Not All Children Breathe Equal
Air pollution hits the most vulnerable infants hardest
We know air pollution is bad for children. That much is settled science. But here is something we understand far less well: it is not bad for all children in the same way. Some infants are barely affected by a spike in fine particulate matter. Others end up in the emergency room. Who are those children, and what makes them so vulnerable? And what can policies do about it?
These questions matter because most air quality regulations work the same way everywhere: set a threshold, reduce pollution in areas that exceed it. But if the health damage is concentrated among specific children rather than spread evenly across the population, then where you cut pollution may matter less than who you protect. Our research suggests this is exactly the case, and that it calls for a fundamentally different approach to environmental health policy.
In a paper just published in the Journal of Environmental Economics and Management with Milena Suarez Castillo and Christine Le Thi, we investigate these questions using administrative data covering about 340,000 children born in France between 2008 and 2017.
The setting
France offers a unique window into this problem. Its universal health care system means that virtually every doctor visit, emergency admission, and prescription is recorded for every child, regardless of family income. There are no insurance gaps, no missing populations. We can observe exactly what happens to children’s health when pollution spikes, across the entire income distribution.
And pollution does spike. Weather events like thermal inversions (when a layer of warm air traps pollutants near the ground) or winds blowing from industrial areas cause sudden, short-lived increases in PM2.5 concentrations. These events are essentially random with respect to individual families, which lets us isolate the causal effect of pollution on health.
What we found
On average, infants exposed to more pollution days in their first year are more likely to visit the emergency room for bronchiolitis. But the average masks enormous disparities.
When we dig into who is actually affected, a striking pattern emerges. The health impacts of air pollution are concentrated among roughly 10% of infants, characterized by two things: lower parental income and poorer health indicators at birth (premature birth, low birth weight). For these children, exposure to pollution increases the likelihood of an emergency admission for bronchiolitis by 1.6 percentage points, about 33% above their baseline risk. For the other 90%, the effects are much smaller or statistically insignificant.
The income gradient is stark. Children from the lowest income decile account for 14% of all bronchiolitis-related emergency visits. When air pollution spikes, that share rises to between 15% and 17%. Meanwhile, wealthier families use more preventive medication and visit pediatricians more often, suggesting they have better tools to manage the same environmental threat.
Rethinking how we target environmental policy
This has direct consequences for how we design air quality regulations. Most current policies target geography: reduce pollution in the most polluted areas. That makes intuitive sense, but our results suggest it may not be the most effective approach. The children who suffer most from pollution are not necessarily the ones who breathe the most of it. They are the ones least equipped to cope with it.
Consider two neighborhoods with the same pollution level. In one, most families have access to pediatricians, preventive asthma medication, and the resources to keep vulnerable infants indoors on bad air days. In the other, families rely on emergency rooms and have little capacity to adapt. The same pollution spike produces very different health outcomes. A policy that treats both neighborhoods identically misses this entirely.
Our results suggest that targeting based on vulnerability, prioritizing areas with high concentrations of premature births and low-income families, could deliver substantially greater health benefits for the same investment. Simple indicators like the share of low birth weight babies or the local poverty rate could serve as practical targeting tools for regulators. This aligns with the EU’s ongoing push to update its Ambient Air Quality Directives, which aim for stricter local thresholds by 2030, and could inform how those thresholds are set at the local level.
The broader lesson is simple but often overlooked: environmental policy is health policy, and health policy is inequality policy. You cannot address one without understanding the others.
Suarez Castillo, M., Benatia, D., and Le Thi, C. (2025). “Air Pollution and Children’s Health Inequalities.” Journal of Environmental Economics and Management, 131, 103149. DOI.
