Lead exposure has long been understood to be one of the most prevalent and pernicious environmental health hazards in the world. Today, the World Health Organization identifies lead exposure as a uniquely destructive public health problem, estimating that it causes over 850,000 preventable deaths a year, contributes to a large percentage of intellectual disabilities, and substantially increases risks of heart disease and stroke. The burden of lead exposure disproportionately falls on poorer countries, and yet in the United States, the richest country in the world, lead exposure remains a vexing issue. The potential benefits of a large-scale governmental effort to drastically reduce lead exposure are vast, and the costs are reasonable given the potential positive impacts on public health. But first it’s important to recognize what has been done and what work remains.

The Impact of Unleaded Gasoline

The outlawing of leaded gasoline stands as the single most important advance in controlling lead exposure in history. This might seem counterintuitive, as few people come in direct contact with gasoline, even when fueling their cars. But the essential exposure in the leaded gasoline era was not direct contact with liquid fuel, but rather exposure to exhaust from cars, which caused ambient gaseous lead and lead in the dust and sediments on roads and around them. Children playing in yards and parks near streets were exposed, as were pedestrians walking in congested areas. The ubiquity of automobiles in 20th century life meant exposure to leaded exhaust was very common.

In the latter half of the 20th century, leaded gasoline was phased out gradually in most of the developed world. The United States formally began a program to phase out leaded gasoline in 1976, after several years of battling industry groups over the proposal. Leaded gasoline was fully phased out in the United States by 1991. A 1994 study found that blood-lead levels in the United States declined precipitously in the 15 year phaseout period, demonstrating how robust environmental regulation can improve public health.

A Contested Connection to Crime

Given lead exposure’s well-known effects on cognitive functioning, including a decrease in impulse control and inhibitions against violence, some have contended that lead exposure and decline in same caused the spike in crime rates in American in the 1960s and subsequent decline in the 1990s and the period following. Though the lead-crime link has been proposed for decades, the Mother Jones writer Kevin Drum’s 2016 cover story on this potential connection has become something of a touchstone in this discussion, and Drum has followed it up with several blog posts further exploring the idea. Drum has tied lead exposure not only to crime rates but also to teen pregnancy, depressed IQ scores, and attention deficit hyperactivity syndrome.

The essential argument runs something like this: lead is a known environmental contaminant which has long been proven to cause cognitive impairments; among those impairments are degraded impulse control and a tendency towards violence; the massive reductions in crime rates since 1990 have coincided with significant reductions in lead exposure, primarily due to the phasing out of leaded gasoline and lead paint.

While this argument is intuitively satisfying, we cannot say with rigor that there is an actual causal impact here. Researchers typically depend on randomized controlled experiments to prove causation. For obvious reasons, we can’t conduct such experiments when it comes to lead and crime rates; deliberately exposing children to lead would be unethical. And while the associations in the data are striking, correlation is not sufficient to prove causation, as you so often hear on the internet. There are other explanations for rapidly-falling crime rates that are frequently touted: the “law and order” policing tactics and attendant high incarceration rates of the 1990s; rising poverty rates, social unrest and upheaval in the turbulent 1960s; and the famous “Freakonomics” argument that the legalization of abortion with the 1973 Roe v. Wade case resulted in fewer unwanted pregnancies and thus fewer children born into unstable homes. None of these explanations is without their own sets of problems.

A major question remains regarding the lead-crime hypothesis: why did crime rates spike in the first place? It’s true that lead exposure likely fell concurrently with declining use of leaded gasoline. But car ownership was already high in the 1940s and 1950s, and in fact use of lead paint, the second most common source of exposure in contemporary times, peaked far earlier in the 20th century than crime rates. We lack quality data on lead exposure rates from past eras, but it’s hard to imagine that the average lead levels were lower in the first half of the 20th century than the latter half. It’s therefore difficult to see how lead could have suddenly caused such a sudden spike in crime and violent crime rates. There are some attempts to address this concern, including pointing out that crime statistics were not as carefully collected in the past as they are today, but questions and criticism about the lead-crime hypothesis remain. Still, some recent research takes advantage of natural experiments and lends more credence to this idea.

High Levels of Exposure Persist

Like so many other facets of American life, lead exposure is stratified along race and class lines. Research demonstrates that those from poorer socioeconomic conditions suffer from significantly higher rates of lead exposure. Meanwhile, African Americans are exposed to lead at significantly higher rates even when controlling for income level.

The water crisis in Flint, Michigan demonstrates these inequalities, and proves that lead exposure remains a pressing public health issue. The city of Flint began drawing from the Flint river for their tap water in 2014; by the end of 2015, the rate of children with elevated blood lead levels in the city had doubled. In addition to an altered appearance and taste, the notoriously unclean water in that area has been tested at levels as high as 13,000 parts per billion, more than 850 times the EPA’s recommended safe level. Recent research suggests that, among other consequences, this level of lead has caused fertility rates in the area to plummet.

The Flint case has galvanized public opinion in part because it so directly demonstrates the socioeconomic realities of 21th century America. Flint was once a populous and economically robust city, standing with Detroit as one of Michigan’s centers of the mid-20th century automotive industry. But automation and offshoring led to a collapse in jobs for uneducated workers, with a subsequent loss of population, as today the city’s ~100,000 residents are less than half the population of the city in 1960. Those left behind are disproportionately poor, with some 40% living below the federal poverty line, and disproportionately black, at over 55%, according to the US Census Bureau. The city’s economic struggles and a state austerity regime led to the city switching its water supply, demonstrating how disadvantage multiplies itself in our system.

Lead exposure in our country’s poorest environs should be enough to provoke a public policy response. But wealthier Americans should understand that they have self-interested reasons to support such actions as well. This year, a Reuters report found that throughout Los Angeles county, even in wealthy enclaves like San Merino, children had far higher levels of lead exposure than commonly assumed. In fact, Reuters has found  more than 3,300 areas in the United States with dramatically elevated childhood lead exposure, areas including all kinds of demographic and socioeconomic backgrounds. So while lead exposure is stratified in predictable ways given this country’s underlying wealth distribution, even those who are among society’s economic winners should support a large-scale lead abatement initiative.

The Reuters investigation cannot be assumed to be nationally representative. Rates of elevated lead levels and exposure can be idiosyncratic and we lack adequate large-scale data, though the CDC estimates that 2.5% of American children have elevated lead levels. Whatever the exact percentage, the extant evidence should be more than enough to prompt governmental response and public investment.

Positive Impacts from Intervention

There is some good news. A recent study demonstrates that intervention programs for children exposed to lead can have a positive impact on their academic and behavioral outcomes. In the study, researchers examined a dataset from the 1990s, comparing outcomes from children who had similar levels of lead in their blood and who either fell just within or just outside of a given cut point for eligibility in a lead exposure management program, forming natural test and control groups.

The interventions included education and awareness campaigns, general medical screenings for overall childhood health, nutritional interventions which may help mitigate the effects of lead exposure, educational interventions, and for higher levels of exposure, efforts to physically locate and remove the sources of contamination (usually lead paint). The effect size of the outcomes from such intervention was fairly modest, at about .12 of a standard deviation for educational impacts and .18 of a standard deviation for behavioral impacts. Still, these positive effects were highly statistically significant, and even minor improvements of this kind can have a major impact on the quality of life of children. The per-child cost of the program was $5,288. To scale such a program up for the 5% of American children estimated by the CDC to have high levels of lead exposure would thus cost about $20 billion. That number may seem high, but bear in mind that the F-35 fighter plane is projected to cost the United States more than 72 times that amount over its lifespan.

Failure to Eliminate Lead Exposure is a Policy Choice

Back in 1991, the Federal Department of Housing and Urban Development estimated that some 57 million American homes were in need of lead abatement, with 3.8 million of them designated high priority cases in which children were likely exposed to lead dust. The department estimated at the time that it would cost between $2,100 and $11,000 per home to remove the leaded paint, for a potential national program cost of almost $500 billion. While many smaller programs have been undertaken in that time, no systematic national attempt to ameliorate lead exposure has taken place. As Flint demonstrates, we are far from a lead exposure-free society 25 years after HUD’s report.

Ultimately, our lead problems are a matter of political will. Whether it caused the late 20th century crime surge or not, lead’s health consequence are clear and troubling. What’s more, we have the necessary expertise and construction infrastructure to clean up lead-contaminated spaces. What we need is a muscular federal policy to combat this problem, funded with public dollars. The potential benefits to public health far outweigh the likely costs.