School enrolment has risen for a quarter-century, and the number left outside the classroom has risen for seven years. Both statements are true. That is the uncomfortable relay message from the 2026 global education monitoring findings.
Since 2000, enrolment increased 30% in primary and secondary education, 45% in early childhood education and 161% in postsecondary education. Completion rates improved at every school level. Yet 273 million children, adolescents and young people were out of school in 2024, roughly one in six. Progress expanded the system. Population, inequality, conflict and weak finance kept the finish line moving.
The long-run gains are substantial
Primary completion rose from 77% in 2000 to 88%. Lower-secondary completion increased from 60% to 78%, while upper-secondary completion moved from 37% to 61%. Those are millions of additional learners reaching milestones their parents may not have reached.
Early childhood participation one year before primary entry reached 75% in 2024. Tertiary gross enrolment more than doubled from 19% in 2000 to 44%. Education systems are serving far more learners and retaining them longer.
These gains matter because a discussion focused only on the current gap can erase the work of teachers, families and public systems. The opposite error is just as serious: using historical progress to excuse present exclusion. An aircraft can have flown a long distance and still be off course.
Education has become larger faster than it has become universal.
The 273 million are not one group
“Out of school” covers a child who never enters primary school, an adolescent who leaves early, and a young person blocked from upper-secondary education. Their causes and remedies differ.
Distance and safety may prevent entry. Fees, uniforms, transport and lost household labour can make nominally free schooling unaffordable. Disability may meet an inaccessible building or an untrained teacher. Conflict can close schools or displace families across borders. Older students may leave for work, marriage or care responsibilities.
A single global target needs local diagnosis. Building more classrooms can help where distance is the constraint. It will do little if students leave because instruction is in a language they do not understand or because the household cannot afford for them to remain.
The statistic should therefore be treated as a route map, not a single queue. Policy must identify who is missing at each education level and at what point they left.
Seven years of rising exclusion is an alarm
The out-of-school population has increased for seven consecutive years. That trend means expansion elsewhere is not offsetting new exclusion. Conflict, displacement, demographic growth and economic shocks can push the number upward even as national enrolment rates improve.
Global averages may also hide regional concentration. A country with declining child population can improve its numbers more easily than a country building schools for a rapidly growing cohort. International comparisons should account for the size and age structure of the population being served.
The seven-year direction is more important than a single estimate. It indicates that current methods are not closing the gap fast enough for the 2030 education goal. The timetable is no longer generous.
Early childhood has a public-provision gap
Three-quarters of children participated in organised learning one year before primary school in 2024. The remaining quarter starts formal education with less preparation or no access at all. The gap can influence language, social development and later learning.
Many governments rely heavily on non-state providers. Only ten countries with data spend more than 1% on pre-primary education, according to the monitoring summary. When families pay most of the cost, access follows income and geography.
Expanding early childhood education is not merely adding miniature primary classrooms. Young children need play-based learning, safe facilities, trained staff and manageable group sizes. A low-cost room with a crowded worksheet routine may raise enrolment without providing the intended benefit.
Completion rose, but learning must be checked
Completion rates are easier to count than knowledge and skills. A student can remain enrolled and still struggle to read, calculate or apply concepts. Systems under pressure may promote learners without enough support.
This does not make completion meaningless. Staying in school creates opportunity and protects many children from harmful work or early marriage. It means the dashboard needs both measures: who finishes and what they can do.
Assessment should guide teaching rather than simply rank schools. If results show that a group is falling behind, resources should move toward tutoring, language support, materials and teacher development. Publishing a low score without support is an announcement, not a solution.
Higher education expanded faster than graduation
Tertiary enrolment rose from 19% in 2000 to 44% in 2024. Graduation moved from 17% to 27%, less than half the pace of enrolment growth. More people reach the gate, but a smaller share of that expansion converts into completed qualifications.
Affordability is one reason. Only one-third of countries have free public universities, and student support varies widely. Living costs, transport and foregone earnings can exceed tuition. Students may enrol, accumulate debt or expenses, and leave before graduation.
Course quality and labour-market relevance matter too. Rapid expansion can strain faculty, laboratories and student services. A university place is not valuable merely because it exists. It should provide credible teaching, a route to completion and knowledge that learners can use.
Policy should track time to degree, dropout by income and subject, and employment outcomes. Gross enrolment is the boarding pass. Graduation and capability are the arrival.
Teacher qualification is an infrastructure issue
Only 78% of primary teachers were academically qualified, down from 89% in 2013. In low-income countries, only 11% require a bachelor’s degree for primary teaching. The figures reveal pressure on recruitment and training.
Formal qualifications do not guarantee excellent teaching, and experienced teachers without a degree may be highly effective. Still, falling qualification levels can signal systems hiring rapidly without enough preparation. Teachers need subject knowledge, classroom methods and support for diverse learners.
Training cannot end at entry. New curricula, technology, multilingual classes and the effects of disruption require continuing development. Short workshops rarely change practice unless coaching, materials and school leadership reinforce them.
Teacher pay and working conditions affect retention. A system that trains teachers and then loses them to other work or overseas recruitment is leaking its most important capacity.
Finance is missing its own benchmark
Only 22% of countries with data met both international education-finance benchmarks in 2023: at least 4% of gross domestic product and 15% of public expenditure. These benchmarks are not magic numbers, but missing them so widely explains why expansion struggles to keep quality.
Spending totals should be paired with distribution. Money may favour universities in capital cities while rural primary schools lack teachers. Construction may be funded while maintenance and textbooks are not. Payroll can consume the budget without reaching understaffed areas.
Efficiency matters, but “efficiency” should not become a polite word for permanent cuts. A school needs enough teachers, safe buildings and learning materials before software or management reform can produce savings.
Conflict makes education both fragile and essential
Schools can be damaged, occupied or closed. Families move without records. Teachers may flee or go unpaid. Remote learning helps only where electricity, devices and connectivity exist.
Education is sometimes treated as a service that can wait until an emergency ends. For children, lost years do not wait. Safe temporary learning, accelerated programmes and recognition of displaced students’ prior work can preserve continuity.
Host communities also need support. Adding displaced learners to already crowded schools without more teachers and facilities can create resentment. Funding should strengthen the shared system rather than build a temporary parallel track whenever possible.
Technology can widen or narrow the distance
Digital lessons, translation and adaptive practice can reach students with limited local options. They can also exclude learners without devices, connectivity, quiet space or digital literacy. The tool inherits the conditions around it.
Technology should support teachers, not become an excuse to remove them from hard-to-serve places. A recorded lesson cannot notice confusion, protect a child or adapt to a classroom’s language without human help.
Procurement should begin with a learning problem. What barrier is the system trying to remove? How will teachers use the tool? Who maintains devices and pays for connectivity after the pilot? A dashboard full of logins can conceal very little learning.
The route to 2030 must become more specific
Governments need disaggregated attendance and completion data by age, sex, disability, location, income and displacement status. National progress can otherwise leave the same communities behind year after year.
They also need early-warning systems for dropout. Repeated absence, declining performance or unpaid fees can trigger support before a learner disappears. The ethical design is assistance, not punishment or surveillance.
Second-chance routes matter. Flexible schedules, vocational programmes and recognition of prior learning can bring adolescents and adults back. A system designed for one uninterrupted path excludes anyone whose life did not follow the timetable.
Progress should be defended without being romanticised
Exclusion often arrives through several doors
A rural girl with a disability may face distance, an inaccessible classroom, care duties and social pressure at the same time. Policies organised in separate departments can address one barrier and leave the others intact.
Inclusive education requires accessible buildings, trained teachers, assistive materials and safe transport. It also requires listening to learners and families. Placing a child in a classroom without support can produce formal enrolment and practical exclusion.
Gender patterns vary by place and age. In some settings girls leave because of safety, marriage or household labour. In others boys are more likely to leave for paid work. National campaigns should not assume one universal story.
Cash support, school meals and transport can reduce household cost. Their design should avoid burdensome applications that exclude the families with least time and documentation. A programme is not accessible merely because an office accepts forms.
The most useful data connect identities without turning students into categories. The purpose is to see where barriers overlap and send help, not to produce a longer table for the annual report.
The world has expanded education dramatically since 2000. Enrolment and completion gains are not statistical decorations. They changed lives. The current emergency is that the expansion has stalled short of inclusion while the out-of-school total rises.
Both truths belong in the same report. Declaring failure erases progress and can weaken support. Declaring success abandons 273 million people outside education and ignores millions more who attend without learning enough.
The next phase is less about a universal slogan and more about precise delivery: affordable early childhood services, qualified and supported teachers, targeted finance, safe access in crises, credible completion routes and useful learning.
There is no single boarding call for 273 million people. Each education ministry needs a route-by-route plan with named responsibility, funding and dates. International partners should support national systems rather than create short pilots that vanish with the grant. Consistency is less photogenic than a new device and far more useful to a child returning on Monday.
The classroom grew large enough to change the world; it has not yet grown fair enough to seat everyone waiting at the door.