Too many children cannot read by the age of ten. Around two out of three of children in low- and middle-income countries cannot read and understand a simple, age-appropriate text. This is a tragedy, and it is also unnecessary. We know how to solve it.
The causes of the learning crisis are many and familiar. Education systems are chronically underfunded. Teachers are underpaid and undertrained. Schools lack books, basic infrastructure, and instructional time. Many children are taught in a language they do not speak at home. All of this matters. But one cause stands out, and it is too rarely named: in many countries, teachers are simply not using the right methods to teach children to read. Classrooms still rely on choral repetition, and whole-language approaches that decades of research have shown to be ineffective. Fixing this will not solve everything. But without fixing it, nothing else will be enough.
Reading is not natural. Unlike spoken language, which children pick up at home simply by being exposed to it, reading is a recent cultural invention—only a few thousand years old—and the human brain is not pre-wired for it. Learning to read requires rewiring neural circuits originally dedicated to other functions so that they can connect visual symbols, the letters, with the sounds of speech, and from there with meaning. Think for a moment about what you are doing right now: distinguishing small marks of ink or pixels, mapping each one to a sound, combining those sounds into a word, and then making sense of the word in a sentence. It appears effortless because you learned to do it long ago and have practiced thousands of times. But it had to be taught systematically. And it cannot be left to children to figure out on their own.
This is what “systematic” means. Skills are taught in a logical sequence. We start with the simplest elements: the sounds of the language, the correspondence between letters and sounds. Then we move, step by step, to syllables, words, sentences, and longer texts. Each step builds on the previous one, and no essential skill is left to chance. This is very different from methods that expect children to “discover” how reading works by being exposed to books, or that teach phonics in a scattered, incidental way. Those approaches fail most children, and they fail the poorest children the most—the children who arrive at school without having been read to at home, without books, sometimes without having heard the language of instruction. For them, the absence of systematic teaching is decisive.
Systematic instruction must also be explicit: the teacher demonstrates and explains each skill clearly before students practice it. And it must be comprehensive, covering six core skills that the evidence has identified: mastery of the oral language, phonological awareness, systematic phonics, reading fluency, reading comprehension, and writing. Each skill reinforces the other. A weakness in any of these areas will hold a child back. Teaching only phonics is not enough. Fostering a love of books is not enough. All six skills are needed.
For decades, the strongest evidence on how to teach reading came from the Global North. In the United States, in the year 2000, the National Reading Panel report reviewed hundreds of experimental studies and concluded that explicit, systematic phonics instruction worked better than whole-language methods. In the United Kingdom, the Rose Report of 2006 reached the same conclusion, and led the English government to embed systematic phonics in the national curriculum. In Australia, the 2005 Teaching Reading report reached the same verdict. And in France in 2006 the Ministry of Education published Apprendre à lire confirmed these principles in a different language and a different pedagogical tradition. More recently, the Eurydice Report on Teaching Reading of the European Commission showed that the same findings hold across diverse European linguistic contexts. Four governments, four independent reviews, four times the same answer.
For the debate in low- and middle-income countries a reasonable objection used to be that most of this evidence came from English and from rich countries. Did the conclusions hold in different languages, in classrooms with fewer resources? A recent report that I coauthored with several colleagues, experts in reading in different languages, and endorsed by the Global Education Evidence Advisory Panel answers that question. Drawing on 151 studies covering more languages across Africa, Asia, Latin America, and the Middle East, the answer is clear. Of the rigorous evaluations, 84% show positive effects on learning. Comparative studies in Guatemala, Malaysia, Thailand, and Zambia confirm that students taught with systematic phonics outperform those taught with whole-language methods. The Learning at Scale study found that the eight highest-performing large-scale reading programs in low- and middle-income countries all share one feature: they apply the principles of the science of reading. The evidence from the Global South clearly confirms what works regarding teaching methods.
Knowing what and how to teach is one thing. Getting it into every classroom, for every child, is quite another. In many countries, the obstacle is no longer pedagogical. The methods are known, and increasingly the materials are too. The obstacles are logistical and managerial. Making sure that a million teachers in tens of thousands of schools have, on the right day, the right teacher guide, the right student textbooks, the right decodable readers in the right language, and the training and coaching to use them—this is an operational challenge of extraordinary complexity. It requires good procurement systems, reliable distribution chains and accurate school-level data. Many education ministries in low- and middle-income countries struggle in these processes. Many structured pedagogy programs have faltered not because the science was wrong, but because the inputs never reached the classroom on time, or in sufficient quantity, or in usable form.
The managerial challenge at the school level is just as demanding. Every school must be organized so that every teacher receives the support needed to change practice—and changing systematically how teachers teach every day is hard for any system. Requires school leaders who understand the new methods, instructional coaches who can observe classes and give feedback, time for teachers to plan and reflect together, and a culture that treats continuous learning as part of the job. None of this happens spontaneously. It has to be designed, financed, and managed.
But support is not enough on its own. Teachers and schools must also be accountable for what children actually learn. This is not about punitive testing or external pressure for its own sake. It is the basic condition for any system that wants to improve. Accountability for learning requires regular, standardized measurement of reading outcomes throughout primary school, and—crucially—the use of that data by teachers, school leaders, and ministries to adjust practice in real time. When a school knows how many of its second-graders cannot yet decode simple words, it can act. When a system knows where children are falling behind, it can reallocate support, revise materials, and intensify coaching where it is most needed. If education systems do not measure learning, they are flying blind. With measurement, schools and systems can learn and improve, year after year.
Too often, reading policy is shaped by tradition, pedagogical fashion, or political preference rather than by what actually helps children learn. Decisions about how to teach a foundational skill should be governed by the same standards of evidence that guide medicine or public health. Where they have been, results have followed. England's shift to systematic phonics produced its best-ever performance in international reading assessments, reaching fourth place among fifty countries in PIRLS 2021. The shift in the State of Ceará in Brazil during that last ten years took it to the top of the Brazilian rankings. The southern US states of Mississippi, Louisiana and Tennessee have improved dramatically.
This report, endorsed by the Global Education Evidence Advisory Panel (GEEAP), shows that ending the literacy crisis is neither a mystery nor a luxury. We know what to teach. We know how to teach it. We know it works in rich and poor countries alike, and across more than 167 languages. What remains is the hard, patient work of getting it into every classroom—and the political decision to do so.