances, tax credits, and means-tested cash transfers because families can receive coordinated support from several legal regimes. Those categories remain institutionally distinguishable. A social insurance entitlement may be linked to contributions or employment; a universal family allowance generally relies on categorical child status; means-tested social assistance relates benefits to income, assets, or family composition; refundable tax credits are mediated by the fiscal system; and in-kind services are resources with valuation and access rules of their own.

The work asks three questions. First, which legal and administrative stages intervene between family-benefit eligibility and actual receipt? Second, how do household structure, targeting rules, non-take-up, taxes, and benefit timing shape vertical and horizontal redistribution? Third, what evidence warrants conclusions about poverty, schooling, health, and long-term child well-being? Each answer is restricted to the relevant population, jurisdiction, period, and identification strategy.

The method is comparative statutory interpretation, descriptive administrative-statistical reconstruction, definitional incidence analysis, and critical synthesis of empirical and quasi-experimental research. No household-level database has been independently reanalyzed for this manuscript. Numerical values below are attributed to identified source publications and are not presented as newly estimated results. Conceptual formulae express accounting relationships and transparent evaluation estimands; they do not introduce an authorial theory.

Conceptual and Historical Foundations

Legal objects and institutional evolution

The historical rationale for family benefits connects maintenance of children with the collective allocation of risks and costs across working lives, household forms, and generations. Family responsibility entered early social-security arrangements through employment-linked and occupation-linked provisions; subsequent tax-financed and residence-based schemes broadened eligibility beyond protected workers. The international standards reflect these successive institutional possibilities without prescribing one uniform national program.

Part VII of the ILO's Social Security (Minimum Standards) Convention, 1952 addresses family benefit. Article 40 identifies the contingency as responsibility for the maintenance of children, Article 41 sets alternative protected-person benchmarks, Article 42 permits periodic payments or specified benefits in kind, and Article 43 allows defined qualifying periods. Article 44 specifies aggregate value benchmarks in relation to a reference wage and numbers of children; it is not a universal individual benefit rate. The Convention's legal force for any country depends on ratification and the accepted parts of the instrument. Textual international minimums should never be mistaken for identical benefit schedules in every jurisdiction.

The later Social Protection Floors Recommendation, 2012 establishes in paragraph 5(b) the normative objective of basic income security for children with effective access to nutrition, education, care, and other necessary goods and services. It allows plural institutional instruments, including universal schemes, social assistance, social insurance, and negative-income-tax arrangements. Recommendation No. 202 is a recommendation, not a directly harmonizing convention; national eligibility and enforcement are established in national law.

In modern systems, three broad routes commonly coexist. A payment can be made to a parent or caregiver because a qualifying child is present; it can be provided as additional support for a household whose assessed resources fall below a threshold; or it can enter as a credit or deduction through the tax system. Each route may use age tests, disability provisions, residence requirements, registration, income tests, or time-limited certification. Their overlap determines the actual family-support budget available to a child.

What constitutes a child and family transfer?

A child or family transfer is a publicly authorized cash, near-cash, or identifiable resource payment designed to support persons bearing the costs of children or the household in which children live. The precise analytic boundary must be specified. A direct child allowance, a parental cash benefit, a conditional family transfer, and a refundable tax credit may be aggregated for an income-distribution exercise only after distinguishing gross payment, income-related withdrawal, and tax effects. School services, public healthcare, and subsidized childcare should be assessed as in-kind services and should enter a total-resource measure only with stated valuation assumptions.

The OECD family-benefits spending indicator classifies government support to families into cash, services, and tax-system support. Its percentage-of-GDP numerator differs from the narrower sum of cash transfers received by households. Comparisons that equate a country's total family-policy budget with its child-cash-benefit intensity therefore combine different instruments and may overstate the cash available to families.

Child-related transfers are connected to both vertical and horizontal equity. Vertical redistribution changes the disposable resources of families at different income levels, typically aiming to concentrate a greater relative benefit among households with fewer resources. Horizontal redistribution changes resources across otherwise similar households with different numbers, ages, or needs of children. A fixed payment per child can be horizontally redistributive even when the cash distribution extends across all incomes. A means-tested payment can be strongly vertical in statutory design while leaving horizontal inequities where families with identical needs differ in access or administrative capability.

The observables that must be kept separate

For program p, period t, and defined population G, let E denote eligible units under the legislation in force; C the eligible units with a registered claim; P the eligible units for which a positive payment was actually made; and N the total potentially relevant population. Basic descriptive rates are:

  • Statutory eligibility prevalence: E / N.
  • Claim take-up among eligible units: C / E.
  • Positive-payment receipt among eligible units: P / E.
  • Positive-payment conversion among eligible claimants: P / C.
  • Population coverage by positive receipt: P / N.

These expressions presuppose consistent units and time. A unit can be a child, applicant, family, household, or tax unit. The equality of any two rates is contingent on each program's institutional rules. Claims may remain active where payment is voluntarily declined or subsequently recovered through taxation. Payments can also be received by units later deemed ineligible; a reconciliation study therefore needs a separate false-positive classification rather than folding every transfer into P.

The ILO's statistical annex on effective coverage describes recipient coverage for defined child-age universes; the World Bank's ASPIRE data methodology includes indirect beneficiaries living with a direct recipient. Both are legitimate conventions, but a percentage of children living in recipient households is not the same quantity as the percentage of registered parents receiving payments. Neither is automatically an estimate of the proportion of all legally eligible households that claimed.

Family composition as a substantive variable

Household size affects both benefit entitlement and the needs-adjusted assessment of resources. Relevant differences include number of children, ages, disability, single- versus two-adult caregiving, shared custody, nonresident parents, multigenerational households, children living with relatives, and households temporarily displaced across jurisdictions. A family program may pay the same nominal amount for each child, a higher amount for younger children, or different amounts by birth order; equivalent disposable income will respond differently to each arrangement.

The legal claimant, cash recipient, household income unit, and child beneficiary may be four different actors. A child can be covered through a caregiver's account while the benefit is assigned to an individual tax filer; a child can share food and housing with other adults not legally represented in the claim. For that reason, an outcome analysis must name both the immediate payment unit and the persons whose welfare is being measured.

Why the historical distinction still matters

Institutional histories explain why apparently comparable benefit schemes can exhibit different take-up. Employment-connected schemes may be effective for stable formal workers yet exclude irregular or informal employment. Residence-based child allowances may be widely accessible through civil registration, while income-tested support requires repeated income verification. A tax credit can be highly valuable to families filing returns but inaccessible in practice to some otherwise eligible nonfilers. The historical trajectory shapes the administrative records available to evaluators and the populations likely to be visible.

The comparative inference is that institutional breadth and benefit generosity must be evaluated alongside access pathways. A generous statutory entitlement that never reaches a child contributes no realized cash resources to that child in the period under study.

Benefit Uptake

Administrative stages of an entitlement

A family-transfer claim moves through several potentially independent stages: identifying eligibility, learning the program exists, recognizing eligibility, assembling identity or income records, submitting a claim, responding to requests for clarification, receiving an eligibility decision, establishing a payment mechanism, keeping the award current, and reconciling or appealing errors. Dropout at any stage changes realized coverage. Design decisions about proactive notification, automatic enrollment, integrated registries, renewal frequency, and appeal procedures can therefore affect distribution without any statutory increase in the benefit amount.

The foundational review by Currie (2006) on take-up describes the importance of administrative barriers, default enrollment, and the cumulative effect of transaction costs. This review draws substantially on U.S. and U.K. experience, not a global causal estimate. Its lasting implication is that the design of application procedures is part of the economic design of a transfer. An administrative rule can change who receives the benefit at given formal eligibility.

A randomized U.S. study of older adults by Finkelstein and Notowidigdo (2019) found SNAP enrollment over nine months of 6 percent in the control group, 11 percent after eligibility information, and 18 percent after information plus application assistance. This was a nutritional assistance intervention among likely eligible elderly individuals, not a direct experiment in child benefits. It supplies rigorous mechanistic evidence that application assistance may alter participation beyond information alone. It does not establish the magnitude that a family-transfer agency would obtain by copying the intervention. The authors' welfare analysis also warns that newly induced participants may differ from existing claimants, so higher enrollment does not automatically imply optimal targeting.

Measuring the denominator of non-take-up

An administrative register is usually strong at counting successful awards but weak at counting eligible people who never applied. The denominator E must therefore be estimated from a census, tax register, representative household survey, linked administrative records, or a microsimulation of current law. Each option introduces a different error structure. Survey-based eligibility may be misclassified by missing assets, income components, custody rules, or residence histories. Administrative datasets may miss informal employment or undocumented family arrangements. Tax records may omit nonfilers. The more complex the statute, the less credible a simple headcount of apparent eligibles.

Operational studies should publish E's definition; the sampling universe; reference year; whether the threshold is measured before or after taxes; the ages and custody status of covered children; and how appeals, late claims, suspensions, and retrospective payments were treated. A measured payment in December does not imply uninterrupted support throughout January–December. A yearly claim count can greatly exceed or differ from a monthly active-recipient count, depending on entries and exits.

An immediate accounting identity helps keep administrative leakage visible. If the statutorily eligible population E is partitioned into eligible recipients P and eligible nonrecipients U, then E = P + U only when one mutually exclusive unit, one benefit period, and one payment definition are used. With underpayments, overpayments, partial-year receipt, and tax recoveries, the proper measure is a vector of actual amounts rather than a binary classification alone.

The United Kingdom as a measurement case

HM Revenue and Customs' Child Benefit statistics for August 2025, published in April 2026, reported approximately 7.55 million families claiming Child Benefit but 6.87 million families receiving payments. It recorded around 685,000 claimant families opting out of payment and estimated that the claim take-up rate among eligible children had declined to 87 percent, compared with 97 percent in 2012. The report defines take-up to include claimed children in families that opt out of payment. Thus the take-up rate cannot be used as the positive-payment coverage rate.

The High Income Child Benefit Charge introduces another distinction. Some families claim while declining payment because a tax charge may claw back the economic value; others refrain from claiming at all. Recorded claiming, gross payments, and net-of-tax resources must be distinguished. The HMRC research on Child Benefit take-up, based on a survey of nonclaimants and delayed claimants and qualitative interviews, reports that widespread awareness does not imply clear understanding of the benefit. The study illuminates communication and behavioral frictions but cannot, on its own, assign causal responsibility for the full national decline.

This case yields a concrete comparability rule: the analyst must report whether a published administrative 'take-up' numerator includes registered recipients with zero current cash receipts. Treating every registered claimant as a cash recipient can bias measured redistribution upward and misclassify children for an outcome study.

Tax-filing and payment-system access

The U.S. 2021 temporary advance Child Tax Credit provides a tax-system example. The Internal Revenue Service payment statistics identify payments distributed from July through December 2021 and describe the taxpayer-recipient basis of the tabulations. Advance payments depended on information available to the tax administration, and final entitlements were reconciled with tax-year filings. A direct deposit or mailed check could reach a family without an additional recurring application, but households outside the existing tax and nonfiler information systems could require active enrollment.

The analytic distinction is between the number of tax accounts receiving advances, the number of qualifying children represented by those accounts, the number of eligible tax units at year-end, and the ultimate net credit after reconciliation. These are not interchangeable. A transfer analyst should also separate the effect of advance-payment timing from the effect of increased annual credit generosity. A monthly payment may relieve short-run cash constraints even if annual entitlement is unchanged; a larger annual entitlement alters annual disposable income even if paid in a lump sum.

Administrative inclusion and exclusion

Income testing can direct finite budgets toward lower-income households but raises the informational demands of screening. Administrative exclusion includes eligible households that fail to apply, are incorrectly rejected, cannot renew, receive no payment despite an award, or incur delays that destroy the benefit's short-term protective function. Incorrect inclusion includes payments to ineligible units, failures to update income, duplicated awards, and errors created by incompatible agency records. Both categories require investigation; reducing one error can increase the other if controls become more burdensome.

The appropriate design question concerns the joint distribution of error, need, resources, and institutional costs. A system that reduces incorrect payment through an expensive verification process may increase eligible nonreceipt. Conversely, an automatic benefit with simple rules may reduce omissions while accepting additional gross payments that are later recovered or offset through taxation. Evaluation should include claimant costs, administrative expenditure, fiscal leakage, and the value of consumption smoothing.

International coverage and country-level interpretation

The global ILO estimates demonstrate the scale of effective-coverage gaps, particularly where financing and administrative infrastructure are constrained. The ILO's coverage findings report only 7.6 percent of children aged 0–18 in low-income countries receiving child or family cash benefits. This is not the statutory non-take-up rate: some children may be outside legislation altogether, while others may be eligible but unpaid. Distinguishing statutory gaps from implementation gaps is essential to prescribing appropriate remedies.

The World Bank's 2025 protection assessment describes a much broader protection universe and warns of inadequate coverage in low-income economies. Combining those numbers with child-benefit rates would produce an invalid composite denominator. A country can have high coverage of older adults and low coverage of children, or high general assistance coverage but limited child-specific income support.

Receipt, timeliness, and continuity

A binary annual indicator conceals the regularity of payments. Define the receipt profile for household h as monthly amounts B(h,1) through B(h,12). Two families receiving the same annual total can have distinct hardship exposures if one receives predictable monthly transfers and the other receives a delayed lump sum. For households unable to borrow cheaply, regular support may be especially valuable around school fees, food expenditures, and illness episodes. Evaluation should record disbursement date, failed transactions, interruptions, recoveries, and adjustments rather than assuming a fully smooth resource stream.

Any causal claim that timing improved outcomes requires a credible research design isolating timing from benefit size, beneficiary selection, and contemporaneous events. The assertion that liquidity matters is an economic interpretation of household constraints; its empirical magnitude remains specific to each studied population.

Household Incidence

Gross allocation, net allocation, and final resources

Distributional incidence begins with the amount legally assigned but is completed only after receipt, withholding, withdrawal, and fiscal interaction. For household h in period t, define the cash-transfer amount actually disbursed B(h,t), the tax liability or charge directly attributable to the transfer T(h,t), and the reduction in other means-tested support attributable to that transfer W(h,t). A transparent partial net measure is:

Net family-transfer contribution(h,t) = B(h,t) − T(h,t) − W(h,t).

This identity is a partial, program-specific accounting definition. It does not capture indirect taxes, labor responses, noncash services, public financing incidence, or general-equilibrium wage and price changes. It must not be used to claim that every tax system legally earmarks a particular charge to a benefit. Where a policy exercise studies the complete tax-benefit system, differences between simulated disposable resources under well-defined policy regimes are more defensible than selectively subtracting isolated tax items.

In an equivalent-income calculation, household net resources are divided by an explicitly defined equivalence scale. Poverty thresholds and benefit effects change if the analyst moves between per capita income, modified OECD scale, a national official measure, or a multidimensional deprivation index. Family composition enters both the transfer schedule and the needs denominator, so one transfer can have different equivalized effects for small and large households.

Benefit incidence and beneficiary incidence

The World Bank's ASPIRE definitions of incidence separate the share of all transfer money received by a group from the share of all beneficiary persons in that group. Benefit incidence for income quintile q is the sum of transfers allocated to that quintile divided by the sum to all quintiles. Beneficiary incidence counts persons in recipient households, including indirect beneficiaries, within the appropriate total. Coverage is a different proportion: beneficiaries in q divided by the population in q. These distinctions resolve a common interpretive error: a program can have many beneficiaries in a poorer quintile but still distribute greater total money to richer quintiles if benefit amounts are larger there.

Pre-transfer ranking and post-transfer ranking can also change the reported incidence. If a transfer moves a household into a higher quintile, ex post ranking may show support concentrated in a higher income group even though the household was initially poor. Comparative tables should state the ranking variable and calculate alternative estimates where feasible.

An illustrative accounting example

Consider four hypothetical households with two children each, all eligible for an annual nominal family allowance of 1,000 currency units. If only three claim and one of those claims yields no positive payment, gross legal entitlement totals 4,000 but observed disbursement totals 2,000. The claim rate is 3/4, or 75 percent; positive-payment receipt is 2/4, or 50 percent; and realized gross disbursement is 50 percent of nominal statutory total. If 400 of the disbursement is subsequently offset by an explicitly attributable withdrawal of another benefit, the partial net contribution is 1,600. These are arithmetic properties of the stated hypothetical, not empirical estimates for any jurisdiction.

The point is substantive: an evaluation using legal eligibility would record four potential beneficiaries, a claims register would record three, and a payment register would record two. An income survey could record an amount smaller than either administrative disbursement if recall is incomplete. Without a common unit and reconciliation rule, analysts may call each number 'coverage' and produce apparently contradictory findings that are perfectly compatible.

Vertical redistribution

A targeted family transfer generally produces a larger proportional gain in disposable income for lower-income recipients than for wealthier recipients when the nominal benefit is comparable. Means-testing can further concentrate gross expenditures. Yet realized progressivity depends on the take-up gradient. Suppose an eligible low-income household has a lower probability of receiving payment because of documentation requirements, while a higher-income eligible household has a stable income record and automated access. The policy's statutory targeting can be more progressive than its realized incidence. Estimated redistribution must incorporate this selection.

The OECD's 2026 review of child-policy spending considers public cash transfers, tax measures, and in-kind services alongside the composition and adequacy of budgets. Its comparative analysis reports that child-benefit design, the scale of expenditure, and targeting interact in poverty reduction. A result for an OECD or European sample is an empirical finding for that sample; it does not determine a global ranking of universal versus means-tested institutions. The fiscal budget available to the scheme and the beneficiary population are indispensable parts of that comparison.

Horizontal redistribution and child costs

Horizontal redistribution addresses families with children relative to households with similar market incomes and fewer dependents. A child-specific cash payment changes the distribution across family structures even when the receiving family is not classified as poor. The magnitude depends on which children qualify, whether payment increases with family size, age or disability, and whether housing or care costs are separately subsidized. A program may treat all children symmetrically per capita while leaving substantial variation in real costs of disability, childcare, transport, or remote residence.

A single-parent family may experience greater absolute relief from an identical payment because its adult earnings capacity and unpaid caregiving constraints differ. A blended or shared-custody family may encounter attribution disputes because the statutory payee and the household paying ongoing expenses differ. Household-level incidence should therefore be complemented by a clear definition of the persons to whom the transfer is attributed.

Tax expenditure versus direct cash support

The OECD expenditure categories highlight that a tax deduction, nonrefundable credit, refundable credit, and direct payment have different income profiles. A tax deduction is most valuable to those with taxable income and a positive marginal tax liability. A nonrefundable credit may be limited by tax owed. A refundable credit can exceed tax liability and generate payment, but its take-up may depend on filing. A flat child allowance can reach a broad set of households if registration is effective; a means-tested allowance can direct higher payments toward low incomes at the cost of additional verification.

A distributional simulation should not assign the full face value of a nonrefundable credit to a household with no tax liability. Nor should it treat a gross high-income charge as irrelevant when estimating final resources. The policy's legal taxonomy determines how the accounting should be performed.

Poverty effects and budget constraints

Let the child poverty headcount be the proportion of children living in resource units below a stated threshold. A conventional static antipoverty contribution compares the headcount with observed tax-benefit resources against a counterfactual in which a defined transfer or transfer group is removed, keeping all other modeled conditions fixed. This is a mechanical accounting comparison, not a causal estimate of behavioral adjustment. If eligibility depends on market income and taxes, the counterfactual rules need to be rerun consistently.

The National Academies' child-poverty assessment evaluates U.S. policies under specified counterfactual program packages, costs, and behavioral assumptions. Such modeling is more informative than a single spending ratio because it connects budget scale, administrative mechanisms, and a poverty threshold. It also illustrates why transfer simulations should disclose fiscal offsets and uncertainty rather than treating every projected poverty reduction as an observed outcome.

Fiscal and institutional sustainability

Child and family benefits are supported through general taxation, earmarked contributions, or mixed fiscal arrangements. Their fiscal cost depends on age structure, number of qualifying children, take-up, indexation, benefit generosity, income distribution, enforcement, and economic conditions. A per-child universal payment can have predictable rules but a broad expenditure base. Means-testing can narrow gross beneficiaries while increasing administrative, appeals, and recertification burdens. Simulations of both arrangements should use the same demographic projections and assumptions on behavioral responses.

The appropriate cross-country unit is defined by the research question. Total family spending as a share of GDP addresses fiscal effort; average transfer per child addresses expenditure intensity; payment divided by recipient-household income addresses adequacy; poverty-gap reduction addresses changes below the poverty threshold. One indicator cannot substitute for the others.

Household bargaining and the allocation to children

Resources recorded at household level are shared imperfectly among individuals. The fact that a transfer is statistically attributed to children in a recipient household does not prove that children receive the full amount through consumption, care, or investment. Recipient identity, caregiving responsibilities, access to financial accounts, and household decision-making may shape use of the transfer. Empirical research needs actual child outcomes or detailed expenditure measures before claiming child-specific benefits beyond the household's income increase.

The analytic consequence is a two-stage distinction: first establish the monetary incidence at the household level, then investigate intrahousehold allocation and child outcomes. Direct allocation of the full cash value to every child as an individual gain would double count resources if multiple children share the same household payment.

Child Outcomes

From income receipt to child well-being

The outcomes of family transfers are causally downstream from a sequence of legal, administrative, financial, and behavioral events. An increase in entitlement can improve material conditions through cash income only when resources reach the relevant household, or when a credible expectation of payment changes its decisions. Once received, the transfer may affect food consumption, housing stability, healthcare use, educational investments, parental work allocation, savings, debt, and the ability to absorb temporary income shocks. Each pathway has its own measurement requirements. A higher amount recorded in a benefit ledger establishes a payment, not automatically an improvement in every child outcome.

The cross-country evidence review by Bastagli and colleagues (2016) documents reductions in monetary poverty and gains in school participation in many programs while finding more varied results for learning, anthropometric measures, and other outcomes. Increased school attendance is an empirical endpoint distinct from improved learning. Likewise, increased food expenditure and objectively measured nutritional status belong to different outcome domains. A sound synthesis preserves the measured endpoint rather than replacing it with the most favorable interpretation.

The effect of transfers also unfolds over time. A family may use a payment to avert a costly loan, maintain housing, finance school transportation, or cover essential medical care. Short-run expenditures and adult outcomes decades later therefore require different observation windows, research designs, and mechanisms. A static incidence table describes resources attributed to households in a particular period; it does not establish a lifelong developmental effect.

Child-benefit variation in Canada

Milligan and Stabile's study of Canadian benefit expansions exploits differences across provinces, years, and family types to evaluate changes in child tax benefits. The authors report improvements in some measures of child test performance, maternal health, and child mental health, with patterns differing by child sex and outcome. The study's identification rests on policy-driven variation rather than an unadjusted comparison of recipients and nonrecipients. It provides evidence that increases in family resources can influence child well-being under the conditions studied.

Those findings do not establish a universal outcome gain per dollar. Cross-country extrapolation would require information on initial incomes, education and healthcare services, the purchasing power of benefits, family composition, policy eligibility, and the amount actually received. Administrative participation and changes in statutory generosity may themselves vary across cohorts.

The Canada Revenue Agency's Canada Child Benefit statistics offer benefit-year data on recipient households, dependent children, income, and family characteristics. These figures describe the distribution of program receipt. A beneficiary table does not, standing alone, count all eligible nonrecipients, and it does not establish a causal child-outcome effect. A defensible participation estimate requires independently determining the legally eligible population for the same benefit year.

U.S. tax-credit income and learning

Dahl and Lochner's Earned Income Tax Credit study uses policy-induced variation in the U.S. EITC to estimate effects of family income on children's test scores. The original authors reported a baseline estimate of approximately six percent of a standard deviation in combined math and reading performance for an additional $1,000 of family income. A subsequent Lundstrom (2017) replication critique identified an income-coding error that reduced the instrumental-variable estimate by about 32 percent. In their published 2017 reply, Dahl and Lochner acknowledged this correction while maintaining that the qualitative conclusion and the reduced-form findings were unchanged. The corrected magnitude is therefore materially smaller than the originally reported baseline estimate. The intervention is an earnings-conditioned tax credit, and the identified response belongs to the study population and its empirical design. It cannot be directly assigned to unconditional universal child allowances or to countries with different labor institutions.

The study nevertheless supplies evidence for an economically important channel: family resources influence educational outcomes in at least some circumstances. Applying the result to a particular reform requires evaluating actual receipt, the distribution of additional income, labor-supply incentives, and the educational environment. The source supports an income–achievement effect under specified variation, rather than a general claim that every child benefit yields identical returns.

Long-run historical evidence

Aizer, Eli, Ferrie, and Lleras-Muney's Mothers' Pension study links historical U.S. program applications to later outcomes. Among male children in the historical sample, receiving a pension was associated with longer life expectancy, higher educational attainment, and better adult economic outcomes; the life-expectancy gain was approximately one year in the authors' comparison. The program operated in a specific early-twentieth-century welfare context, and the evidence does not license a universal contemporary lifetime-return coefficient. Its scientific contribution is to make the possibility of long-run effects of early family income support empirically testable.

Hoynes, Schanzenbach, and Almond's childhood Food Stamp exposure study examines geographic variation in the historical introduction of food assistance. The authors find improvements in selected later-life health outcomes and economic self-sufficiency measures among women. Food Stamps are a nutrition-assistance intervention, distinct in law and delivery from a general child-cash allowance. The study is relevant to the broader question of childhood resource protection precisely because the outcome and intervention are identified, not because all programs can be merged into a single effect.

These studies motivate examination of cumulative exposure, sensitive developmental periods, and the distinction between temporary and sustained support. They do not demonstrate that the timing, magnitude, and composition of every transfer can be ignored. Later-life outcomes may also depend on local opportunities, food and healthcare prices, and the services available alongside a transfer.

Advance tax-credit payments and the poverty metric

The temporarily expanded U.S. Child Tax Credit in 2021 illustrates the interaction between tax rules, advance disbursement, and statistical poverty accounting. The IRS advance-payment statistics document payments issued during July–December 2021. These administrative numbers establish a flow of payments to tax units; they do not, by themselves, estimate a causal reduction in child poverty. Final annual entitlements depend on tax-year eligibility and reconciliation, and the unit receiving an advance may not be identical to the child population later measured in a poverty survey.

In the U.S. Supplemental Poverty Measure, refundable tax credits are counted among family resources, subject to the SPM's specified income, tax, benefit, and expense definitions. The 2025 Census poverty report, released in September 2026, reports an official overall poverty rate of 10.2 percent and an official child rate of 13.4 percent for 2025; the overall SPM rate was 13.1 percent. The official measure and the SPM use different resource and threshold definitions. The numerical difference between them must not be interpreted as a stand-alone estimate of the effect of family transfers.

The Census Bureau's 2026 SPM correction notice documents revised threshold inputs affecting earlier data years. This creates a concrete data-vintage requirement for historical comparisons. A time series assembled from pre-correction and post-correction tables may contain differences attributable to statistical revision rather than policy change. Reproducible child-poverty analysis must identify the SPM vintage, the reference year, and whether thresholds or underlying data were revised.

The analytically decisive point is that tax-benefit entitlement, advance cash delivery, final tax reconciliation, and survey-based poverty estimates are four distinct sources of evidence. Their linkage must be established before an analyst attributes a measured poverty change to an identified child-transfer reform.

Brazil: household targeting, children, and continuity

Brazil's Bolsa Família combines a means-tested income-protection structure with benefit components and administrative rules that reflect family composition, including children. In its December 2025 payment announcement, the Ministry of Development and Social Assistance reported support for approximately 18.7 million recipient families and an average payment of R$691.37. These figures are monthly administrative observations for the reported period. They are not a count of all eligible families, all qualifying children, or annual continuous recipients.

The Brazilian program also illustrates how income transitions can affect continuity of support. The ministry describes a protection rule permitting some families whose income rises above the ordinary entry criterion to retain partial payments for a prescribed period. Such a rule modifies the discontinuity between eligibility and ineligibility and may matter for earnings incentives and household budgeting. It must be evaluated under the provision effective in the period studied; historical benefit rules, benefit values, and transition requirements should not be transferred uncritically between years.

An administrative family count and an estimate of protected children require a family-composition bridge. One recipient family can contain several children or none, depending on the applicable program rules and case mix. Child-level distributional incidence demands child counts, ages, and residence within the household, as well as payment amounts.

Deprivation, conditionality, and household allocation

Income poverty headcounts indicate whether an income aggregate crosses a threshold, but persistent or episodic deprivation may involve food insecurity, unstable housing, debt, childcare shortages, transport interruptions, and healthcare barriers. A transfer can reduce a household's poverty gap without moving it above the threshold; conversely, a measured reduction in monetary poverty does not establish that all nonmonetary needs are met.

The OECD's 2026 child-policy analysis considers how cash, taxes, and services interact in reducing child poverty and disadvantage. Its comparative approach illustrates the need to evaluate not only spending, but also benefit composition and institutional fit. Child-specific outcomes require defined education, health, deprivation, and economic indicators, not a generic claim that greater expenditure improved every form of well-being.

Some family-transfer systems link payment or continued eligibility to school participation, health visits, or administrative reporting. Conditionality can alter behavior through incentives and engagement with services, but it may also increase the procedural burden faced by families when services are distant or unavailable. The cash-transfer evidence review stresses the role of design and implementation. For causal attribution, a bundled transfer-plus-service intervention must be distinguished from the effect of money alone.

Finally, measured household receipt is not proof of equal distribution among household members. Payment recipient identity, caregiving, shared custody, family decision-making, and individual needs may shape how resources are spent. Household-equivalized income is a useful distributional convention, but it does not directly observe the consumption or welfare of each child. The monetary-incidence stage and the child-outcome stage therefore require separate analyses.

Evidence Verification and Sensitivity

Evidentiary hierarchy and verification scope

This dossier combines four evidence classes. International legal instruments establish normative obligations, institutional possibilities, and the meaning of social protection categories. Official administrative reports and statistical releases establish quantities and documented program procedures for specific periods. Methodological manuals define indicators and data transformations. Peer-reviewed empirical research supplies evidence about selected mechanisms and outcomes within stated identification designs. A legal text cannot demonstrate actual payment; a payment register cannot identify an otherwise unobserved eligible population; and an aggregate before-and-after comparison does not, without further identification, demonstrate a causal outcome effect.

For the primary legal sources, the analysis examines the ILO family-benefit Convention and the Social Protection Floors Recommendation. For international measurement, it uses the ILO's age-specific protection report and the ASPIRE data documentation. For domestic practice, it compares the documented purpose of HMRC claimant and payment counts, tax-based payment records in the United States, Canada Revenue Agency beneficiary tables, and the selected Brazilian monthly announcement. The empirical work represents outcomes in precisely identified study settings, including outreach experiments, tax-credit variation, and historical pension and nutrition-assistance cohorts.

The research establishes neither a new global statistical sample nor a fresh household microdata replication. It makes no original numerical estimate of eligible nonclaimants worldwide, no newly estimated benefit-incidence curve, and no new causal parameter for child outcomes. The original analytical contribution is the reconciliation of the quantities established by these different classes of evidence and the explanation of their implications for comparative inference.

Legal time and statistical time

Every legal statement has a jurisdiction, applicable population, instrument, effective period, and rule of interpretation. The Convention specifies internationally recognized standards while national coverage depends on ratification, accepted branches, and implementation. Recommendation No. 202 provides a framework for national protection floors without itself creating one worldwide statutory allowance. A country-level comparison therefore requires the particular legal text in force for that reporting period.

Statistical source dates require separate treatment. HMRC's 2025 Child Benefit data describe an August 2025 administrative stock but were published in 2026. Brazil's ministry reports payments for December 2025, so the numbers belong to that administrative cycle. The IRS counts tax-credit advance payments during a defined six-month interval in 2021. International reports may publish a 2024 document with selected observations from 2023 or earlier years. A manuscript must always name the reference year rather than silently substituting the publication year.

For the Supplemental Poverty Measure, the 2026 threshold revision notice makes exact statistical vintage part of the evidence itself. A historical trend can change when a statistical agency corrects its series. Researchers should preserve both the year analyzed and the edition of the table or methodology consulted.

Population and observation-unit integrity

The central comparative hazard is denominator substitution. A child coverage percentage, a family-claim percentage, a recipient-household total, and a payment count may all be accurate within their proper universe. Their numerators and denominators refer to different populations.

A program evaluation should explicitly identify whether its numerator is eligible claimants, families paid positive cash, children living with paid households, tax filing units, or all payment transactions. It should then identify whether its denominator is legally eligible units, all resident children, all families with children, a sampled household population, or a modeled tax-filing population. The World Bank ASPIRE framework distinguishes individual coverage, beneficiary incidence, and benefit incidence. It also counts indirect beneficiaries in beneficiary-household analysis. That convention is useful for welfare allocation, but does not mean that each household member is a distinct legally registered claimant.

A period is equally important. A household receiving one monthly installment is not necessarily covered for all 12 months. Family composition and eligibility can change during the year, and retrospective reconciliation can alter gross amounts. For annual equivalized income analysis, the appropriate record is the flow of payments and tax consequences over the defined period, accompanied by a rule for changing family composition. For operational monitoring, monthly active paid caseloads and continuity measures may be more informative.

Survey and administrative error

Surveys can underreport benefits, omit informal income, miscode program categories, and exclude households with unstable accommodation or language barriers. Administrative systems can contain duplicate identifiers, delayed registration changes, incorrect eligibility classification, and gaps in linked civil, tax, and labor records. A larger administrative sample is not automatically a complete population frame.

Two data sources should be reconciled through a documented program codebook and unit definition, with adjustments for the time window and known reporting lags. If a household survey captures social assistance but cannot distinguish a child allowance from a general minimum-income payment, the analyst must not label all reported transfers as child-specific without corroboration. If the administrative register records awards but excludes people who never filed, its caseload cannot independently determine take-up.

Eligibility simulation introduces a separate error structure. Income reporting may be incomplete; custody, legal residence, disability status, and asset provisions may be difficult to reconstruct. The denominator for eligible nonreceipt should therefore be accompanied by sensitivity analysis under alternative assumptions rather than treated as an exact observed count.

Distributional ranking and weighting

The choice of income ranking changes apparent progressivity. If households are grouped using disposable income after transfers, recipients moved upward by the benefit can appear in higher income quintiles. Ranking by pre-transfer income may better identify the group initially targeted, but it too depends on how the counterfactual tax-benefit system is constructed. Researchers should report which income variable determines the quintile and calculate both rankings where the research question requires it.

Household and child weighting answer different questions. Under household weights, a family with four children counts once; under child weights, it contributes four children. A program designed to pay larger amounts to large families can therefore appear to have different benefit incidence under these conventions. Equivalence scales also influence measured poverty and relative need. No international table should compare outcomes across countries without recognizing the scale and threshold definitions.

Gross monetary benefit incidence, net tax-benefit incidence, and comprehensive fiscal incidence should remain separate. The first allocates observed transfers. The second evaluates effects on disposable income after relevant fiscal interactions. The third considers the broader distribution of tax financing and public services, requiring additional behavioral and valuation assumptions. The source data necessary for each level are materially different.

Causal identification and external validity

A family may receive a benefit because of low income, poor health, unemployment, separation, or other risks that independently affect child outcomes. Consequently, comparisons of recipient and nonrecipient children may be confounded, even with many observed controls. Credible causal work depends on exogenous variation, valid assignment or identifying assumptions, and transparent reporting of the outcome and treatment.

The SNAP take-up experiment identifies the effect of outreach and application assistance for a defined older adult population. It demonstrates an access mechanism but does not estimate a universal treatment effect for child benefits. The Canadian benefit study identifies effects of policy variation within its own historical and institutional conditions. The EITC child-achievement study estimates a response to income changes generated by its tax-credit design. These investigations cannot be pooled as though they randomize the same policy.

For any future causal evaluation, researchers should describe treatment assignment, participation or compliance, attrition, spillovers, geographic relocation, anticipation, changes in labor supply, unobserved family shocks, and the timing of outcomes. Historical linkage studies additionally require scrutiny of selection into surviving or successfully matched records. Point estimates need uncertainty intervals and a clear definition of the population to which they refer.

Reproducible sensitivity tests

A comparative study grounded in individual-level data should carry out the following diagnostic comparisons before making a major distributive claim:

  1. Estimate claim registration, positive cash payment, and net-of-tax receipt separately, where administrative data permit.
  2. Define eligible-population denominators under current law and state how household surveys or linked registers approximate unobserved nonclaimants.
  3. Recalculate benefit incidence under pre-transfer and post-transfer rankings, using household weights and child weights.
  4. Recalculate monetary poverty under at least one documented equivalence-scale alternative and report poverty gap alongside headcount.
  5. Separate annual entitlement from within-year payment timing, delayed installments, retrospective awards, and recoveries.
  6. Examine differences by child age, family size, caregiver status, disability, low-income status, and relevant geographic or documentation barriers.
  7. Test sensitivity to benefit underreporting in surveys and eligibility misclassification in administrative sources.
  8. For causal studies, identify the intervention, estimand, counterfactual, sampling frame, standard errors, exposure period, and all material co-interventions.

These are ordinary research controls derived from the evidence and definitions above. They are not advanced as a new formal model or independent methodology.

Evidence limits and editorial implications

The verified evidence supports a set of conditional conclusions: legal protections have internationally recognized foundations; effective child-benefit coverage remains incomplete; administrative participation is sensitive to access and application procedures; and particular income-support programs have improved measured outcomes in particular studies. The sources do not establish an invariant country-independent response to each unit of benefit expenditure.

The principal unresolved scientific question in the field is how to obtain internationally comparable eligible-population denominators linked to net payment records and child outcomes. That is a research frontier rather than a factual defect in the present synthesis. The manuscript explicitly records the absence of new household-level data, country-by-country legal audit, and independently reproduced econometric estimates. Its citations establish the scope of every empirical claim and keep institutional design, descriptive measurement, and causal inference distinct.

Scientific Synthesis and Research Agenda

The relationship between entitlement and incidence

The research yields a coherent interpretation of the path from law to economic effect. A statutory provision defines potential inclusion. Administrative rules influence whether the household knows about, claims, and can maintain its entitlement. Payment systems determine whether and when cash is delivered. Taxes and other benefit rules alter the change in disposable resources. The composition of the household and the needs of children determine how that change is represented in distributional accounts. Behavior and access to public and private services influence the measured consequences for children.

Each stage is supported by a different type of evidence. International labor standards and national statutes describe legal possibilities. Administrative reports document registered and paid populations. Household surveys and microsimulation establish income distribution under explicit assumptions. Experimental and quasi-experimental studies identify selected causal consequences. The interpretive task is to connect these stages without pretending that their data form one interchangeable measure.

An expansion of eligibility changes the set of potential beneficiaries. Automatic enrollment or administrative assistance may change actual receipt within that set. A benefit increase changes payments conditional on receiving them, as well as the associated fiscal requirement. Improving payment reliability alters liquidity and the exposure of children to short-run income gaps. Changing benefit taper rates affects net-income incentives and the distribution of support around eligibility thresholds. These interventions target different problems and therefore require different evaluations.

Statutory and realized redistribution

A statutory benefit may be progressive in formula while its realized distribution is less progressive because eligible low-income families face more difficulties claiming it. The converse is also possible in a particular population: automatic registration of vulnerable families could generate greater realized progressivity than a simple legal-rate chart would suggest. Which outcome occurs is a question for evidence, not an assumption.

The relevant comparison is between simulated legal payments under full take-up and the observed distribution of cash and net resources. Differences can arise from missing claims, withheld benefits, tax charges, incomplete recertification, delayed payments, or other program interactions. The UK Child Benefit administrative distinction provides a particularly clear warning: a registered claim can exist without a positive payment.

An analysis of realized redistribution should therefore state the policy counterfactual, identify whether tax and other benefit interactions are modeled, and describe the distribution of nonreceipt among statutorily eligible units. Where the tax-benefit system is complex, a documented microsimulation such as the European Commission's EUROMOD framework can help make such comparisons reproducible for covered countries and years. Model coverage, policy rules, survey inputs, and calibration remain essential boundaries.

Vertical, horizontal, and temporal redistribution

Three aspects of redistribution appear throughout the evidence. Vertical redistribution concerns the distribution of resources across income and need. Horizontal redistribution concerns differences associated with child number, age, disability, caregiving, and comparable household circumstances. Temporal redistribution concerns the regularity and life-course timing of support. A scheme can perform differently in each domain, and a single aggregate expenditure indicator is therefore insufficient.

A means-tested transfer may achieve a high share of spending among low-income recipient households while imposing recertification burdens and difficult eligibility transitions. A categorical allowance may reach a wider population with simpler rules yet provide limited purchasing-power protection to children in deep poverty unless the benefit level is sufficient. A refundable tax credit can convert fiscal entitlement into cash but may depend on filing and reconciliation procedures. None of these empirical propositions establishes a universal design ranking; program success depends on the objective, financing, population, and administrative institutions.

The temporal dimension is particularly relevant to families with restricted access to credit. Two households can receive equal annual sums but experience different hardship if one receives consistent monthly payments and the other faces unpredictable delays. Assessing this channel requires recorded disbursement dates, household financial conditions, and outcomes observed at a time resolution capable of detecting income instability.

Implications for program evaluation

An analytically complete evaluation should report the following elements in a shared framework: the statutory qualifying population; the eligible population estimated under actual law; claims and application completion; positive receipts by period; household and child denominators; gross and net monetary support; poverty headcount and poverty gap; fiscal and administrative cost; and clearly named child well-being outcomes. Each variable should be supplied with a data source, reference year, and jurisdiction.

The OECD's child-policy research places monetary support alongside services and other institutions, while the World Bank ASPIRE methodology explains how different incidence measures answer different allocation questions. The synthesis of these approaches supports evaluation that treats reach, adequacy, distribution, and outcomes as jointly relevant. A program can be assessed as successful in delivering claims but less successful in producing positive payments, or successful in reducing poverty gaps while leaving learning outcomes unaffected. These judgments are meaningful when the criteria are declared in advance.

Causal evaluation adds a second layer. Changes in benefit formulas, enrollment systems, or payment schedules should be studied using variation plausibly independent of potential outcomes, or evaluated through explicitly justified designs and robustness checks. Findings should be assigned to the studied program rather than extrapolated to every transfer named 'family benefit.'

Counterfactuals suitable for future research

Four comparative simulations are particularly informative. The first holds statutory rules fixed and assigns payments to every eligible nonrecipient, showing the mechanical maximum available through full legal take-up, subject to program interaction rules. The second varies the administrative probability of claiming and payment, using empirical evidence on application costs and assistance when available. The third changes the payment schedule, taper, or benefit value while holding receipt assumptions transparent. The fourth changes the timing or continuity of payments without altering annual nominal entitlement. Each isolates a different policy margin, and their differences help identify whether poverty reduction is constrained mainly by legal exclusion, nonreceipt, insufficient amounts, or liquidity.

For every scenario, reporting should separate additional gross expenditure from changes in net disposable resources and from any modeled behavioral response. Analysts should report both family- and child-weighted effects, recognize financing needs, and document the sensitivity to household composition and eligibility misclassification. Child outcome changes that are not directly estimated should be labeled projections or hypotheses, rather than empirical results.

Research priorities

The first priority is better linkage between administrative awards, actual payments, and the eligible population. Household surveys are needed to observe nonrecipients, while administrative records are essential for accurately measuring payment timing and amount. Such linkage should respect national data-protection requirements and use only the data necessary for the research purpose.

The second is measurement of non-take-up gradients by socioeconomic vulnerability. Studies should estimate whether low-income, migrant, remote, disabled, single-parent, or otherwise administratively constrained households experience different probabilities of payment conditional on legal entitlement. Their policy interpretation must account for the precise national eligibility rules and the availability of alternative support.

The third is direct investigation of children’s experience of timing and continuity. Exposure to a regular transfer may matter differently for a household with ample savings than for a household with recurrent rent arrears. Evidence should test the timing channel and avoid attributing all observed differences to annual benefit generosity.

The fourth is the net distributive interaction of benefits, tax charges, and withdrawals. Administrative receipt is an intermediate variable. Research should estimate whether household disposable income actually rises, by how much, and for how long, while preserving the distinction between the accounting counterfactual and behavioral adjustments.

The fifth is the establishment of comparable indicators across legal regimes without erasing local institutional meaning. The ILO's child-level effective-coverage rate, the UK's eligible-child claim rate, Brazil's recipient-family count, and a U.S. tax-unit payment measure describe different units. A useful international analysis should display the units side by side and reconstruct comparability only when the necessary data exist.

Authorial analysis and scientific contribution

The authorial contribution of this monograph is a source-grounded synthesis connecting eligibility, claims, positive disbursement, tax-benefit interactions, family composition, incidence, and child outcomes; an explanation of why official indicators that appear contradictory may be simultaneously correct under their respective definitions; an explicit treatment of benefit timing as a distributive property; and a documented agenda of sensitivity tests and empirical counterfactuals. The distinctions and accounting examples arise from standard legal, fiscal, and statistical relationships. No new theory, doctrine, principle, or independently branded conceptual system is claimed.

The strongest scientific interpretation is that child and family transfers achieve their distributive purpose only through effective and measurable delivery. A generous benefit with weak receipt may produce limited realized support; a universal payment with high coverage but low adequacy may fail to close deep poverty gaps; and a payment that arrives late may leave crucial periods of child need unprotected. Each empirical problem calls for a different instrument of diagnosis and evaluation.

Conclusion

Child and family transfer take-up is a central determinant of realized redistribution. Legal entitlement, claim registration, positive disbursement, net disposable-resource contribution, and child well-being are different quantities. Each must be linked to a defined population, observation period, jurisdiction, and evidentiary source. These distinctions make it possible to compare contributory family benefits, categorical child allowances, means-tested social assistance, and tax-based support without conflating their legal or statistical structures.

International legal instruments establish recognized protection objectives. Global coverage data reveal substantial gaps in effective child-family support. National administrative evidence shows how claimants, payments, and families with eligible children can diverge. In the United Kingdom, the separation of claimed Child Benefit and positive payments is directly observable. In Brazil, the selected Bolsa Família release records monthly payments to recipient families under a specific set of income and administrative conditions. In Canada, benefit statistics describe receipt, while policy-variation studies address child outcomes. In the United States, tax-credit payments, tax reconciliation, and the Supplemental Poverty Measure demonstrate why fiscal records and poverty estimates require consistent reference dates and definitions.

The causal literature shows that administrative assistance can improve access in studied populations and that additional family resources can improve some educational, health, and later-life outcomes in settings with credible identification. The evidence also shows meaningful differences among endpoints and programs. It supports careful empirical assessment of take-up, timing, adequacy, and service interactions rather than a single outcome coefficient or a context-free ranking of universal and targeted schemes.

A rigorous evaluation therefore begins with statutory eligibility, measures actual claiming and receipt, reconstructs net resources, and connects those resources to child outcomes through appropriate observation and causal designs. Its credibility depends on correctly matched numerators and denominators, source dates, legal rules, and transparent assumptions. The present work provides a comprehensive international synthesis and a reproducible analytic agenda grounded in those requirements. Its public release remains a separate publisher decision and technical event.

References