Princeton Election Consortium

A first draft of electoral history. Since 2004

Environmental lead (Pb) and crime

January 4th, 2013, 6:30pm by Sam Wang


At Mother Jones, Kevin Drum has an in-depth article on the hypothesis that environmental lead  is a likely root cause of the increase in crime a generation ago. Even more importantly, removal of lead from gasoline may be responsible for the decrease in crime in the last few decades. Last night, Kevin and I discussed these ideas with Jay Ackroyd on Virtually Speaking.

Support for the environmental-lead/crime association has been building for some time, as described for example in these articles by Rick Nevin and David Carpenter. Two key elements are there: (1) historical epidemiological data from many countries, and (2) a plausible biological mechanism (PDF): effects of lead on the developing brain, especially the prefrontal cortex, and on IQ.

This two-pronged argument is reminiscent of the linkage between smoking with cancer. I find it considerably more persuasive than the speculation by Steven Levitt (Freakonomics) that legal abortion was responsible for decreases in crime. That idea is based on just one study, lacks a convincing mechanism, and does not account for earlier increases in crime, which the lead hypothesis does.

After Kevin left, Jay and I continued our discussion, branching off into child development, individuality, and many other topics from my book Welcome To Your Brain (currently on sale!). It was fun – check it out here.

Update, Saturday: here’s just the part of the program on lead and crime.

Tags: 2008 Election

9 Comments so far ↓

  • Amitabh Lath

    I always thought the Freakonomics authors were courting controversy with the abortion – crime correlation.

    Can you give links to some rat and/or primate studies on this Pb issue?

    PS: There is a lot of lead used in particle physics. I helped build a lead brick wall (to shield muon counters) as one my first tasks as a postdoc. Picking up a lead brick with gloves on is difficult.

  • Pat B

    This is a fascinating topic for me – a lay person/ordinary citizen: That such a correlation between lead and crime has been documented by many studies, even world wide. I have been reading about it in many articles in the last week or so.

    It needs to be publicized enough so that law enforcement and lawmakers will take it seriously and we can begin to address solutions – such as Kevin Drum suggests in the Mother Jones article.

    What were possible effects of early lead exposure on the brains of these mass shooting perpetrators? Would it be possible to find out?
    Is there a relationship?

  • Amitabh Lath

    As another lay person/ordinary citizen, I want to see this followed up a little more. There have been too many people wanting to claim credit for the drop in crime rates (broken windows theory, Compstat, the Roe-vs-Wade correlation, welfare reform…).

    That being said, this Pb poisoning theory looks promising. Plausible mechanism gives this theory credence. Otherwise it’s just another correlation. Like the correlation of fluoride in the water and cancer rates (it turned out the towns that had fluoridation were richer towns, where people lived longer, and cancer is basically a disease of old age).

    Bringing up kids in NJ, a state with several brownfield sites, we have to stay vigilant to all sorts of environmental contaminants. Heavy metals are one of the worst.

  • mediaglyphic

    i have also heard that the large majority of violent crimes are commited by men between the ages of 15 and 30. As this cohort becomes a smaller part of the population, crime comes down. There probably are multiple drivers of things like crime. (which itself is perhaps tough to define, for example were execs at large banks who through faulty supervision or other indirect means, partially responsible for the economic meltdown, criminals?)

  • Amitabh Lath

    I had a thought on the Pb vs crime stats. Comparing
    datasets separated by time has problems, since other
    factors (abortion, policing tactics) cannot be cleanly
    factored out.

    But there must have been a transition period where
    some new schools were built without lead paint, while
    older schools still had it. If we can find a time
    window where there were basically equal numbers of
    old lead paint schools and new non-lead paint schools
    (in some large metropolitan area) then you can do a
    clean test. Of course, old and new schools might be
    correlated with rich and poor districts. But unless
    all schools got fixed at once, there should be some
    time window that makes sense.

    As for leaded gasoline, I think there could be a
    visible effect when you take cities that have a
    lot of wind (Chicago) vs. cities which have still
    air (LA).

    • Tom Johnson

      Amitabh, read the Drum article that Sam linked. It provides references to several interesting articles that speak somewhat to your point. The lack of a SUPER good natural experiment is somewhat mitigated by a really impressive number of fair natural experiments.

  • Dave Kliman

    so i wonder what the implications are for gun owners, what with all the lead they’re constantly exposed to….

Leave a Comment