Dozens of countries and states have adopted, drafted, or proposed social media bans for youth to address widespread concern that heavy use of social media harms adolescent mental health. But are such bans effective? This working paper studies outcomes from the first country-wide ban on social media for youth under 16, which began in December 2025 in Australia, and offers sobering evidence for policymakers who are planning similar bans.
Australia’s Online Safety Amendment (Social Media Minimum Age) Act sets a mandatory minimum age of 16 for holding an account on 10 major social platforms: Facebook, Instagram, Snapchat, TikTok, Threads, X, YouTube, Reddit, Twitch, and Kick. Messaging services such as WhatsApp and online games are excluded. Importantly, parents cannot bypass these restrictions. Enforcement is driven through the social media platforms: companies face civil penalties of up to $49.5 AUD million for failing to take “reasonable steps” to prevent under-16 accounts.
Unlike alcohol or tobacco restrictions, which are enforced at a physical point of sale, social media can be accessed from any device at any time, and the Australian law carries no penalties for individual users. Behavior change under the ban could theoretically occur through three channels: perceived sanctions for non-compliance, increased difficulty of access, and a shift in the peer environment that raises the social cost of breaking the norm. To study the ban’s effects, researchers surveyed 507 Australian teenagers aged 14–18 in March and April 2026, alongside a contemporaneous sample of American teenagers and an additional survey focused on mechanisms. The authors find the following:
- Compliance is low. Approximately 27% of banned 14–15-year-olds comply.
- Most banned teens believe their peers are still using the platforms, most find circumvention easy. Most non-compliers cite social reasons, that is, friends remaining on the platforms and fear of missing out, for their continued use.
- Compliers report difficulty keeping up with friends and feeling more bored.
The authors also ran a companion survey that asked teens two important questions: what did teens believe was the compliance rate of their peers, and what share of peer compliance is required before they, themselves, would comply? They find the following:
- Across different question framings, the mean stated threshold was around three-quarters, well above the observed 27% compliance rate.
- Combining stated thresholds with current peer-compliance beliefs, the only level at which compliance would be self-sustaining, that is, where the share of teens whose threshold is met equals the share complying, is around 18%, far below today’s rate.
- Finally, given that the current compliance level is substantially below two-thirds, compliance is more likely to erode than to rise going forward, absent supply-side or regulatory responses such as a change in enforcement.
What does the survey reveal about the social composition of those who comply vs. those who do not? Survey respondents perceived compliers as less popular than non-compliers, meaning that teens with the greatest social influence are more likely to remain on the platforms, keeping social media the socially dominant space. This contrasts with patterns observed in cigarette cessation among young adults, where well-connected individuals tended to quit together.
Bottom line: Sustaining high compliance would require not only a sufficient share of teens complying, but also that those who comply continue to find it preferable to do so, and whether that condition can be met will depend on whether the composition of compliers and noncompliers shifts under sustained enforcement, and whether social norms eventually realign.