In 2000, the Brazilian municipality of Sobral had a problem that seemed unsolvable. Nestled in Ceará, one of Brazil’s poorest states, only 49% of second-graders could read at grade level.1 By 2004, that number had reached 92%.1 Today, Ceará has the lowest learning poverty rate in Brazil, with 10 of the country’s top 20 performing municipalities.1
Sobral’s transformation wasn’t magic. It was method: structured teaching materials, intensive teacher support, and results-based financing that tied 18% of tax transfers to educational outcomes.1 The approach spread across the state, proving that even the most disadvantaged communities can achieve what wealthy nations often struggle to deliver.
We open with Sobral because this story of evidence-based intervention producing dramatic results is being replicated across the developing world. In Kenya, literacy rates nearly doubled after a national reading program reached 23,000 schools.2 In India, a simple approach of grouping children by skill level rather than age has reached 76 million students with some of the largest learning gains ever measured in education research.3
These success stories matter because they illuminate a path through one of the most significant, and solvable, challenges in human development today.
The Gap Behind the Classroom Door
Here’s a number that should reshape how we think about global education: seven in ten children in low- and middle-income countries cannot read and understand a simple text by age 10.45 The World Bank calls this “learning poverty,” and it represents something profound: the gap between getting children into schools and actually teaching them to read.
This isn’t about access anymore. Decades of global effort successfully expanded enrollment, and most children now have a seat in a classroom. The challenge is what happens once they’re there. We’ve achieved schooling without learning, and the consequences ripple across entire societies.
The numbers vary dramatically by region, but the pattern is consistent. In Sub-Saharan Africa, 89% of children experience learning poverty: nine in ten cannot read at age 10.6 Latin America saw rates jump from 52% to an estimated 80% following pandemic school closures averaging 225 days.4 South Asia, with the world’s longest closures at 273 days, moved from 60% to 78%.4
When we look at root causes, three factors emerge repeatedly across contexts.
Teachers are stretched impossibly thin. UNESCO projects the world needs 44 million additional teachers by 2030, including 15-17 million in Sub-Saharan Africa alone.7 The funding required reaches $120 billion, against current spending of just $55 per student annually in low-income countries versus $8,532 in wealthy nations.8 That’s a 155-fold gap in investment per child.
Children learn in languages they don’t speak. Between 37-40% of students in developing countries receive instruction in languages different from what they speak at home, rising to 90% in some contexts.9 In Peru, native Spanish speakers are seven times more likely to achieve satisfactory reading than indigenous students learning in Spanish as a second language.9
Traditional teaching methods fail foundational literacy. Teacher-centered instruction dominates despite evidence of poor outcomes. Curricula assume knowledge children don’t possess. Many teachers lack training in evidence-based reading instruction and receive no ongoing coaching or support.10
What’s at Stake, and Why It’s Worth Solving
The economic scale is significant. The World Bank’s most comprehensive estimate values learning poverty at $21 trillion in lost lifetime earnings for the current generation, equivalent to 17% of global GDP.114 Flip this around: solving it represents one of the largest opportunities in human development. For Africa specifically, closing the learning gap could unlock an estimated $6.5 trillion in economic opportunities.6
But beyond economics, this is about human potential. Under the UN Convention on the Rights of the Child, every child has a right not just to education, but to education that develops their capabilities.12 SDG 4’s framing as “quality education” explicitly recognizes this, and the good news is that proven solutions exist to achieve it.
The intergenerational dimension makes action especially valuable. UNESCO estimates 171 million people could be lifted from poverty if all students in low-income countries achieved basic reading skills.12 Foundational literacy opens doors to everything else: the technical skills modern economies need, the agency to participate in civic life, the capacity to break cycles of disadvantage.
The Interventions That Actually Work
What gives us hope is that we now have robust evidence for what works, and it’s being implemented at scale. The solutions share common features: they focus on foundational skills, support teachers with practical tools, and adapt to local contexts while maintaining evidence-based principles.
Structured Pedagogy: The Strongest Evidence Base
Structured pedagogy programs provide teachers with detailed lesson guides, student workbooks, intensive training, and ongoing coaching support. The Global Education Evidence Advisory Panel classifies these as a “Great Buy” based on exceptional cost-effectiveness.3
The results are striking. Across developing countries, structured pedagogy produces average improvements of 0.44 standard deviations, twice the effect size of similar programs in the United States.10 Kenya’s Tusome (“Let’s Read”) program began with randomized trials in 400+ schools finding students were three times more likely to meet national benchmarks.2 Within two years, it scaled to 23,000 public primary schools with literacy rates nearly doubling.2
Analysis found each additional $100 in spending produced 15 more students reaching benchmarks, an exceptional return on investment.2
Teaching at the Right Level: Meeting Children Where They Are
India’s Pratham NGO developed an elegantly simple insight: group children by actual skill level, not age. A child who can’t recognize letters needs different instruction than one who can decode words, regardless of what grade they’re enrolled in.
Six randomized trials documented effects that J-PAL describes as “some of the largest rigorously measured in education literature.”3 In Uttar Pradesh, children reading paragraphs or stories doubled.3 The Teaching at the Right Level (TaRL) approach has now reached 76 million Indian students through government partnerships and expanded to 20+ countries.3
Mother-Tongue Instruction: Building on What Children Know
UNESCO’s 2025 data confirms what cognitive science predicts: children taught in their home language are 30% more likely to read with understanding by end of primary school.9
Counter-intuitively, this extends to second-language acquisition too. Mali’s Pédagogie Convergente found students in mother-tongue schools actually performed better in French than those taught only in French.9 Strong foundations in a first language transfer to second-language learning. The World Bank now recommends at least six years of mother-tongue instruction before transitioning.9
Early Childhood Investment: The Highest Long-Term Returns
The earlier we intervene, the greater the impact. Jamaica’s home visiting program produced 37% higher earnings at age 31 for participating children.13 Meta-analyses show quality early childhood education reduces special education placement by 8.1 percentage points, grade retention by 8.3 points, and increases high school graduation by 11.4 points.13
In Sub-Saharan Africa, every dollar invested in tripling pre-primary enrollment could generate $33 in returns, exceeding virtually any alternative investment.6
School Feeding: Addressing Hunger to Enable Learning
Hungry children cannot learn effectively. With 200 million children under five affected by poor nutrition, the cognitive foundations for learning are often compromised before schooling begins.14 School feeding programs address this directly.
Systematic reviews document 5-6 percentage point increases in girls’ enrollment and higher attendance rates.14 A Kenya study found students receiving meals with meat improved 57.5 points across subjects compared to controls receiving no food.14
Scaling What Works
The 2024 Africa Foundational Learning Exchange brought delegates from 39 countries to commit to achieving zero learning poverty by 2035.6 It’s an ambitious target, but the October 2025 Global Education Evidence Advisory Panel report, synthesizing approximately 120 studies across 170+ languages, confirms we know what effective reading instruction looks like.10
Countries succeeding in reducing learning poverty share common features: sustained political commitment, use of existing government structures for scale-up, results-based financing, continuous monitoring, and investment in teacher support.12 These aren’t mysterious ingredients; they’re implementation discipline applied to proven interventions.
The main constraint is funding. The $97 billion annual gap between what’s needed and what’s available cannot be closed through domestic resources alone in the poorest countries.8 Yet education aid fell 7% between 2020 and 2021, with Sub-Saharan Africa experiencing a 23% decline.8 African governments now spend more on debt servicing than on education and healthcare combined, a structural barrier that needs international attention alongside domestic commitment.8
The Path Forward
Learning poverty represents a fundamental gap in what Doughnut Economics calls the social foundation: children without the basic capability to decode written language, which cascades into every dimension of human flourishing.
But unlike many global challenges, this one has proven solutions. Sobral’s transformation from 49% to 92% literacy in four years wasn’t an anomaly; it was a template. Kenya scaled evidence-based reading instruction to 23,000 schools. India reached 76 million children with targeted instruction. These aren’t pilot programs anymore; they’re proof of concept at national scale.
The research tells us that each additional year of quality schooling generates 9-10% higher earnings.11 Every dollar invested in early childhood education can return $33.6 Structured pedagogy delivers twice the learning gains at a fraction of the cost of interventions in wealthy countries.10
What remains is deploying what we know works, at the scale the opportunity demands. The 800 million children currently learning to read aren’t waiting for new innovations. They’re waiting for the political will and coordinated investment to bring proven solutions to every classroom.
Sobral, Kenya, and India proved it’s achievable. The research shows us how. The question now is whether we’ll act on what we’ve learned, and the evidence suggests we absolutely can.