Cash for Mothers: Ecuador’s Flagship Anti-Poverty Programme Shows Only Modest Links to Maternal Survival

Every two minutes, somewhere in the world, a woman dies from complications of pregnancy or childbirth. Global maternal mortality has fallen by roughly forty percent since 2000, yet that progress has stalled, and Latin America and the Caribbean stand out as one of the few regions where deaths have recently been rising rather than falling. Against this troubling backdrop, a new fourteen-year study from Ecuador offers a sobering assessment of one of the region’s most popular policy tools: the conditional cash transfer. The research, published in SSM – Population Health, examined whether Ecuador’s flagship anti-poverty programme, the Bono de Desarrollo Humano, was associated with reductions in maternal mortality across the country’s municipalities between 2010 and 2023. The answer, in short, is that the programme reached the right places but delivered far less than many had hoped.

The Bono de Desarrollo Humano, or BDH, sits at the heart of Ecuador’s social protection system. Launched in 2003 as a successor to an earlier unconditional transfer, the programme provides monthly cash payments to low-income households, primarily mothers, on the condition that they attend preventive health visits and keep their children in school. Payments of fifty US dollars per month, rising to as much as one hundred and fifty dollars depending on the number of minors in the household, are distributed in cash through more than nine thousand authorised points across the country. At its peak, the programme reached over one million families and accounted for nearly 0.7 percent of Ecuador’s gross domestic product. In theory, the conditions attached to the money should push pregnant women toward at least five antenatal check-ups, while the income itself should ease the financial barriers that keep poor women from seeking care.

To test whether this theory holds in practice, researchers led by Sara Toros and Ana L. Moncayo assembled a panel dataset covering Ecuador’s 221 cantons, the second-level administrative divisions of the country, from January 2010 to December 2023. Because maternal deaths are rare events and vital registration quality varies across Ecuador, the team restricted their primary analysis to the 147 cantons with intermediate or high-quality birth and death registration, a classification previously validated using indicators such as registration completeness and the proportion of ill-defined deaths. Maternal deaths were identified according to World Health Organization criteria and the tenth revision of the International Classification of Diseases, capturing deaths during pregnancy or within forty-two days of its termination from pregnancy-related causes. The outcome measure was the maternal mortality ratio, expressed as maternal deaths per 100,000 live births, calculated annually for each canton.

The statistical approach was deliberately conservative. The researchers used negative binomial regression models for panel data, a choice driven by the over-dispersed distribution of maternal deaths across canton-years, and included fixed effects for both canton and year. Canton fixed effects absorb all time-invariant characteristics of each municipality, such as geography, culture and historical legacies, while year fixed effects account for national shocks affecting the whole country simultaneously. The exposure was measured in two ways: coverage of the eligible population, capturing how many qualifying families were actually enrolled, and coverage of the total canton population, which reflects the broader penetration of the programme within a community. Models were adjusted for poverty measured by the multidimensional Unsatisfied Basic Needs Index, adult illiteracy, physician density and hospital bed availability, and the team ran an extensive battery of sensitivity analyses to probe the robustness of their findings.

The descriptive picture is striking. Across the analysed cantons, the mean maternal mortality ratio almost halved over the study period, falling from 41.4 to 21.4 deaths per 100,000 live births. Yet the BDH moved in the opposite direction: coverage of the total population dropped from 42.3 percent to 24.1 percent, and coverage of the eligible population fell from 60.5 percent to 49.1 percent, largely because a 2013 update restricted eligibility to households in extreme poverty, cutting the number of recipients by about fifty-seven percent. Meanwhile, poverty, illiteracy and physician density all improved substantially, with physician density more than doubling from 9.6 to 22.5 per 10,000 inhabitants. Disentangling the effect of a shrinking programme from these concurrent social changes is precisely the challenge the fixed-effects design was built to address.

The headline result is a near-null association. In models adjusted for socioeconomic and health-system factors, each percentage-point increase in BDH coverage was associated with only a one percent reduction in the maternal mortality ratio, with confidence intervals that grazed the null value. Stratified analyses revealed little evidence of effect modification across poverty quintiles or dominant ethnic groups, though the strongest inverse associations appeared in cantons in the fourth poverty quintile and in Afro-Ecuadorian-majority cantons, where coverage of the total population was associated with a three percent lower mortality ratio. Urban cantons showed a modest protective association, while rural and mixed cantons showed none. In other words, where the programme seemed to matter most was where health services were actually available to absorb the additional demand for care.

Two contextual findings add nuance. After the 2013-14 eligibility update, a one-percentage-point increase in total-population coverage was associated with a three percent reduction in maternal mortality, a significantly stronger association than before the reform, and formal interaction tests supported this difference. Even more striking, during the COVID-19 pandemic years of 2020 and 2021, when Ecuadorian health services buckled and prenatal, delivery and postnatal care fell by forty-six, twenty-eight and thirty-eight percent respectively, higher BDH coverage was associated with two to three percent lower maternal mortality for both coverage measures. This suggests that cash transfers may act as a buffer during crises, cushioning households against the catastrophic income shocks and care disruptions that the pandemic imposed, even though the programme alone was insufficient to prevent an overall rise in maternal deaths during that period.

Why did Ecuador’s programme deliver so much less for maternal survival than its celebrated counterparts in Brazil and Mexico? The authors point to a crucial difference in implementation. Brazil’s Bolsa Família operates alongside the Family Health Strategy, a consolidated primary care system that actively monitors health conditionalities, including antenatal attendance, and has shown synergistic effects with the cash transfers. Mexico’s Prospera generated sustained demand that was matched by improvements in the supply of health services. In Ecuador, by contrast, compliance with the BDH’s conditionalities has never been systematically monitored, and sanctions for non-compliance have rarely been applied. The requirement of five antenatal check-ups, the very mechanism designed to detect hypertensive disorders, infections and other leading causes of maternal death, exists largely on paper. Without enforcement or linkage to a strong primary care network, the demand-side pathway from cash to care quietly breaks down.

The study has limitations that the authors acknowledge candidly. The ecological design measures exposure at the canton level and cannot capture individual or household-level mechanisms, although it is well suited to estimating population-level effects, including potential spillovers onto non-beneficiaries. Maternal deaths are recorded by place of occurrence rather than residence, which likely underestimates mortality in rural, high-coverage cantons and overestimates it in better-resourced referral centres, a non-differential misclassification that would bias estimates toward the null. Residual confounding from time-varying unmeasured factors, such as health reforms or concurrent social programmes, cannot be excluded, and ethnicity and urbanicity were treated as time-invariant proxies derived from the 2022 census. Yet sensitivity analyses, including province-level models, restriction to cantons with hospital beds and alternative census years, produced consistent results, lending credibility to the central conclusion.

The broader lesson extends well beyond Ecuador. Conditional cash transfers have become the default social protection instrument across Latin America and much of the developing world, and their documented successes in boosting school enrolment and service utilisation are real. But this study adds to a growing body of evidence that cash alone is unlikely to substantially reduce maternal deaths, a rare and multifactorial outcome shaped by the quality of obstetric care, emergency referral systems, skilled birth attendance and deep-rooted structural inequities that no monthly payment can dissolve. Among the 147 cantons analysed, forty-four had no hospital beds throughout the entire study period, and specialists remained concentrated in Quito, Guayaquil and Cuenca. The authors argue that the BDH should be seen not as a failed intervention but as a valuable platform, one that could be strengthened by enforcing health conditionalities, integrating culturally sensitive community-led services for Indigenous and Afro-Ecuadorian women, and pairing income support with sustained investment in equitable, high-quality maternal healthcare. Cash, in other words, is a beginning, not a cure.

Subject of Research: The association between Ecuador's conditional cash transfer programme and maternal mortality

Article Title: Long-term association of a conditional cash transfer programme and maternal mortality in Ecuador: An ecological longitudinal study, 2010-2023

Article References: Long-term association of a conditional cash transfer programme and maternal mortality in Ecuador: An ecological longitudinal study, 2010-2023. (n.d.). https://doi.org/10.1016/j.ssmph.2026.101963

Image Credits: AI Generated

DOI: 10.1016/j.ssmph.2026.101963

Keywords: maternal mortality, conditional cash transfers, Ecuador, Bono de Desarrollo Humano, social protection, public health, health inequalities, antenatal care, fixed-effects regression, COVID-19, Latin America, health policy