The Lead-Crime Hypothesis—that the toxic effects of leaded gasoline explain big spikes in crime—offers convincing evidence and helps explain why violent crime fell so much from the 1990s to today.
Stunning that action didn’t happen earlier. Tetraethyl lead, just from its name, a chemist should immediately be able to tell this is a highly toxic compound. If one was tasked with finding a chemical to efficiently deliver lead to biological systems, tetraethyl lead might be a prime candidate, might even be the first one to try.
But proving causation like you mention is really difficult, especially for humans. We have a whole lot of ethical considerations - can’t just assign people to groups and feed some of them a bunch of poison to study the effect. One cannot control isolate an individual’s environment scientifically for any significant length of time either. So, you don’t have complete data to prove a case.
And if you have an interested party on the other end that is printing money by marketing the dangerous product, they can be very effective running interference, stymie public awareness enough to prevent regulation. We saw this with tobacco a generation ago and climate change for the past several decades. It now seems there is PR misinformation playbook that is very effective at enabling continuation of harmful business practices.
Yes, what’s maddening here is everyone knew. This is the stuff where I think a lot of people just assume the government will catch it if it’s really harmful...but that’s often not true.
Such a shame. Sarah Kendzior has written some neat ideas about ‘normalcy bias’ as it applies politically, largely in reference to elite accountability, corruption and MAGA. The thinking is ‘if it were that bad and dangerous as it appears, certainly someone would have done something about it by now…so it must not actually be that dangerous.’ Her contention is the longer no action is taken, the more normalcy bias can be exploited by propagandist/influencers. Eventually, even leaders who may have originally resisted or opposed a crime or corruption can essentially become co-opted into defending it, or at least defending lack of serious action against it. And when things get bad enough, lack of significant action in opposition start to become complicity.
One added thought regarding use of lead by the Romans. This not only occurred in plumbing, but also in the use of lead cups for drinking. Lead was used to give weight, leaf gold to make the cups also look like gold. Over time and with the chemical effects of wine, the gilding gradually wore off and the average consumer drank wine laced with lead. There is a theory that this was a significant contribution to the decline of the Roman Empire.
I didn’t know that. This is what I love about this community! I do all this research and then learn stuff after I publish from you. And what a fascinating hypothesis...
I don’t think it usually works to use broad social factors as a criminal defense, but the question is apt because it exposes the flaws of simplistic assumptions of blame. Does a criminal who has had brain damage from environmental toxins and became violent as a result deserve blame? Scrutinising moral culpability becomes tricky very fast in the complexity of the real world, a subject I’ll be writing on more soon.
Brian, as an PhD economist by training and operations research and power system engineering professional we are always caught in the trap of being “more rigorously imprecise” where by we create ever more complex models that explain little or nothing. I have been seeing this everyday for over 25 years in the power industry.
From an economics standpoint so much is driven by simplistic Econ 101 ideas of markets and how they should be, and ignore that facts that many convenient assumptions we use simply do not hold true in the real world. In this view policy makers and regulators are too simplistic.
On the opposite end are engineers who double as “economists” who think they know the complex elements of mechanism design (in full disclosure one of my first year profs, Leonid Hurwicz, was awarded a Nobel prize for his work in this area), and complex economic models but know nothing. In this case they can do math, but do not have the sense of social scientists to ask questions and be skeptical. In short, they lack a sense of knowing what it is they do not know.
Your lead poisoning case is one I was aware of and you did a great job of summarizing it. As well as the history of Mr Midgely. Sometimes, the evidence is just overwhelming and we ignore it because we do not really want to believe it. Sometimes we want to make things more complex than they really are, just as much as we make things simpler than they really are.
This also shows the hubris of one man in conjunction with economic greed and incomplete information of large economic actors, can lead to such awful trends as you showed.
Now a related question from this article: the lagged timelines also match up with urban decline that coincided with the worst of pollution seen in the late 60s and 1970s. Did leaded gasoline also lead to flight from the urban centers to the suburbs and exurbs? And conversely, did the phase out of leaded gasoline lead to the renewal of many urban areas up to the COVID epidemic? How are these related to crime that may have been caused by lead poisoning?
When I was in college in Providence, RI circa 2000, lead removal was a big topic locally. Many residents planted sunflowers on the belief that they sucked lead out of the soil, and the flowers were then sent to a special disposal facility. I lived with environmental science majors, and one of my friends later worked for a nonprofit that educated about lead paint exposure and mitigation. Anyway, regarding this historical narrative, I do wonder about the possibility of correlation rather than causation: after all, the railroads didn't only bring paint, but also industrialization, capitalism, etc. Lots of factors contributed to violence. Then again, as you say, countries wised up to lead and phased it in and out at different times, which helps isolate the variable.
The debate, as far as I can tell, is about how much it's the causal variable, not whether it is one. Nevin puts stock in it as the primary cause; others as a contributing factor. I don't think many people think it had no effect.
Let's not ignore Calvin Coolidge and his role. When the petroleum companies panicked over having lead gasoline banned because of its obvious health effects on those who made it, good ol' "no regulation is good regulation" Coolidge formed a study group, packed it with industry reps, and viola, it's no longer a problem so long as the workers wore masks and gloves. As for the effect on the public. "We'll worry about that later."
Stunning that action didn’t happen earlier. Tetraethyl lead, just from its name, a chemist should immediately be able to tell this is a highly toxic compound. If one was tasked with finding a chemical to efficiently deliver lead to biological systems, tetraethyl lead might be a prime candidate, might even be the first one to try.
But proving causation like you mention is really difficult, especially for humans. We have a whole lot of ethical considerations - can’t just assign people to groups and feed some of them a bunch of poison to study the effect. One cannot control isolate an individual’s environment scientifically for any significant length of time either. So, you don’t have complete data to prove a case.
And if you have an interested party on the other end that is printing money by marketing the dangerous product, they can be very effective running interference, stymie public awareness enough to prevent regulation. We saw this with tobacco a generation ago and climate change for the past several decades. It now seems there is PR misinformation playbook that is very effective at enabling continuation of harmful business practices.
Thanks again for another eye opening article!
Yes, what’s maddening here is everyone knew. This is the stuff where I think a lot of people just assume the government will catch it if it’s really harmful...but that’s often not true.
Such a shame. Sarah Kendzior has written some neat ideas about ‘normalcy bias’ as it applies politically, largely in reference to elite accountability, corruption and MAGA. The thinking is ‘if it were that bad and dangerous as it appears, certainly someone would have done something about it by now…so it must not actually be that dangerous.’ Her contention is the longer no action is taken, the more normalcy bias can be exploited by propagandist/influencers. Eventually, even leaders who may have originally resisted or opposed a crime or corruption can essentially become co-opted into defending it, or at least defending lack of serious action against it. And when things get bad enough, lack of significant action in opposition start to become complicity.
Brian, a brilliant paper! Congratulations.
One added thought regarding use of lead by the Romans. This not only occurred in plumbing, but also in the use of lead cups for drinking. Lead was used to give weight, leaf gold to make the cups also look like gold. Over time and with the chemical effects of wine, the gilding gradually wore off and the average consumer drank wine laced with lead. There is a theory that this was a significant contribution to the decline of the Roman Empire.
I didn’t know that. This is what I love about this community! I do all this research and then learn stuff after I publish from you. And what a fascinating hypothesis...
Could explain many other issues that we are plagued with as well. Excellent research.
As an afterthought, it just occurred to me that I quit having migraine headaches after “ethyl” was outlawed.
Intriguing. Thank you, Brian.
I wonder if it’s ever been adduced as a criminal defence?
I don’t think it usually works to use broad social factors as a criminal defense, but the question is apt because it exposes the flaws of simplistic assumptions of blame. Does a criminal who has had brain damage from environmental toxins and became violent as a result deserve blame? Scrutinising moral culpability becomes tricky very fast in the complexity of the real world, a subject I’ll be writing on more soon.
Thanks Brian. I’ll look forward to reading that.
Brian, as an PhD economist by training and operations research and power system engineering professional we are always caught in the trap of being “more rigorously imprecise” where by we create ever more complex models that explain little or nothing. I have been seeing this everyday for over 25 years in the power industry.
From an economics standpoint so much is driven by simplistic Econ 101 ideas of markets and how they should be, and ignore that facts that many convenient assumptions we use simply do not hold true in the real world. In this view policy makers and regulators are too simplistic.
On the opposite end are engineers who double as “economists” who think they know the complex elements of mechanism design (in full disclosure one of my first year profs, Leonid Hurwicz, was awarded a Nobel prize for his work in this area), and complex economic models but know nothing. In this case they can do math, but do not have the sense of social scientists to ask questions and be skeptical. In short, they lack a sense of knowing what it is they do not know.
Your lead poisoning case is one I was aware of and you did a great job of summarizing it. As well as the history of Mr Midgely. Sometimes, the evidence is just overwhelming and we ignore it because we do not really want to believe it. Sometimes we want to make things more complex than they really are, just as much as we make things simpler than they really are.
This also shows the hubris of one man in conjunction with economic greed and incomplete information of large economic actors, can lead to such awful trends as you showed.
Now a related question from this article: the lagged timelines also match up with urban decline that coincided with the worst of pollution seen in the late 60s and 1970s. Did leaded gasoline also lead to flight from the urban centers to the suburbs and exurbs? And conversely, did the phase out of leaded gasoline lead to the renewal of many urban areas up to the COVID epidemic? How are these related to crime that may have been caused by lead poisoning?
Great questions! I don’t know the answer re: the possible knock-on effects of leaded gasoline (pun intended, I suppose). I’m
100% with you on the models that explain nothing. It’s a core reason I’ve started doing a lot more reading outside social science for fresh ideas.
When I was in college in Providence, RI circa 2000, lead removal was a big topic locally. Many residents planted sunflowers on the belief that they sucked lead out of the soil, and the flowers were then sent to a special disposal facility. I lived with environmental science majors, and one of my friends later worked for a nonprofit that educated about lead paint exposure and mitigation. Anyway, regarding this historical narrative, I do wonder about the possibility of correlation rather than causation: after all, the railroads didn't only bring paint, but also industrialization, capitalism, etc. Lots of factors contributed to violence. Then again, as you say, countries wised up to lead and phased it in and out at different times, which helps isolate the variable.
The debate, as far as I can tell, is about how much it's the causal variable, not whether it is one. Nevin puts stock in it as the primary cause; others as a contributing factor. I don't think many people think it had no effect.
Insightful - fantastic research presented in a readable & understandable format.
Let's not ignore Calvin Coolidge and his role. When the petroleum companies panicked over having lead gasoline banned because of its obvious health effects on those who made it, good ol' "no regulation is good regulation" Coolidge formed a study group, packed it with industry reps, and viola, it's no longer a problem so long as the workers wore masks and gloves. As for the effect on the public. "We'll worry about that later."