There are literally hundreds of estimates of the effect of schooling on earnings (actually 1,120 per one recent review). But there is just one single published estimate of the effect of foundational skills on earnings in a low- or middle-income country (by Glewwe, Song, and Zou, in China). This is pretty surprising, given the global policy focus on the learning crisis, and associated emphasis on improving foundational skills over other educational goals.
There's a good reason for the lack of evidence on long-run gains from learning. To estimate the effect of schooling you can simply ask an adult how long they went to school for. But you can’t really ask an adult how good their reading was when they were 10 years old, or at least you couldn’t rely on their answer. You need to have actually measured it 15-20 years prior.
In a new CGD working paper, I do just this, using longitudinal data from the Indonesia Family Life Survey. Longitudinal surveys that follow individuals over long periods of time are rare in low- and middle-income countries. I found 22 surveys, and only one which both tested children on their foundational skills in primary school, and tracked the same individuals through to adulthood and asked about later outcomes.
Key findings
- Higher earnings: Children who score well in reading and maths (after accounting for their family background) go on to earn roughly 11 percent more as adults.
- More than just schooling: Children with better foundational skills tend to stay in school for longer, but this only partially explains the earnings premium. This suggests that foundational skills can equip individuals with skills valued directly in the labor market, beyond just formal qualifications.
- Broader life outcomes: Beyond earnings, individuals with stronger foundational skills as children were slightly less likely to have had children themselves by early adulthood, indicating a potential link between early learning and reproductive choices. While we didn't find consistent correlations with other health outcomes in this study, the impact on earnings and fertility highlights the broad influence of foundational skills.
Why this matters for policy
If these relationships can be interpreted as causal—and the robust set of controls in this analysis suggest they might—then the magnitude of these returns implies a large positive benefit-cost ratio for investments in foundational skills.
For policymakers and practitioners in education, these findings provide a strong economic rationale for prioritizing foundational literacy and numeracy interventions. But aid donors are doing the opposite. The UN has projected that global aid for education will fall by a quarter between 2023 and 2027, with an even steeper decline expected for schools. The US and UK—until recently the largest bilateral donors for basic education—are both expected to almost entirely cut that spending. In the UK, overall aid is set to fall by 40 percent, and the development minister has signaled that aid for education will be cut by even more. Meanwhile, donors like Japan, who are increasing their overall aid budgets, spend far more on university scholarships—either for foreign students in Japan or for students in poorer countries—than on helping young children in those countries access a basic education.
The size of these estimated returns suggests that improving foundational learning yields such significant economic benefits that even major increases in investment—five or ten times current levels—could still represent excellent value for money. At a time when global support for basic education is falling, that’s a message policymakers can’t afford to ignore.
