Smartphones and Children's Mental Health
What happened when English communities delayed their children's smartphones.
A research summary for policy makers and journalists, July 2026.
The working paper is currently under peer review.
The question
Children's mental health has deteriorated across high-income countries since smartphones became a fixture of childhood. Governments are responding with school phone bans and social-media age limits. Yet there has been no causal evidence on the step many parents care about most, delaying the first smartphone. The Smartphone Free Childhood movement launched in the UK in February 2024. It organizes parents through a “Parent Pact,” a public pledge to delay smartphones until age 14. By early 2026, over 165,000 parents across more than 15,600 UK schools had signed the pact. The change on the ground is striking. At one of the early adopting schools, Year 6 smartphone ownership fell from 75 percent to 3 percent in two years.
The study
I compare how mental-health needs changed at English primary schools where at least 20 percent of families signed the pact with similar schools where almost no one signed. Children's mental health is measured in two independent administrative systems: (1) the school census, where schools record each pupil's social, emotional and mental health (SEMH) needs, and (2) child psychiatric disability benefits (DLA), which government caseworkers determine with no direct involvement from schools.
The findings
In high-signup communities, recorded mental-health needs fell by about 8 percent within two years, relative to similar comparison schools. Total child psychiatric disability claims fell by about 6 percent in the same communities. Both effects grow over time and with the fraction of families that signed. The effects are about twice as large for girls. These headline numbers are conservative, because they include older children, all psychiatric conditions, and all families, whether they signed or not. The actual effect for behavioral conditions in younger children may be two to four times larger.
Is this just engaged parents?
High-signup communities are more affluent and their children may have lower mental-health needs for other reasons. The analysis addresses this concern in three different ways.
- Like is compared with like. Each high-signup school is compared with schools that looked the same before 2024, with the same recorded mental-health levels, similar affluence, and similar size. The estimates compare changes within high-signup schools to the changes within similar low-signup schools. Average differences between schools and communities are removed.
- Two independent systems agree. The effects are similar whether they are measured by schools or the Department for Work and Pensions. It is unlikely that the decision-making processes at two different agencies change at the same time only in engaged communities.
- The pattern matches a phone delay rather than general engagement. I find effects only for behavioral conditions such as ADHD. In contrast, I find no effect on autism, developmental disabilities, or adult depression in the same neighborhoods. The effects are also much stronger for 5 to 10 year-olds, who are less likely to already have a phone, than for older children, who mostly already own one. It is unlikely that general parental engagement alone could produce this pattern.
What it means
Parents fear that withholding a smartphone will socially isolate their child. The pact solves this coordination problem by converting a private intention into a public, school-year-specific commitment. The findings give parents causal evidence that delaying the first smartphone through the primary-school years improves children's mental health, and they give governments a near-zero-cost complement to school phone bans and statutory age limits. The findings are also relevant for the UK fiscal situation, where the outcomes studied here drive two of the fastest-growing items: child disability benefits (over £4 billion a year) and high-needs education (£12 billion). Two cautions apply when interpreting these results. The evidence covers only the movement's first two years. The adopting communities are relatively affluent, and the effects in more disadvantaged areas could be larger or smaller.
Contact: gregory.veramendi@rhul.ac.uk. The full paper, with all methods, robustness checks, and data descriptions, is available from the IFS.