Three quarters of the world’s children today enter a pre-primary structure in the year before school — but this raw figure overestimates reality: UNESCO’s 2026 GEM report clarifies that 27% of these children were already enrolled in primary school, meaning only approximately 60% of children actually had access to at least one year of pre-primary education. This is a false victory, doubly so: behind this figure lies massive underinvestment that UNESCO’s 2026 Global Monitoring Report on Education documents across 119 countries. Median global spending on early childhood education stagnates at 0.43% of GDP. Only ten countries cross the 1% threshold. Meanwhile, economists have agreed for twenty years that no other moment in life produces as much return per euro invested. The issue is not whether early childhood deserves budgetary effort. It is understanding why such a well-documented lever remains so systematically ignored.

The Essentials

  • Median global spending on pre-primary education reaches 0.43% of GDP, against a recommended threshold of 1%, according to the 2026 UNESCO report covering 119 countries
  • Only 10 countries out of 119 exceed this minimal threshold, despite two decades of academic consensus on the economic return of this education stage
  • Doubling public spending per child is associated with tripling participation, according to the same UNESCO data
  • Lithuania illustrates this mechanism: it simultaneously increased spending per child and advanced enrollment rates
  • The obstacle is structurally political, not economic: future beneficiaries do not vote, costs are immediate, and gains are measured over fifteen to twenty years

Academic Consensus Is Not Enough

James Heckman is a Nobel Prize-winning economist. He spent part of his career measuring the return on educational investment according to age of intervention. His conclusion, since replicated in dozens of studies across multiple continents, is that returns are diminishing: they are maximal between zero and five years, and they decrease as one advances through the curriculum. The mechanisms are known. The young child’s brain exhibits exceptional synaptic plasticity. Cognitive and socio-emotional skills acquired early serve as the foundation for all subsequent learning. Investing late means attempting to build without foundations, at higher cost for lesser result.

This is not a marginal thesis. The OECD, the World Bank, and UNESCO have referenced it since the 2000s. The Perry Preschool Program, followed for forty years in the United States, showed that disadvantaged children who benefited from a quality pre-primary program had, in adulthood, higher incomes, lower crime rates, and less dependence on social assistance than those in the control group. The initial investment was recouped several times over through savings generated subsequently in the justice system, health, and social transfers.

The consensus is there, solid, long-standing, and internationally shared. Yet it does not translate into budget policy.

0.43% of GDP: Twenty Years of Stagnation

UNESCO’s 2026 data are striking in their precision. Over two decades, median global spending on pre-primary education progressed only from 0.29% to 0.43% of GDP. It is an increase, but it remains well below the recommended 1% threshold. Above all, it masks extreme dispersion: countries investing little stagnate, those already investing more progress, and the gap widens.

Of the 119 countries covered by the report, only 10 exceed the 1% threshold. It is not exclusively a matter of national income. Some middle-income countries have made the political choice to dedicate significant resources to it. Other rich countries remain below the threshold. The correlation between GDP per capita and pre-primary spending exists, but it is far from mechanical. It is a matter of priority, not capacity.

Perhaps the most telling figure in the report is this one: doubling public spending per child is associated with tripling participation. The elasticity is massive. When supply exists and its cost for families decreases, children attend. It is not a problem of social demand. It is a problem of supply built by public authorities.

Lithuania Demonstrates What the Rest Refuses to Attempt

In this landscape of general stagnation, Lithuania stands as a methodical exception. The country simultaneously increased its spending per child in pre-primary and recorded a rise in enrollment rates. This is not a coincidence: it is the mechanism described by UNESCO on a global scale, but applied deliberately.

What makes the Lithuanian case instructive is less the scale of the amounts committed than the underlying political logic. The country treated early childhood education as an infrastructure, much like a road network or health system: an investment whose return is deferred but documented, which justifies stable and predictable public spending. This approach contrasts with the typical treatment of this budget item, often seen as an optional social expense rather than as a structuring economic investment.

Other examples exist. France maintains near-universal enrollment from age three, a legacy of a public and free kindergarten system whose coverage is exceptional on a global scale. Uruguay has built a network of integrated early childhood centers (the CAIFs) that combine childcare, stimulation, and nutritional monitoring for children from disadvantaged backgrounds. These models are diverse, but they share a common characteristic: they result from an explicit political choice, maintained over time, independent of governmental changes.

The Structural Bias Against Long-Term Investments

If the economic argument is known and operational models exist, why do nine countries out of ten remain below the threshold? The answer lies in a structural distortion of political time.

The beneficiaries of investment in early childhood are the children themselves, in fifteen or twenty years. They do not vote. They do not lobby. Their parents vote, but they are rarely organized into a pressure group powerful enough to weigh against better-structured interests. Conversely, the costs of investment are immediate and visible: construction of facilities, training of educators, recurring budget. Within an electoral cycle of four to five years, an education minister who dedicates resources to pre-primary will never see the results of his or her mandate.

This logic is universal. It explains the chronic underrepresentation of early childhood education in budget arbitration, even in countries that publicly display their adherence to SDG 4 objectives (quality education for all). The 2026 UNESCO report notes that the global objective of guaranteeing universal access to quality pre-primary education by 2030 will not be achieved at the current pace. This is unsurprising: the trajectories observed over two decades made it predictable.

There is also a measurement bias. The effects of early childhood education are dispersed across time and space: they appear in school statistics at age 10, in employment data at age 25, in health indicators at age 40. Political accounting systems are poorly equipped to capture this type of long causal relationship. What one measures easily, one defends easily. What one measures poorly, one cuts first.

The phenomenon is not specific to education. The same structure is found in arbitrations over medical prevention, long-term infrastructure, or energy transition. As the Draghi report highlighted for the European economy, shared diagnoses do not automatically translate into action: institutional mechanisms are needed to protect long-term investments from short political cycles.

Inequality Is Decided Before the First Day of School

The consequences of underinvestment are not distributed uniformly. They strike first children from disadvantaged backgrounds. In countries where pre-primary supply is limited, affluent families find private solutions. Poor families do not. The result is that the cognitive and socio-emotional gap between rich and poor children is already substantial on the first day of primary school. Much of the subsequent effort of the education system consists, in part, in catching up a gap that should never have widened.

This mechanism has been documented with precision in several national contexts. In the United States, Sean Reardon’s work at Stanford shows that the gap in results between rich and poor children at the start of primary school has increased over the past forty years, partly because affluent families invest increasingly early in their children’s development, while public supply remains limited for others. In France, despite universal kindergarten, studies show that inequalities in language acquisition at the start of CP remain significant and largely follow the lines of social inequalities.

Social mobility, a central theme in education debates, largely passes through the early years of life. An education system that arrives too late, after initial inequalities have settled in, works against itself. It spends more for lesser results, because the foundations are lacking.

UNICEF estimates that 250 million children worldwide do not reach their full development potential before age five, due to lack of access to adequate health, nutrition, and early education services. This is a figure that must be read as what it is: massive waste of human capital, with economic and social consequences that span decades.

What Countries Making Progress Have in Common

Countries progressing on this issue do not share a single model. They share a method. The first characteristic is institutional continuity: early childhood policies only work if they last. Positive effects do not manifest in the two or three years following program implementation. Governments that launch initiatives and then abandon them in the next cycle waste their initial investment.

The second characteristic is service integration. The most effective programs do not separate childcare, education, health, and nutrition. They combine them. Uruguay with its CAIFs, Finland with its neuvola (integrated pre- and post-natal health centers), or Jamaica with its stimulation program at home for malnourished children, show that impact is maximal when the intervention is comprehensive.

The third characteristic is the quality of professionals. Access to a pre-primary structure is not enough if educators are undertrained and poorly paid. Countries achieving the best results have also invested in the professionalization of the early childhood sector, recognized as a demanding profession and valued accordingly. This is an additional cost, but the data suggest it conditions return on investment.

This last point raises a question that extends beyond early childhood and runs through entire education systems: how do we train, attract, and retain the professionals who transmit fundamental knowledge? An issue taking on new urgency as digital tools redefine what it means to teach and learn, as suggested by debates on the place of technology in schools.

The next decade will decide whether the official figure of 75% of children accessing some form of pre-primary education in the year before primary — a figure that the 2026 GEM report itself acknowledges overstates reality, the actual rate being approximately 60% — represents a floor or a ceiling. Achieving universal access before 2030, an objective the United Nations has set for itself, would require doubling or tripling spending in dozens of countries in less than five years. The pace observed over twenty years does not suggest this movement will be spontaneous. What has worked elsewhere is when political actors decided to treat early childhood as a non-negotiable budgetary priority, protected from short-term arbitrations. The question is not whether it is possible. It is who decides to do it.

Sources

  1. UNESCO, Global Monitoring Report on Education 2026: https://www.ei-ie.org/fr/item/32242:rapport-mondial-de-suivi-sur-leducation-2026-les-progres-en-matiere-dacces-a-leducation-ne-sont-pas-a-la-hauteur-de-lobjectif-de-developpement-durable
  2. James Heckman, University of Chicago, works on the return on early childhood educational investment (heckmanequation.org)
  3. UNICEF, data on early childhood development
  4. Sean Reardon, Stanford Center for Education Policy Analysis, studies on educational inequalities in the United States
  5. Institute for Public Policy (France), studies on language acquisition inequalities at the start of CP
  6. UNESCO GEM Report 2026 - Access and Equity (primary report): https://www.unesco.org/gem-report/en/publication/equity-and-access
  7. World Education Blog GEM 2026 - Pre-primary education financing: https://world-education-blog.org/2026/05/06/ineffective-financing-is-holding-early-childhood-education-back/
  8. Education International - GEM Report 2026 Summary: https://www.ei-ie.org/en/item/32242:global-education-monitoring-report-2026-progress-in-ensuring-access-to-education-falls-short-of-sustainable-development-goal
  9. HighScope Perry Preschool Study - Results at 40 years: https://highscope.org/project/perry-preschool-study/
  10. Heckman Equation - Diminishing returns by age: https://heckmanequation.org/resource/invest-in-early-childhood-development-reduce-deficits-strengthen-the-economy/
  11. SDG 4 - Target 4.2 universal access to pre-primary education: https://unric.org/en/sdg-4/
  12. UNICEF - Early Childhood Development Vision for Every Child: https://www.unicef.org/media/145336/file/Early_Childhood_Development_-_UNICEF_Vision_for_Every_Child.pdf
  13. Stanford CEPA - Sean Reardon, income achievement gap: https://cepa.stanford.edu/content/widening-income-achievement-gap
  14. Institute for Public Policy (IPP) - Education page: https://www.ipp.eu/thematiques/education/