RESEARCH DOSSIER
Maternity Cash Benefit Universality and Financing
Abstract. Maternity cash benefits insure or otherwise compensate income interrupted by pregnancy, childbirth, and related leave, while interacting with employment law, reproductive health, family policy, and social protection. Their claim to universality must be judged by actual receipt among all women giving birth, including women working informally, independently, intermittently, without wages, and across changes of employer or residence. This monograph examines the historical and normative foundations of maternity income security, eligibility architecture, risk-pooling and taxation arrangements, fiscal and actuarial accounting, international comparative experience, and the limits of coverage statistics and causal policy evidence. The International Labour Organization (ILO) estimated that 36.4 percent of women with newborns worldwide received a maternity cash benefit in 2023, compared with 29.6 percent in 2015; the remaining population exposes an important gap between labor-law protection and population-wide financial security. Comparative cases demonstrate meaningful differences between government payment, government reimbursement of employer advances, compulsory contributory insurance, and noncontributory transfers. The analysis establishes that legal availability, claim eligibility, administrative access, receipt, replacement adequacy, and continuity of financing require separate measurements. It derives accounting and measurement implications from existing social-insurance and public-finance concepts, examines the assumptions behind commonly proposed reforms, and identifies source-grounded research priorities. All empirical figures are tied to the reporting periods and populations of their original sources. No new doctrine or independently validated causal estimate is claimed.
Keywords: Maternity cash benefits; maternity protection; social insurance; universal social protection; effective coverage; financing; employer liability; informal employment; contributory eligibility; general taxation; replacement rates; fiscal incidence; take-up; maternity allowance; risk pooling; SDG 1.3.1.
Introduction
The central question of maternity cash-benefit policy is how a society maintains income when childbirth temporarily interrupts earnings and care responsibilities expand. In social-insurance terms, the event is relatively foreseeable for some individuals but uncertain in timing and incidence across a broad contributor pool. For families, the consequence is a simultaneous loss of labor income and increase in needs. For firms, obligations can arise unevenly across workplaces and years. For governments, the allocation of financing determines the distribution of cost, resilience of benefits, labor-market incentives, and likelihood that women beyond standard employment will receive support.
Maternity cash protection consists of several linked institutions. Employment-protected maternity leave concerns a right to be absent from work; the maternity cash benefit concerns money paid during the relevant interruption; maternity medical benefits address healthcare services; a birth grant is generally a lump-sum family transfer; parental benefits may be shared or reserved for particular caregivers. The OECD distinction among leave types is essential because cross-country tables sometimes combine maternity-specific and parental provisions, and many countries integrate income support in parental rather than separate maternity programs (OECD, 2026). Combining nominal leave weeks across unlike entitlements conceals who qualifies and who actually receives payments.
The most important international observation is the gap between program architecture and receipt. The ILO's global maternity coverage assessment reports 36.4 percent effective coverage in 2023, equivalent in its executive summary to approximately 85 million women with newborns lacking maternity cash benefits (ILO, 2024). These are estimated population-wide measures, not the percentage of insured workers satisfying a claim test. They should not be interpreted as the failure rate of a single statutory program, and the ILO cautions that estimates across its successive reporting editions reflect methodological changes and country revisions.
This dossier evaluates the relationship among legal coverage, financial design, and benefit receipt. Its evidence base consists of the texts of ILO conventions and recommendations, the European Union's maternity directive, the ILO World Social Protection Report 2024–26, OECD comparative definitions, World Bank legal-measurement methodology, official national benefit descriptions, and selected causal empirical studies. A national administrative page is treated as evidence about the population and period specified by that page, rather than as a universal or historically permanent rule. The analytical synthesis compares mechanisms already established in public-finance and social-insurance scholarship; illustrative equations are identities or scenarios, not estimated causal laws.
Conceptual and Historical Foundations
Maternity income protection as a distinct social contingency
In the earliest employment-based provisions, maternity protection was commonly tied to industrial labor, an identifiable employer, and a period of absence around childbirth. The Social Security (Minimum Standards) Convention, 1952 (No. 102) established maternity as a social-security branch alongside other contingencies. Its Part VIII addresses medical care and suspension of earnings; Article 50 requires a periodic cash payment calculated under the applicable standard-beneficiary provisions, and Article 52 allows specified duration limits, with conditions concerning nationally authorized longer absence (ILO, 1952). These provisions distinguish replacement of earnings from payment for medical treatment.
The revised Maternity Protection Convention of 1952 made collective financing especially explicit. Article 4(4) referred to compulsory insurance or public funds; Article 4(7) specified that insurance contributions or payroll taxes financing maternity protection should be levied for men and women without distinction by sex; Article 4(8) barred individual employer liability for the benefit. The historical importance of these provisions lies in separating the collective social cost of reproduction from the realized births of a particular firm's employees (ILO, 1952b).
The Maternity Protection Convention, 2000 (No. 183) expanded the substantive framework while retaining collective finance as its central design. Article 4 establishes at least fourteen weeks of maternity leave for women to whom the convention applies; Article 6(3) requires earnings-related cash benefits of at least two-thirds of previous earnings counted for benefit calculation. Article 6(4) provides comparability for other benefit-calculation methods. Article 6(5) requires qualifying conditions achievable by a large majority within the convention's protected population; Article 6(6) provides for adequate social-assistance benefits, subject to the applicable means test, for women who fail to qualify. Article 6(8) directs provision through compulsory social insurance or public funds, or a nationally determined method, and restricts direct employer liability while specifying historical and collectively agreed exceptions. This is a nuanced legal standard: it does not establish a single worldwide administrative program, nor should its two-thirds benchmark be presented as the observed replacement rate in every jurisdiction (ILO, 2000a).
The accompanying Maternity Protection Recommendation No. 191 recommends raising cash benefits, where practicable, to full prior earnings; it also recommends at least eighteen weeks of maternity leave. These recommendations are analytically distinct from the convention's minima and from national entitlements (ILO, 2000b). The Social Protection Floors Recommendation No. 202 places basic income security in maternity within a broader life-course guarantee, alongside essential maternity healthcare, nationally defined standards of adequacy, and universality grounded in solidarity (ILO, 2012). Its institutional perspective is wider than a contract of employment or a contribution record.
The Transition to Formality Recommendation No. 204 asks countries to extend maternity protection progressively to informal-economy workers in law and practice, adapting insurance administration, benefits and contributions to contributory capacity (ILO, 2015). This is directly relevant because most employment-based eligibility systems assume documented earnings and periodic contributions, precisely the evidence that many informal workers cannot reliably provide.
Universalism has several measurable meanings
Universality is often asserted in different senses. Population universality means all women experiencing the defined maternity contingency are protected by an income guarantee, subject to lawful residency and other applicable criteria. Occupational universality means all covered occupational categories have a route to entitlement. Formal legal universality concerns statutory scope. Administrative universality requires practical enrollment, information, and non-discriminatory claims processing. Financial universality requires that benefits actually reach those eligible without an intolerable income or access burden. The distinctions matter because an entitlement that technically extends to independent workers may operate through voluntary contributions that systematically leave low-income mothers outside the insured pool.
A full assessment therefore follows the sequence: statutory entitlement; potential qualifying status; actual enrollment or recognized residence; fulfillment of conditions; completed application; accepted claim; receipt of payment; adequacy and timing. The transitions can fail independently. A benefit can be generous for a narrow insured population and weak as a universal measure. Conversely, a basic tax-funded transfer may reach many mothers while replacing only a small fraction of prior earnings among higher-wage recipients. Evaluating schemes on a single axis conceals this structure.
Comparative normative architecture
Three legal benchmarks should be kept distinct. Convention No. 102 defines minimum social-security standards for covered classes, including a maternity cash contingency and specified adequacy conventions; Convention No. 183 advances maternity protection and its cash-benefit replacement benchmark; Recommendation No. 202 supplies guidance for nationally defined basic social-protection floors. The ILO's comparison of maternity standards identifies different protection populations, duration standards and reference-beneficiary methods across the instruments (ILO, 2024). A legal compliance assessment must establish whether the state ratified the relevant convention, accepted the relevant branch of Convention No. 102 where applicable, and how the domestic legal order implements it. A recommendation is normative guidance; a convention's obligations depend on its status and terms for the state.
Within the European Union, Directive 92/85/EEC provides for at least fourteen weeks' maternity leave for pregnant workers and related categories within its scope. Article 11 requires pay maintenance or an adequate allowance, as defined by reference to income available during health-related absence and subject to national ceilings; it allows qualifying rules but prevents prior-employment conditions exceeding twelve months immediately before the presumed date of childbirth. It is a regional employment protection instrument, and it does not establish that all residents receive one common EU-level cash payment (Council of the European Communities, 1992). The EU approach illustrates how a floor of rights can coexist with nationally diverse financing arrangements.
The financing logic of solidarity
Employer-liability systems link the cost of maternity leave to the employer of the woman who gives birth. Collective insurance and general-revenue transfers distribute financing across a wider group, including, depending on the design, workers of all genders, employers as a group, taxpayers, or public budgets. This redistribution serves several functions: reducing the concentration of childbirth expenditure in small workplaces, lowering the dependence of a claim on a particular employer's liquidity, and broadening the base over which maternity-related costs are borne.
The economic effect is conditional. A contribution-financed arrangement can still impose an effective tax wedge on particular labor categories; an employer can still act as payment intermediary; and an insurance fund can become underfinanced or administratively inaccessible. Broad financing improves risk pooling only to the extent that obligations, registration, cash-flow continuity, and reimbursement mechanisms are enforceable. Moreover, collective financing alone does not guarantee universal entitlement when benefit eligibility still depends on stable covered employment. The ILO's comparative account of financing methods documents precisely this coexistence of social insurance, public funds, employer liability and mixed arrangements (ILO, 2024).
Maternity Eligibility
The unit of entitlement
An eligibility rule specifies who can claim, which event triggers payment, how earnings or residence are established, what lookback period is relevant, and how overlapping benefits are coordinated. Statutory workers, domestic workers, independent earners, platform workers, seasonal and agricultural workers, cross-border workers, migrants, recent labor-market entrants, unemployed insured persons, and people engaged principally in unpaid care face distinct evidentiary challenges. A robust comparative record reports each group explicitly instead of saying that the scheme covers “women” or “workers” without qualification.
The relevant unit may be the claimant, a birth, an adoption or other legally recognized caregiving event, or a period of leave. A person-level beneficiary count, a count of paid claims, and a count of childbirths can therefore differ. Multiple births, repeat claims within a reporting year, shared parental arrangements, pregnancy loss and adoption eligibility make the numerator definition material. For international effective maternity coverage, the ILO reports the share of women giving birth who receive maternity cash benefits, with its modelled methodology and national source inputs; that population denominator should remain visible whenever the number is quoted (ILO, 2024, coverage indicator discussion).
Contributions, work histories and discontinuous employment
Contribution thresholds link present eligibility to past insured earnings. Such thresholds may prevent opportunistic short-duration participation, help establish employment attachment, and facilitate record verification. They also create discontinuities. A worker with a few hours below the threshold can face zero entitlement while another above it receives a substantial replacement payment. Eligibility is particularly sensitive to pregnancy-related employment interruptions, short-term contracts, employer changes and seasonal work.
Canada's federal Employment Insurance system illustrates this mechanism. The official maternity eligibility rules require, for the specified regular employee pathway, a reduction in regular weekly earnings exceeding 40 percent for at least one week and 600 insured hours in the applicable qualifying period, normally 52 weeks or since the most recent claim, whichever is shorter (Government of Canada, 2026a). The maternity benefit schedule provides up to fifteen weeks at 55 percent of average insurable weekly earnings, subject to the specified cap; the posted 2026 maximum is C$729 per week. This is a time- and program-specific illustration, not a universal Canadian entitlement: Quebec has a separate parental insurance regime, and self-employed people encounter different contribution and enrollment conditions.
Canada also offers a special-benefits pathway for self-employed people, demonstrating how separate affiliation rules can widen nominal occupational coverage while retaining participation constraints (Government of Canada, 2026b). These arrangements show why country-level coverage cannot be inferred from a headline maximum benefit. Measurement requires the proportion of eligible pregnant persons who meet the insurance-hours or self-employment participation criteria and actually obtain the payment.
Employment-related and supplementary access channels
The United Kingdom separates employer-administered Statutory Maternity Pay (SMP) from Maternity Allowance available to some people excluded from SMP. The statutory maternity payment schedule provides up to thirty-nine paid weeks, consisting of six weeks at 90 percent of average weekly earnings and thirty-three weeks at the lower of the statutory weekly amount or 90 percent of earnings. The posted 2026–27 weekly flat-rate limit is £194.32 (UK Government, 2026a). The Maternity Allowance eligibility guidance identifies self-employed people, employees who do not qualify for SMP, people who recently stopped working, and specified unpaid work for a partner's business as potential beneficiaries; detailed tests vary with their circumstances (UK Government, 2026b).
These pathways illustrate a financing and eligibility distinction. Multiple schemes can reduce gaps left by employment attachment, yet they may impose different earnings tests, filing procedures, payment rates and contribution conditions. A claimant moving from short-term paid work into self-employment can be classified differently from an otherwise similar salaried claimant. Any claim that the system is universal must be tested against people who qualify for neither route and against delays or nonreceipt among those who do.
Insurance membership, categories and insured status
Brazil's INSS salário-maternidade guidance, updated in July 2025, specifies a maternity benefit typically lasting 120 days after childbirth and lists employees, domestic workers, individual contributors, optional contributors, and unemployed persons retaining insured status among potentially eligible groups (INSS, 2025). It also distinguishes claims submitted through an employer from claims submitted to the social-security authority. This provides an example of broad insured-category architecture and the need to differentiate statutory inclusion from actual contributions, recorded status, and payment receipt. The correct contemporaneous rules for qualifying periods and procedures must be established from the applicable benefit and regulatory instruments for a specified claimant and date; the public agency overview is insufficient to infer that every resident receives the benefit.
The point generalizes. Broad category lists may coexist with relatively strict minimum contribution requirements for some categories, different evidence standards for rural work, and distinct administrative obligations for employees and independent earners. Official lists should be coded together with their qualifying rules and the actual population distribution across categories.
Administrative rights and the meaning of access
A benefit with an accessible legal definition can still fail at registration, birth certification, application, verification, bank-account delivery, or appeal. An eligibility review must ask whether the claimant can apply before birth, whether a claim can be retroactive, what identity documents are accepted, whether rural or informal work can be evidenced, and whether dependent migrants or seasonal workers can establish required ties. The World Bank's Parenthood methodology distinguishes legal frameworks, supportive frameworks and expert evaluations of implementation; its standardized reference person is important for interpreting any legal score (World Bank, 2026). These measures supply structured comparisons but do not measure the complete distribution of actual maternity benefit receipts.
Eligibility also interacts with gender equality. A direct employer obligation can create incentives to screen applicants perceived as likely to require expensive leave, although the magnitude of any discriminatory behavior is an empirical question requiring design-specific evidence. Labor-market segmentation may concentrate insured positions among higher-income urban employees while informal and own-account workers remain outside. A policy can accordingly exhibit excellent compliance within covered firms and a low population-wide rate of income protection.
To make entitlement verifiable, a country study should record its legal effective date, protected person definition, insured work categories, qualifying weeks or contributions, event definitions, replacement wage base, caps and floors, application pathway, payer, final financing entity, enforcement mechanisms and appeal rights. Those fields provide the minimum context needed to connect legal design to observed take-up.
Payment Financing
Who pays, who disburses, and who bears the risk
A financing description must distinguish the legal benefit provider from the first disbursing agent and the final economic financier. Under direct employer liability, the firm pays the claimant and retains the net expense. Under reimbursed statutory pay, the employer may transmit the payment but recover much or all of its cost from public funds. Under social insurance, a pooled institution finances cash benefits from contributions, sometimes supplemented by tax transfers. Under noncontributory systems, governments can pay directly through general revenues. A mixed regime combines two or more mechanisms; classifying it as either entirely public or entirely private discards important financial information.
The distinction is observable in the United Kingdom. Employers usually pay SMP through payroll, but under statutory payment recovery rules can generally recover 92 percent of qualifying payments from HM Revenue and Customs, and qualifying small employers can recover 109 percent (UK Government, 2026c). These percentages describe statutory reimbursement, not the worker's replacement rate. In the 2026–27 regime, small-employer relief is subject to an applicable previous-year Class 1 National Insurance contribution threshold of £45,000. Thus, calling SMP “employer-financed” solely because it appears on the firm's payslip would systematically overstate the employer's retained maternity expense. Its employer administration and collective public reimbursement must be recorded separately.
New Zealand offers an alternative administrative design. The government-funded parental payments are paid through Inland Revenue to qualifying employees and self-employed persons for up to 26 weeks. The country's public guidance expressly separates the right to take leave from eligibility for a leave payment, and notes that an employee who does not qualify for a particular leave entitlement may nevertheless qualify for the public payment if the applicable work and nonwork conditions are met (Employment New Zealand, 2026). The program integrates childbirth-related support into a parental-benefits framework, underscoring why it should not be coded automatically as a separate maternity-only benefit in a cross-country dataset.
South Africa's Unemployment Insurance Fund maternity guide identifies a claim route for insured workers taking maternity leave and a benefit duration described as seventeen weeks, subject to the program's statutory provisions and eligibility conditions (South African Department of Employment and Labour, 2013). This is an example of placing the maternity contingency within a broader contributory social-insurance institution. The administrative guide alone does not establish a current population-wide receipt rate or all historical changes in payment formulas. In a full country evaluation, claim data, contribution registration, rejected applications, and statutory amendments would have to be reconciled for the same period.
These cases demonstrate three dimensions of public responsibility: public law defines the entitlement, a public or private agent disburses cash, and public budgets or a contribution pool may ultimately bear the cost. A financing typology that ignores any one dimension is incomplete.
The arithmetic of contributory financing
For a pay-as-you-go maternity pool over a defined accounting period, the basic funding identity is:
Every item requires a consistent accounting period and cash or accrual basis. “Net financing” may include borrowing or reserve depletion, with the relevant sign convention stated; an institution's total revenue is not identical to contributions alone. The actuarial estimate of benefit cost can be constructed by applying expected claims and average benefit costs to distinct groups:
Group subscripts distinguish salary bands, employment types, demographic exposures, and legal benefit formulas. The expression is a cost-accounting decomposition, not a new statistical model. Actual claim frequency may correlate with wage histories, age, labor-force transitions and policy changes. Administrative cost and claims-processing lags require separate provisions. When a pool covers many contingencies, maternity expenditure must be disentangled from sickness, unemployment or general family payments to avoid double counting.
A useful stylized projection divides the expected number of compensated maternity episodes by active contributors and multiplies by the mean cash benefit per episode. The resulting implied contribution rate depends on the insured wage base. If the rate is applied to all employees' wages, the denominator is total covered payroll, not only the payroll of women expected to give birth. If a government finances the same outgo from general revenues, the incidence of its taxes depends on the full tax system and macroeconomic adjustments. Financial sustainability therefore cannot be inferred from the statutory name of a tax or contribution; it requires a projection of flows, administrative cost, reserves, compliance and distribution of who ultimately bears contributions.
Consider an expressly illustrative scenario, not an estimate for any country. A pooled scheme expects 10,000 paid episodes in a year at an average cash cost of 4,000 units per episode, giving 40 million units in annual benefit outgo before administration. If total covered payroll is 4 billion units, a 1 percent payroll levy would finance that expected benefit outgo in the simplified scenario. If only half of expected episodes are legally insured or paid, the apparent required levy may fall while substantive population coverage also falls. A financially balanced narrow scheme can therefore coexist with substantial social underprotection. Conversely, extending eligibility without financing the additional expected payments creates a cash-flow gap even when the enlargement advances universalism.
Employer liability, financing solidarity and labor-market effects
An individual firm that is required to pay the direct cash benefit experiences a concentrated expense when one of its workers gives birth. Small firms with few employees face substantial year-to-year variability relative to payroll and may be more sensitive to cash-flow pressure than diversified large employers. Such arrangements can create incentives for avoidance, substitution toward uncovered contractual categories, or pregnancy-related discrimination. These mechanisms are economically plausible, and the ILO financing comparison identifies employer liability as an obstacle to access and adequate protection in many jurisdictions (ILO, 2024). The precise size of any hiring effect should be established by credible firm-level or policy-reform evidence; it cannot be read directly from the classification of a scheme.
In the ILO's 2023-or-latest legal inventory, 157 of 217 countries and territories had periodic maternity cash benefits established in national law; 49 countries relied exclusively on employer liability, and 13 had no statutory periodic cash benefit. The report identifies social insurance in many other countries, including combinations with employer liability or means-tested assistance. These are counts of legal scheme types, not beneficiary-weighted effective coverage estimates (ILO, 2024, legislative scheme inventory). A country with very few births and one with millions of births have equal weight in a count of jurisdictions. In addition, categories such as social insurance plus employer liability describe hybrid legal mechanisms; they should not be re-added as disjoint categories without examining the classification.
Risk pooling addresses cost concentration but has additional consequences. Mandatory funding through a wide wage base removes the direct benefit expense from the firm experiencing the claim, while allowing statutory liability for ordinary payroll contributions. Some forms of insurance may still require women or self-employed claimants to opt in or pay substantial contributions; administrative contributions can be regressive when charged at a flat amount. Shared financing should be evaluated against its incidence across workers, employers, taxpayers, and family structures rather than assumed equitable by definition.
General revenue financing and the universality claim
Tax-financed cash benefits can protect mothers with limited contribution records, including people outside paid employment, and can therefore fill a structural gap left by conventional occupational insurance. The feasibility of this design depends on legal residence rules, payment adequacy, the taxation system, and the stability of budget transfers. A nominally universal flat benefit reduces variation across insured earnings but creates different replacement ratios for beneficiaries at different wage levels. An earnings-related benefit generally performs better at stabilizing previous consumption for covered workers, whereas a flat amount may be easier to extend to people with unrecorded earnings. The choice is a real policy tradeoff requiring explicit objectives and distributional analysis.
The Social Protection Floors Recommendation supplies a normative rationale for income security during maternity without making prior formal contributions the sole basis of protection (ILO, 2012). A basic floor may complement, rather than displace, contributory income replacement. A two-component financing package, for example, can fund a basic payment from general revenues and an additional earnings-related payment from an insurance pool. This is a familiar policy architecture rather than a new theoretical model. Its combined costs must be accounted for once, and interaction rules must explain whether the flat component is added, offset, or withdrawn as the insurance payment increases.
In countries with extensive informality, financing through a payroll base alone reaches a limited proportion of total labor. Moving some protection to general taxation can distribute costs across a wider revenue base, but that revenue may itself be vulnerable to fiscal volatility or indirect-tax regressivity. A complete incidence assessment incorporates both sides of the public budget: who pays, who receives, benefit take-up, and indirect effects through labor-market and consumption adjustments.
Financial resilience and demographic sensitivity
The number of births, maternal labor-force participation, contribution compliance, wage growth, payment formulas, inflation and employment shocks drive budget requirements. A recession may reduce payroll contributions at the same time that household finances become more fragile. The maternity contingency is not necessarily contemporaneously countercyclical in claim frequency, but the revenue base can contract independently of births. Inflation erodes fixed statutory weekly benefits unless adjustments are legislated or automatic. Replacement-rate ceilings can cause higher earners' effective replacement ratios to decline even where the headline percentage is unchanged.
Public administrators therefore need consistent projections under alternative assumptions for birth numbers, claims per birth, qualifying histories, average insured earnings, claims processing times, and inflation. A deterministic scenario table may be supplemented by actuarial reserve and stress testing where the financing institution actually holds capital. The choice of a capital framework must reflect whether obligations sit in a public budget, a contributory social-insurance fund, a private insurer, or an employer; it is misleading to apply the same solvency standard to each without examining its legal risk bearer.
A financing comparison
| Mechanism | Initial disburser | Ultimate funding source | Common reach constraint | Principal measurement risk |
|---|---|---|---|---|
| Direct employer liability | Employer | Individual employer | Formal employment and employer compliance | Mistaking an obligation for paid benefits |
| Employer-administered reimbursed pay | Employer | Public reimbursement plus unrecovered share | Employment, payroll and benefit qualification | Classifying gross payroll payment as private financing |
| Compulsory social insurance | Fund, public agency or delegated payer | Covered contributors; possible budget support | Insured attachment and qualifying histories | Ignoring unregistered and informal workers |
| Tax-financed flat or means-tested benefit | Public agency | General taxation or dedicated public revenue | Residency, means and administrative access | Treating eligibility as effective receipt |
| Hybrid protection | Multiple payers | Contributions, taxes and possibly employers | Coordination and interaction of conditions | Double-counting recipients and cash transfers |
These are analytical mechanisms, and countries may combine them. National system labels alone rarely specify who bears the marginal cost of an additional claimant.
Effective Take-Up
The measurement problem
The effective coverage question is whether a person receives the benefit when the contingency occurs. For a defined population and reporting period, the simplest population-coverage ratio is:
The calculation depends on the agency's definition of receipt, reference period, episode attribution, and demographic denominator. Multiple programs must be deduplicated at the person level. Claims paid after the reporting cut-off should not be silently omitted or included without stating the lag adjustment. When coverage is measured among legally eligible claimants rather than all women giving birth, it answers a different question: administrative take-up conditional on eligibility.
The ILO 2024–26 global report places global cash maternity coverage at 36.4 percent in 2023 and estimates approximately 85 million women with newborns were unprotected. Its regional discussion shows marked inequality: the global increase since 2015 was 6.8 percentage points, while the detailed maternity chapter reports effective coverage in 2023 of roughly 79.4 percent in Europe and Central Asia, 38.4 percent in Asia and the Pacific, and 6.7 percent in Africa (ILO, 2024). These figures are not direct indicators of the adequacy of any single payment; a claimant receiving a very small eligible cash transfer may count as covered.
The data show why the population denominator matters. An insurance program covering nearly every registered salaried employee could still have low maternity coverage if registered employees represent only a fraction of all women giving birth. Conversely, a tax-financed program can have high recipient counts but weak income replacement. Cross-country assessment must report coverage, adequacy and payment timeliness together.
A transparent decomposition of nonreceipt
A diagnostic accounting decomposition can express the number of people paid as the number in the maternity reference population multiplied by successive observed proportions: proportion legally within scope, proportion satisfying effective qualifying conditions, proportion filing a valid claim, proportion approved, and proportion actually paid in the specified period. This is a chain identity if each stage uses the preceding stage's denominator and each person is counted once. It offers a practical audit structure; it does not establish a causal law of take-up.
The distinction between stages changes policy diagnosis. If most excluded mothers work in categories outside legal scope, digitizing the claim portal cannot by itself close the main gap. If legal entitlement exists but contributors fail to qualify because of interrupted work histories, contribution-credit or eligibility reform becomes relevant. If eligibility is broad yet approved claims are systematically delayed or unpaid, budget execution and administrative procedures are immediate concerns. If beneficiaries receive on time but amounts are insufficient, benefit adequacy rather than take-up is the critical defect.
An analogous distinction applies to legal coverage: having one statutory scheme for a worker category is not proof that all members are enrolled. Enrollment denominators should be drawn from labor-force and household evidence, not exclusively from the administrative register that already omits uninsured people. Otherwise the research sample reproduces the exclusion that it intends to measure.
Unequal exposure across work arrangements
Informal and self-employed workers often lack stable insured earnings, formal payroll records, or the ability to fund voluntary contributions during income interruptions. Migrant and cross-border workers face residency and portability questions; domestic and agricultural workers face recording and enforcement obstacles. Unpaid caregivers may have no contribution record at all. Eligibility and payments must be analyzed with appropriate distinctions in the applicable legal system, rather than assuming that all such workers are excluded everywhere.
The ILO's Recommendation on transition from informality addresses the extension of social security and maternity protection to informal workers in both law and practice (ILO, 2015). Empirically relevant policy features include simplified enrollment, contribution subsidies, predictable benefits, low-cost mobile or in-person application, support for nonstandard wage evidence, and credible grievance procedures. Effectiveness should be assessed through observed take-up and changes in undercoverage rather than inferred from launch announcements.
The World Bank's Parenthood indicator design helps document the legal and institutional design for a standardized woman in the main business city, but its reference-case assumptions mean that a favorable score cannot substitute for field measurement among undocumented, migrant, self-employed or remote rural mothers (World Bank, 2026). Differences in population definitions account for some divergence between global legal indicators and maternity cash-benefit receipt statistics.
Benefit adequacy and the quality of coverage
Adequacy has at least three dimensions. Earnings-replacement adequacy compares cash benefits with prior earnings on an explicit gross or net basis. Subsistence adequacy compares the benefit with a relevant poverty or essential-consumption benchmark. Timing adequacy asks whether payments arrive during the income interruption, when they can actually sustain consumption and permit leave. Each measure can change independently.
The OECD's full-rate-equivalent comparison normalizes duration and mean payment rates by expressing paid leave as weeks paid at the equivalent of full previous earnings (OECD, 2026). For a fixed replacement rate, twenty weeks at 50 percent is ten full-rate-equivalent weeks. This is useful for comparing some entitlements but cannot recover the experience of workers who do not qualify, may mislead where ceilings and floors apply, and should not treat gross-based and net-based rates as identical. The OECD itself explains its comparison assumptions and the different underlying payment bases.
The United Kingdom illustrates within-program variation: its first six weeks of SMP use a percentage of historical earnings and the next thirty-three weeks involve a flat-rate ceiling, producing different effective replacement rates across wage levels (UK Government, 2026a, SMP payment rules). The Maternity Allowance schedule can grant up to thirty-nine weeks with different amounts for self-employed claimants depending on qualifying contributions (UK Government, 2026d). A comparative “39 weeks paid” headline describes duration but not the amount received by each employment group.
Likewise, Canada's maximum stated replacement rate of 55 percent applies to insurable earnings under the federal program and is constrained by a weekly benefit cap; it does not imply that all claimants receive 55 percent of their complete prebirth income (Government of Canada, 2026c, EI benefit calculation). International analysis requires effective benefit ratios for representative wages and family structures, measured separately for qualifying and nonqualifying populations.
Observed consequences and causal evidence
The welfare rationale for maternity payments includes the capacity to maintain basic consumption and to take time away from work during pregnancy and recovery from childbirth. Potential broader outcomes include infant health, parental mental and physical health, employment continuity, gender inequality and children's long-run development. These effects are empirically distinct. Evidence that leave duration improves a child outcome under one policy change does not establish that every method of financing produces the same result.
Carneiro, Løken and Salvanes (2015) studied Norway's 1977 transition from twelve weeks of unpaid maternity leave to four months of paid leave alongside an expanded unpaid entitlement. Their natural-experiment findings associate the reform with a two-percentage-point decline in high-school dropout and a five-percent increase in earnings at age thirty among the affected children, with stronger responses among mothers previously likely to take little unpaid leave. The design identifies consequences of this historical entitlement bundle under its institutional conditions. It does not isolate the contribution-financing mechanism from the expansion of paid time, and its average effects should not be mechanically extrapolated to contemporary low-income or informal labor markets.
Nandi and colleagues (2016) analyzed Demographic and Health Surveys covering approximately 300,000 births across twenty low- and middle-income countries between 2000 and 2008. Their difference-in-differences study reported an association of an additional month of paid maternity leave with 7.9 fewer infant deaths per 1,000 live births (95 percent confidence interval: 3.7 to 12.0), with much of the difference after the neonatal period. The authors recognized the possibility of unmeasured coincident reforms. This is important evidence for the potential health consequences of paid leave, but the treatment is a national policy entitlement and the outcomes do not directly measure who received cash, when it arrived, or which group financed it. Causal and distributional claims about universal maternity benefits require additional evidence on exposure and take-up.
A credible research synthesis should therefore separate five potential pathways: money replacing lost earnings; employment protection enabling time away; medical services accompanying childbirth; selection into formal insured work; and behavioral responses to the financing burden. Treating these mechanisms as interchangeable can falsely attribute the effect of one reform component to another.
The United States as a reminder about missing federal cash entitlements
In the United States, the federal Family and Medical Leave Act guidance describes up to twelve workweeks of unpaid, job-protected leave for eligible employees of covered employers, subject to statutory conditions (U.S. Department of Labor, 2026). The existence of this federal leave right does not mean that a national paid maternity cash-benefit system accompanies it. State paid family and medical leave programs and employer-sponsored paid leave create additional protections that must be studied by jurisdiction and date. A nationwide average should never confuse federal unpaid leave with universal federal cash coverage. For international comparisons, the United States is particularly instructive because paid and unpaid entitlements can be layered across federal law, state programs and workplace benefits, producing very different coverage experiences among otherwise similar families.
What observed coverage can and cannot establish
A rising share of paid mothers supports a claim of expanded effective access under the reporting definition. It does not, by itself, prove improved maternity health, reduced income poverty, a particular labor-market response or successful financing sustainability. Differences between 2015 and 2023 may reflect real program extensions, changes in household structure, data availability or estimation revisions. The ILO's methodological notes on global estimates explicitly warn that new estimates are not perfectly comparable with older report editions (ILO, 2024). Proper evaluation combines consistent coverage data with legal chronology, payment records, population denominators and independently evaluated outcomes.
Evidence Verification and Sensitivity
Research design and document hierarchy
This dossier uses comparative institutional analysis, explicit statutory comparison and critical empirical synthesis. It does not use an original claimant-level dataset. The hierarchy of evidence is issue-specific: original treaty provisions and current consolidated legislation govern the interpretation of legal rules; official administrative guidance explains operation but can lag amendments or simplify exceptions; international statistical releases provide harmonized descriptive estimates; peer-reviewed quasi-experimental articles inform carefully delimited causal questions; and the author's own algebra supplies transparent accounting implications. Each evidentiary category answers a different question. A technical page returning a functioning URL establishes accessibility, not automatically the legal validity or scientific adequacy of a claim.
The core legal sources were examined at their operative provision level, notably Articles 4 and 6 of Convention No. 183, Articles 49–52 of Convention No. 102, and Articles 8 and 11 of the EU pregnancy directive. Their legal scope, qualifying conditions, financing provisions and adequacy thresholds were distinguished from national practice. The dates of original adoption were retained rather than conflated with the date a website was crawled.
The descriptive evidence centers on the ILO World Social Protection Report 2024–26, whose population coverage measures refer principally to 2023, with stated exceptions and modelled estimates (ILO, 2024). The same report provides national legislative scheme counts based on 2023 or the latest available year. The OECD leave systems profile provides a complementary entitlement taxonomy and comparative payment-rate framework, with underlying policy reference dates specified in its tables and notes (OECD, 2026). These statistics are placed side by side for interpretation, never pooled into a synthetic count.
The case studies rely on the primary public sources shown in References. They provide verified examples of operative program design at the indicated snapshot, with country-specific restrictions. They do not amount to a complete audit of all national amendments, contributory legal disputes, subnational benefits, administrative determinations or court decisions. This restriction affects what can be inferred, and it is preserved throughout the comparative claims.
Numerical reconciliation and denominator audit
The increase in global maternity cash coverage from 29.6 percent in 2015 to 36.4 percent in 2023 is an arithmetic increase of 6.8 percentage points. It is not a 6.8 percent relative increase; the relative difference calculated against the 2015 base is approximately 23.0 percent. These two statements measure different changes. Similarly, if 36.4 percent of the reference population receives cash, the complementary fraction of 63.6 percent refers to nonreceipt under that same definition. It is not a poverty rate, an uninsured-labor-force rate or proof of noncompliance by every state.
The ILO's estimate of about 85 million women with newborns unprotected is a modelled population count attached to the same reporting framework, not a directly enumerated register of eighty-five million denied applications (ILO, 2024, global coverage summary). It should not be reverse-engineered into a precise total number of births without accounting for rounding, age definitions, source coverage and model assumptions. Regional percentages likewise refer to their region-specific population weighting, while legislative scheme counts refer to individual jurisdictions. These different denominators make an apparently exact arithmetic reconciliation inappropriate.
The figure of 157 jurisdictions with statutory periodic maternity cash benefits and the figure of 49 relying solely on employer liability answer distinct legal-classification questions. Neither should be divided by the number of births to infer effective protection. The ILO describes the underlying universe as 217 countries and territories and its classification as based on legal status in 2023 or the latest available information. Treating the labels as an exhaustive, strictly mutually exclusive partition without checking hybrid categories would introduce category errors (ILO, 2024, scheme type classification).
Measurement sensitivity
Several sensitivity analyses are essential to a future statistical comparison. First, define the reference population as all women giving birth in the jurisdiction and period, or as insured employment episodes, and show how the result changes. Second, separate the prevalence of qualifying insured employment from the conditional claim success rate. Third, test the effect of delayed payment and retroactive awards on the reported year. Fourth, distinguish nominal statutory weekly cash from inflation-adjusted benefit values, and mean from median receipt. Fifth, reweight national results using population, births, insured workers or total beneficiaries, specifying which inference each weighting supports.
In multinational comparisons, define maternity, parental, sickness and child benefits carefully. A country that supplies a childbirth-related payment through a general parental allowance may look unprotected if the researcher searches exclusively for the administrative name “maternity benefit”; the opposite error occurs when an undifferentiated family allowance is counted as maternity earnings replacement. The OECD distinction between maternity and parental leave demonstrates that international nomenclature is insufficient to infer functional equivalence (OECD, 2026). Program function, recipient, event, legal period and payment rules must all be coded.
Gross and net wage bases create additional comparison risks. A benefit replacing 70 percent of net income may offer a very different consumption level from one replacing 70 percent of gross earnings, especially when benefit tax treatment differs. Wage ceilings, minimum payment floors, waiting periods, benefit offsets and temporary employer wage top-ups further complicate effective replacement measurement. Studies should report at least an explicit wage scenario and apply consistent tax assumptions.
Cross-country causal analyses require attention to adoption and enforcement of policies, anticipation, concurrent childcare or health reforms, changes in formal employment, and survey identification of benefit recipients. The Norwegian reform evaluation and the multicountry infant mortality analysis offer more informative research designs than cross-sectional correlation, but identify effects for their specific reforms and outcome populations. Unmeasured contemporaneous interventions and treatment uptake remain relevant threats to transferring effect sizes to other settings.
Source-to-claim discipline and reproducibility
Every linked external source in the scientific chapters is associated with a bibliographic entry below and a source-specific checking record in the companion EVIDENCE.json. The file distinguishes institutional legal authority from policy overviews and empirical research. The manuscript uses short contextual links attached to the cited proposition; internal evidence identifiers are reserved for the machine-readable ledger. The full source ledger should be treated as a navigation and verification aid, not a substitute for checking a changing law or administrative rate on the date of a later publication.
The manuscript's evidence scope is intentionally explicit. No new interviews, original population microdata analysis, actuarial model calibration, or independent national claims audit was undertaken. There is no invented regression coefficient, cross-country benefit total, unnamed statute, or universal claim about national legal applicability. The analytic identities and stylized financing example can be reproduced directly from their stated inputs. A later policy-specific release must revalidate mutable rates and legal provisions in the relevant national context if its purpose includes giving up-to-the-minute operational guidance. These are normal time limits of secondary documentary research, not unfinished claims about the historical and analytical findings made here.
Scientific Synthesis and Research Agenda
Universality is a relationship between financing and reach
The evidence establishes that broad financing and broad entitlement are separate institutional decisions. Collective financing can distribute maternity cash costs away from individual employers, lowering concentration of the expense and facilitating payment continuity. Yet if benefit entitlement remains conditional on recognized wage employment or sufficient insured hours, the uninsured mother still receives nothing from that contributory scheme. Conversely, a tax-financed floor may extend protection to women outside formal work while providing a smaller amount than earnings-related insurance. The system-wide question is therefore how the eligibility architecture interacts with the distribution of the financing burden.
The international instruments suggest a coherent interpretation of this relationship. Convention No. 183 states minimum maternity cash-benefit protections for the workers within its scope and favors collectively financed provision, subject to specified exceptions. Recommendation No. 202 extends the normative concern toward basic income security throughout the population. Recommendation No. 204 addresses the transition from informal work and adaptation to limited contribution capacity. Their combined treatment in the ILO report makes the institutional connection visible while preserving the legal distinction among binding conventions and recommendations (ILO, 2024).
From the evidence, an operationally meaningful account of universal maternity cash benefits requires a definition of the target population, a legally enforceable entitlement route for members of that population, financing capable of sustaining timely payments, and administration that makes the entitlement reachable. The descriptive strength of this conclusion rests on institutional comparison and coverage measurement. It is a synthesis of established concepts, not a newly named theoretical architecture.
Financing should be evaluated against both incidence and benefit distribution
A contribution paid by an employer is an accounting charge, but its long-run economic burden may be shared through wage adjustments, employment choices, prices and returns to capital. General taxes likewise distribute economic burdens differently depending on their tax base. Without an incidence study, “employer-financed” and “tax-financed” remain legal and fiscal classifications. A comparative assessment should pair each with observable policy features: contribution base, rates, subsidy structure, benefit floors and ceilings, employer reimbursement, fiscal transfers and rights of appeal.
Distributional analysis should cover at least the following groups: employees with continuous insured histories; employees with interrupted or casual histories; self-employed and informal earners; people without earned income; migrants and cross-border workers; and beneficiaries at different wage levels. Each group should be evaluated for entitlement, realized receipt and replacement adequacy. A single national average can otherwise conceal that pooled financing largely protects relatively stable earners while highly vulnerable mothers remain without income during childbirth.
The case comparisons reveal concrete mechanisms. Canada's federal EI framework shows how a benefit attached to insured hours can provide substantial temporary wage replacement for some workers but requires close analysis of qualification and self-employment participation. The United Kingdom's public reimbursement arrangement demonstrates that employer payroll disbursement can operate alongside extensive public cost sharing. Brazil's INSS example shows the complexity of insured categories and claimant pathways. New Zealand demonstrates a government-funded parental payment encompassing both employees and independent earners under stated work conditions. South Africa shows a maternity channel in a broader insurance fund, while the United States demonstrates that job protection at one level of law can exist without a federal paid cash entitlement. These comparisons support the general analytic requirement to disaggregate eligibility, payment administration and financing responsibility.
Three empirical questions for the next research cycle
The first question concerns additionality of access: when a maternity cash system expands from formal employees to informal or independent workers, how many additional women actually receive payments, and at what fiscal cost per additional recipient? A defensible evaluation needs a clear pre-reform population denominator, administrative payments reconciled to vital statistics, and a comparison or interrupted-time-series strategy capable of distinguishing the extension from secular changes in births and employment.
The second concerns the financing transition: does moving the direct cash cost from employers to a social-insurance pool or public reimbursement arrangement change women's hiring, employment stability, compliance and maternity benefit receipt? The causal unit could be the employer, worker or birth cohort depending on policy implementation. Reforms phased by firm size, industry or contribution thresholds may provide evaluation opportunities, subject to pre-trend, spillover and policy-selection tests. The hypothesis is worth examining because accounting incentives change; the direction and magnitude of behavioral responses must remain empirical findings rather than assumed outputs.
The third concerns adequacy under discontinuous earnings: what mix of basic income protection and earnings-related insurance most effectively preserves consumption around birth for people whose labor-market attachments are irregular? The relevant outcomes include realized after-tax cash, maternal and child well-being, labor-force attachment and the public financing burden. Experimental or quasi-experimental methods and administrative-household record linkage could test these questions, with privacy and consent safeguards. Their results would help evaluate established policy mechanisms in context.
An evidence-grounded comparative research specification
For each selected country and year, the proposed observational record includes: legal event and eligible claimant categories; first effective date and amendments; employment and contribution thresholds; payment duration; cash formula and cap; source of contributions; amount of public budget transfer; employer gross payments and reimbursements; fund financial accounts; claim applications, approvals, rejections and late payments; female births and relevant employment denominators; inflation and wage measures; and appeal outcomes. Each variable should carry a date, source link, definition, coverage note and quality flag. The ILO source inventory and indicator notes offer an existing foundation for such reconciliation (ILO, 2024).
A comparative research program could begin with representative institutional types rather than attempting a universal country ranking. It would test whether apparent differences survive adjustments for childbirth denominators, financing classifications, benefit overlap and incomplete registration. The resulting evidence could inform practical policy analysis while retaining the legal and fiscal individuality of each jurisdiction.
Institutional implications
For policymakers, the research implies that expanding the phrase “covered worker” in legislation is only the first part of expansion. Financing appropriations, insurer eligibility records, claims administration, information access and appeals must align with the new scope. For insurance administrators, a claim paid once should be linked to its financing classification and reporting period so that pooled risk, reimbursement and coverage can be reconciled. For statistical offices, counts of eligible contributors, paid claims and women giving birth should be separately available. For employers, clear reimbursement arrangements reduce uncertainty about retained cost and administrative responsibility. For researchers, distinguishing a legal entitlement from its realized value is the precondition for meaningful causal and distributional inference.
Maternity cash benefit financing belongs at the intersection of social insurance, public economics and labor-market institutions. Analysis that integrates those disciplines while keeping their distinct measures visible offers a clearer account of what universality means in practice.
Conclusion
Maternity cash benefits address an identifiable interruption of income connected to childbirth and related leave. Their public purpose is realized through a combination of substantive entitlement, reliable payment, adequate support and sustainable financing. The legal history from ILO Convention No. 102 through Convention No. 183 and later social protection recommendations establishes the importance of collective financial responsibility, benefit adequacy and the expansion of protection beyond conventional formal employment. Domestic implementation remains institution-specific.
The ILO's 2023 global effective coverage estimate of 36.4 percent documents the scale of nonreceipt among women with newborns. Its meaning is precise: a population-weighted estimate of cash-benefit receipt under a defined social-protection indicator, with important methodological and country-source limitations. The statistic establishes a global access gap; it does not independently measure the adequacy of every payment or the causal effect of any financing scheme.
The comparative cases demonstrate that final financing and initial payment may be different, that multiple cash-benefit pathways may coexist within a jurisdiction, and that eligibility conditions can limit access even where benefits are collectively funded. Employer liability, insured payroll contributions, tax-financed transfers and mixed reimbursement systems create different fiscal and labor-market exposure profiles. Their performance should be judged through reproducible measures of receipt, value, timing, administrative cost and economic incidence.
The principal scientific conclusion is that the universality of maternity income protection must be evaluated jointly across legal scope, effective access, benefit adequacy and financing capacity. Each dimension can be assessed with existing legal, administrative, actuarial and statistical methods. Future empirical research can determine which reforms change each dimension and for whom, while maintaining a strict distinction among observed outcomes, documentary legal rights and causal estimates. This monograph supplies that source-grounded comparative synthesis for the Research Dossier and preserves the unresolved empirical questions as research questions rather than claims of completed new theory.
References
Carneiro, Pedro, Katrine V. Løken, and Kjell G. Salvanes. 2015. “A Flying Start? Maternity Leave Benefits and Long-Run Outcomes of Children.” Journal of Political Economy 123 (2): 365–412. Publisher and DOI.
Council of the European Communities. 1992. Council Directive 92/85/EEC of 19 October 1992 on the Safety and Health at Work of Pregnant Workers and Workers Who Have Recently Given Birth or Are Breastfeeding. Official Journal of the European Communities L 348. Legal text.
Employment New Zealand. 2026. “Parental Leave Payments.” New Zealand Government, online administrative guidance, consulted November 2026. Program guidance.
Government of Canada. 2026a. “EI Maternity and Parental Benefits: Eligibility.” Employment and Social Development Canada, online program guidance, consulted November 2026. Eligibility requirements.
Government of Canada. 2026b. “Benefits for Self-Employed People.” Employment and Social Development Canada, online program guidance, consulted November 2026. Self-employment coverage.
Government of Canada. 2026c. “EI Maternity and Parental Benefits: How Much You Could Receive.” Employment and Social Development Canada, 2026 rate and calculation guidance. Benefit calculation.
Government of Canada. 2026d. “EI Maternity and Parental Benefits: What These Benefits Offer.” Employment and Social Development Canada, 2026 program summary. Payment schedule.
Instituto Nacional do Seguro Social (INSS). 2025. “Salário-Maternidade.” Brazilian federal social security agency, official guidance updated July 31, 2025. Benefit overview.
International Labour Organization (ILO). 1952a. Social Security (Minimum Standards) Convention, 1952 (No. 102). Geneva: ILO, especially Part VIII, Articles 46–52. Convention text.
International Labour Organization (ILO). 1952b. Maternity Protection Convention (Revised), 1952 (No. 103). Geneva: ILO, especially Article 4. Convention text.
International Labour Organization (ILO). 2000a. Maternity Protection Convention, 2000 (No. 183). Geneva: ILO, especially Articles 4, 6 and 7. Convention text.
International Labour Organization (ILO). 2000b. Maternity Protection Recommendation, 2000 (No. 191). Geneva: ILO. Recommendation text.
International Labour Organization (ILO). 2012. Social Protection Floors Recommendation, 2012 (No. 202). Geneva: ILO, official institutional summary and principles. Recommendation summary.
International Labour Organization (ILO). 2015. Transition from the Informal to the Formal Economy Recommendation, 2015 (No. 204). Geneva: ILO, especially paragraphs 18–20. Recommendation text.
International Labour Organization (ILO). 2024a. World Social Protection Report 2024–26: Universal Social Protection for Climate Action and a Just Transition. Geneva: ILO, chapter 4, especially maternity cash benefits and financing. Maternity chapter.
International Labour Organization (ILO). 2024b. World Social Protection Report 2024–26. Executive Summary: coverage results, definitions, estimated unprotected population. Executive Summary.
International Labour Organization (ILO). 2024c. World Social Protection Report 2024–26. Annex 2 and maternity protection standards comparison. Standards annex.
Nandi, Arijit, Mohammad Hajizadeh, Sam Harper, Alissa Koski, Erin C. Strumpf, and Jody Heymann. 2016. “Increased Duration of Paid Maternity Leave Lowers Infant Mortality in Low- and Middle-Income Countries: A Quasi-Experimental Study.” PLOS Medicine 13 (3): e1001985. Article and DOI.
Organisation for Economic Co-operation and Development (OECD). 2026. “PF2.1. Parental Leave Systems.” OECD Family Database, updated 2026 edition; tables identify underlying policy reference dates. Indicator and methodology.
South African Department of Employment and Labour. 2013. “Basic Guide to UIF Maternity Benefits.” Official Unemployment Insurance Fund claimant guidance, page last modified February 12, 2013; consulted November 2026. UIF maternity guide.
UK Government. 2026a. “Maternity Pay and Leave: Pay.” GOV.UK, 2026–27 statutory payment guidance. SMP rules.
UK Government. 2026b. “Maternity Allowance: Overview.” GOV.UK, consulted November 2026. Allowance eligibility pathway.
UK Government. 2026c. “Get Financial Help with Statutory Pay: What You Can Reclaim.” GOV.UK, 2026–27 employer reimbursement rules. Employer recovery.
UK Government. 2026d. “Maternity Allowance: What You'll Get.” GOV.UK, 2026–27 payment guidance. Allowance payment amounts.
United States Department of Labor. 2026. “FMLA Frequently Asked Questions.” Wage and Hour Division, federal unpaid-leave guidance, consulted November 2026. Federal leave rules.
World Bank. 2026. “Women, Business and the Law: Methodology.” Parenthood topic, legal and supportive framework indicators, online methodology consulted November 2026. Parenthood methodology.