Rationally Speaking #224 - Rick Nevin on "The long-term effects of lead on crime"
Rationally Speaking Podcast- 904 views
- 7 Jan 2019
This episode features Rick Nevin, an economist who is known for his research suggesting that lead is one of the main causes of crime. Rick and Julia discuss: how do we know the correlation between lead and crime is a sign of a causal relationship? Has the lead-crime theory made any successful predictions? And is it possible that getting rid of lead could reduce the crime rate down to zero?
This episode of rationally speaking is brought to you by Stripe Stripe builds economic infrastructure for the Internet. Their tools help online businesses with everything from incorporation and getting started to handling marketplace payments to preventing fraud. Stripe's Culture puts a special emphasis on rigorous thinking and intellectual curiosity. So if you enjoy podcasts like this one and you're interested in what Stripe does, I'd recommend you check them out. They're always hiring. Learn more at Stripe Dotcom. Welcome to, rationally speaking, the podcast, where we explore the borderlands between reason and nonsense.
I'm your host, Julia Gillard, and I'm here with today's guest economist McNevin. Rick is a senior economist at ICF International. He's done research on a bunch of things, but maybe what he's most well-known for is his research on the effects of lead exposure at a young age.
He's written a lot of papers on the topic, including or in addition to a book titled Lucifer Curve's The Legacy of Lead Poisoning. That is what we're going to talk about today. Rick, welcome to rationally speaking. Thank you.
So first off, what, Rick, is the phenomenon that the lead hypothesis was developed to explain, you know, what's the the pattern in need of an explanation?
Well, there are several patterns, but probably one people are most interested in is crime. We experienced a long, seemingly relentless crime increase from the early 1960s through the early 1990s in the US. And then we've seen a precipitous decline since then. And it is well known that early childhood lead exposure affects brain development and lowers IQ, but also affects the brain in many other ways that affect impulsivity and behaviour. And what I found I started working on this in around 1994, is that if you look at the rise and fall of crime and map it against the rise and fall of lead exposure with a twenty three year time lag, it was an unbelievably close that.
And that might sound like a coincidence, but we have I've since discovered exactly the same pattern in almost a dozen other nations around the world. And that international comparison is especially striking because the crime rate started rising much faster and earlier in the US, where the LED emissions started rising faster and earlier. But the US phased out lead in gasoline over the 1970s and experienced a dramatic crime decline over the 1990s. Britain didn't start getting led out of gasoline until really the late 1980s, and they experienced a soaring violent crime rate over the 1990s.
So it follows the same pattern in every country. And there have been a number of other studies by other researchers at the state level, the city level, even at the zip code and suburb level, all of which shows a striking association between variations in early childhood lead exposure and crime rates about two decades later.
And when we talk about lead exposure, are we talking only or primarily about leaded gasoline or are we talking about lead paint or, you know, other other uses of lead?
Well, all that exposure is additive. By far, the two most pervasive sources of exposure in the 20th century were let in paint and let in gasoline. And one of the striking things in my first study that that showed the effect from either source is that we have murder rate data going back to nineteen hundred four from official mortality statistics. And what I found was that the murder rate increased almost tenfold from nineteen hundred up through about 1930 and then came down to an interim low in the nineteen fifties before it took off again and got that close to the peak.
It was that in 1930. And when you, when you mapped the rise and fall of lead paint and the rise and fall of blood and gasoline, you see that it is tracked both of those trends with about a two decade. It was actually twenty one years, over more than a hundred year time period. And a lot of people, when they think about letting gasoline, are thinking only about inhalation, but in fact that by far the biggest source of exposure for their young children is lead contamination in household dust.
The brain is in a critical stage of development at exactly the age when children are beginning to crawl on the floor and engaging in hand-to-mouth activity. And the deteriorated lead paint in a house settles as lead in dust and the lead in gasoline emissions would settle as lead in dust. And that was the primary way in. A number of studies have shown this that led adults is the key pathway for the exposure of very young children. They've gotten their hands in their mouth or wet.
Then they're crawling and picking up the dust, putting the. Back in their mouth and just dosing themselves with this toxic chemical. So there's a few links in this causal chain, right? There's increase or decrease in, let's just say, increase in inlet in the surrounding environment, then causes increase in inhalation or exposure to led some kind of consumption of lead by children that causes changes in their developing brain and that causes rises in violent behavior. So we've talked about correlational evidence, you know, connecting that first node in the chain to the last note in the chain.
Do we have any any evidence for the link going from increase in lead in the environment to brain changes?
Like do we have any evidence showing that, you know, before we get to the actual crime rates, which is an effect, we think of the brain changes. We've seen any evidence for brain changes in children exposed to more lead than other children.
Yes, the University of Cincinnati has done MRI studies that showed striking evidence that children with higher levels of lead exposure have reduced gray matter in the prefrontal cortex, which is the area that that really affects what what they call executive functions. Behavior, planning and impulse control also found that. Early lead exposure in pairs and reduces the development of myelin, which is a substance that that is deposited on the the the neural connections between the connections between synapses. I've heard it described as it's like insulation on a wire.
And as the brain grows, white matter increases. And that's really the myelin being deposited on the the connections between neurons. And a number of researchers have documented that growth as we age and associated with more impulsive behavior among, you know, teenagers and even young adults because the brain isn't fully wired, is connected, has completely as it is as you get older, because that myelin growth continues through the age of 50. But that's particularly important in the late teens and early 20s.
So the impacts that have been demonstrated for lead exposure on brain growth are actually very consistent with what we already know from crime rate data about the ages of offending being in the teenage years and in the early 20s. And the other thing that is particularly striking over the last 20 or 30 years that I've documented is that we've seen a shift in that peak age of offending the the most dramatic declines that we have seen in arrest rates by age have been among juveniles and then substantially among people in their 20s and 30s.
And although the overall crime rate is coming down, it actually would be coming down even faster, except that we are still seeing an increase in arrest rates for older adults. In absolute terms, people in their 40s and 50s are still less likely to be arrested than people in their teens or early 20s. But in temporal terms, people in their 40s and 50s today are more likely to get arrested than they were 30 years ago, whereas there's been an 80 or even 90 percent decline in some crime categories for juvenile arrest rates.
So what you're seeing is that arrest rates are still being affected, even among the oldest age groups by the years before the phase out of lead in gasoline, because that's when those groups were born. So.
I don't yet understand how that how that connects to the thing I'm about to say, but but I reading about the blood and crime hypothesis and potential objections to it, one of the main things that came up was the claim that in the 1990s, the crime rate, you know, in the latter half of the decade and onward, the crime rate went down. But it didn't just go down among people who were born in 1970, mid early, mid 70s or later, it went down among a bunch of different age groups, which seems like evidence against the hypothesis that it's like a cohort problem, that it's like a group of people who are more likely to commit crimes as opposed to some period issue where like there's something about, you know, cities in that era that makes people more or less likely to commit crimes like, you know, a crack epidemic or something like that.
Yes, there is a there was a study that that brought up that point in 2002. And it was it was a good it was a good question to ask. And it was a good study in many ways. But they actually focused rather narrowly on homicide and and then they focused actually on homicide victimization rates by age and.
Oh, as opposed to committing homicide. Yes.
And well, there there is no data that is perfect. And of course, the problem with when you're looking at arrest rates, they might go up from one year to the next, even if the overall crime rate is going down because the police might catch more of the people who have committed crimes. So they focused more on the victimization rate by age. And and they noted that there was a very strong correlation historically between arrest rates and homicide victimization rates by age, that that correlation mean that people who commit homicides are likely the age of the perpetrator is correlated with the age of the victim.
Is that what you're saying?
Correct. Correct. And of course, the most obvious example is street gangs. I mean, that that that was the probably the biggest source of concern, particularly in the 80s and the 90s. But but one of the things I point out in Lucifer curves is that the that correlation has weakened and the percentage of juveniles who are murdered by other juveniles has dropped substantially. And those who are still being homicide victims are more likely to be victims of adult homicide offenders.
But more importantly, looking at that other broader category of arrest rates over a longer period of time, I've found an incredibly consistent relationship because the crime decline has now spread around the world, with most countries having phased out lead in gasoline during the 1980s. And if you look at arrest rates by age, not only is it very clear that we've seen a much steeper decline in youth and juvenile arrests in the United States, you're seeing exactly the same pattern in Canada, in Britain, in Australia, in New Zealand.
So it's it's a very clear pattern that has become apparent since those studies questioned that cohort effect. Another point on that subject is that and we have an sometimes heated debate over incarceration in the United States is going on now. And it's been a little frustrating for me that people have not noticed the same cohort effect in the prison population. The overall U.S. prison population has been declining for the last few years. But people feel like, you know, it should be much lower than it is.
What you know, what kind of sentencing reforms can we enact? And some of those might be very good changes to make, but we should be aware of the fact that we've seen absolutely stunning declines of 70 percent or more in the male incarceration rate for 18 and 19 year olds. We've seen declines of more than 50 percent in the incarceration rates for men in their 20s and the overall and once again, the overall incarceration rate is not declining as fast because the incarceration rates are increasing for people over the age of 45 or 50.
Hmm.
Why would they be increasing like prison? Well, the the latter hypothesis doesn't seem like it would predict that. Well, what what is happening, unfortunately, is that recidivism rates among released prisoners are quite high, particularly among state prisoners. And what is happening is that the percentage of people over the age of 50 that are in prison is increasing in in no small part because they have been to prison once or twice or more during the course of their lives and have racked up another serious offense and have gone back to prison as they've gotten older.
And here again, here again, in absolute terms, the incarceration rate for people over the age of 50 is lower than it is for men in their 30s. But in relative terms, over time, the incarceration rate is actually increasing over the last 20 or 30 years for people over the age of 50 as it has been plummeting for juveniles and young adults.
Were there any predictions that you were or that this theory made when it was first being formulated in the late 90s, early 2000s that were borne out by the data since then? You know, in 15, 20 years?
Yes. In the in the first study, I was anticipating continuing declines in the crime rate because you're looking at roughly a two decade lag. Twenty three years for violent crime. And I found it was closer to 18 years for property crime, which is actually another source of evidence because property crimes are much more likely to be committed by teenagers than violent crime is. But the. The implicit forecast was that there would be considerable ongoing declines in crime rates and especially in arrest rates for younger adults, and that has been borne out in the United States.
My 2007 study that looked at seven or eight different countries around the world. Had the same implicit forecast and then Lucifer curves, I actually show the crime trend up through the last year of my peer reviewed published analysis and then show this stunning, in many cases, 50 percent decline or more in burglary and robbery rates in half a dozen different countries that have tracked what the earlier decline in LED exposure suggested would happen. So it's had better predictive value than any other criminal justice theory I'm aware of.
Is it surprising, though, conditional on the lead hypothesis being being correct and explaining, you know, majority or a large plurality of the variance in crime? Is it surprising that crime continues to go down?
You know, in the 2010s, if the lead, you know, lead was phased out decades ago?
Well well, we've we've always had a distribution of of blood lead levels and there's no lower threshold that any research has suggested below this level. There's no impact. And as blood that levels continue to decline with ongoing progress in reducing lead paint hazards and other sources of lead exposure, I'm I'm keeping an eye on the percentage of children that still have a blood level above five. At least one study that was very carefully done, I think that is also by the University of Cincinnati found that there was an elevated risk of criminal behavior, at least above a level of five will in the 1970s.
Five was an incredibly low blood level. So like a lot of other public health risks, there is almost certainly an interaction between the environmental exposure and individual variations in biological vulnerability that we we might not fully understand. And what that might mean is that among a group of of young children with blood lead levels above 30, you might have an incredibly horrific making numbers up. But just for illustration, it might be 50 percent of those children end up in the criminal justice system.
If you look at children with blood lead levels of 10 to 20, it might be 10 percent of them end up in the criminal justice system. It could be that you're still getting four or five percent of children with blood lead levels of five to 10 that end up in the criminal justice system. And we'll find out how much farther this goes, because we have seen an encouraging, ongoing, substantial decline in the percentage of children with blood levels above five and.
The fascinating thing that is not in my book, but part of my ongoing work is that the juvenile arrest rates are continuing to decline, tracking that earlier decline in the percentage of children above five. And if it continues on this relatively consistent trend that it's been on over the last twenty five years, juvenile restraints could be remarkably close to zero, sometime around twenty, twenty five. We'll see.
Oh, so you're saying that levels of lead have in fact been going down. So it's not surprising, conditional on the hypothesis that we see crime still go?
Yes, we're definitely the level of concern that the CDC has now set and they're likely to lower it in the near future. But they've said anyone above five and I should pause here because as much as as a public health policy, it's extremely important from my perspective. I think we want all children under two, not just under five. Yeah, the the the level they tracked over the years has declined. In the 1960s, a child was not called blood poisoning unless they had a blood level level above 60.
Well, then it dropped that it was worth it. Then it dropped to 40 when they had large scale screening programs in several major cities in the early 1970s, almost a third of the children tested were above 40. Well, then it dropped to 30 and then to twenty five and then 10. And now it's at five am likely to go lower as all the evidence continues to show, there's no lower threshold where there is no impact at all. But I am a little bit concerned when you hear I read news stories and several other prominent researchers have noted the same thing in published reports, that if you are in a situation like we've had in Flint, where children are described as lead poisoned and are being given the impression that they are irreparably damaged because they had a blood lead level above five, well, the first national survey of children's blood lead levels in the late 1970s found that ninety nine point eight percent of all children had blood lead levels above five.
Huh. So I don't want anyone I don't want any child to have a blood level above five. I don't even want anyone to have a blood one level above two. But I don't want to scare the living hell out of parents or children who have one blood test of six or seven, because a lot of people have done just fine after having that experience.
And that is a really good point. I'm actually surprised, given that I forget what you said this was, but the time when a large proportion of children were tested above 40. I'm surprised the crime wasn't even higher than than it actually was if that was the level of exposure.
Well, yeah. And that was in cities. And the air levels in cities were much worse than they were in suburbs. And this is another part of a piece of the evidence that supports this theory. The murder rate declined since the 1990s has been especially steep in the largest cities. And we have the AirLand data from 20 years earlier showing that the greater traffic congestion in those cities naturally led to much higher ambient air lead levels in large cities than you ever saw in suburbs.
So you have the concentration of crime that has been in cities as opposed to suburbs and rural areas in decades past was also consistent with the much higher air lead levels that had been recorded in cities 20 years earlier. And we've seen the decline go down much more steeply in those areas where the airbed levels have come down the most steeply. This also has had very important racial impacts because black children were disproportionately concentrated, first in dilapidated cities, slum housing that had horrible lead paint roofs, and then in cities in general where the Araba levels were higher and then in particular in public housing projects that we unfortunately built in many cases right beside new highways, where the near fallout from the street was much worse than the overall ambient air lead levels.
Yeah, and this led to especially high arrest rates among black juveniles in the 80s and early 1990s.
And did we see those especially elevated arrest rates near the highways or are we talking about in general among black people?
Well, you don't have you don't have any systematic data. But in my 2007 study, I specifically noted the notorious housing project on the south side of Chicago that was eventually torn down and it had been built right beside. I think it's the Dan Ryan Expressway that is one point is 16 lanes wide. Wow. And I just I described what we know about how much more severe the near fallout is. And by the time they started emptying out that project and just decided to tear it down, it was accounting for I forget what the percentage was, but it was a shockingly substantial percentage of all of the murders in Chicago were occurring in that housing project.
So there's anecdotal evidence that is very consistent with near road exposure being especially severe and affecting not just the children, but, you know, children are especially affected by blood and dust. That's their main pathway. But they're also affected before birth and through the mother's blood type level. And pregnant women in those housing projects were breathing a horrific amount of lead in and out right all the time that they were pregnant. So all of the anecdotal evidence continues to line up with that.
And as an aside, people are generally unaware of the fact that black Fellaini black juvenile felony interest rates now are less than one half of what the white juvenile felony interest rates were in the early 1980s. And you can see the much steeper decline in black juvenile interest rates over the last 20 or 30 years. That is, again, completely consistent with the much steeper decline and more severely elevated blood lead levels among black children 20 years earlier.
So the idea is that exposure to lead was greater among young black people than among young white people. And therefore, reducing lead in the environment produced a sharper decline in crime among young black people than white people. Exactly, yeah.
OK, and that that that disparity, that racial disparity in LED exposure and especially in sports, severely elevated blood levels has been very well documented since the 1970s. So we know that happen.
So at the beginning, toward the beginning of this conversation, I was asking about the link in the causal chain between exposure to lead and changes in the brain.
And you talked about some of the neurological evidence we have for for brain changes. But do we have any any other evidence in the population of behavior like. If the reason that lead increases crime is because it makes people, you know, more impulsive, give, gives them lower impulse control, etc., do we have any evidence of of trends that would also result from low impulse control, et cetera? Like, I don't know, people going into more debt or something like making worse financial choices, that kind of thing.
There are two, actually. I mean, the phase out of of lead in gasoline was actually the cost was justified by the anticipated benefit of better educational achievement and higher lifetime earnings as a result of that. And some a researcher named Joel Schwartz, who actually won the MacArthur Award, I believe, for his work on this subject, put together this perspective that there was a strong association between IQ levels and education attainment and subsequent lifetime earnings. And there is a lot of research even as early as the 1960s and especially in the 70s, that showed that LED clearly had an impact on IQ scores and putting those two together.
He said if we if we reduce childhood lead exposure, you're going to increase education attainment and and significantly increase lifetime earnings as a result. Well, one of the things I've looked at, it's not as easy to map or analyze statistically, but the trends are pretty clear that we had significant increases in high school graduation rates up through the 1950s at sometime in the 1960s, about 20 years after the AirLand levels took off after World War Two, the progress in educational attainment mostly stalled for almost 20 or 30 years and in some cases went down.
What we've seen going on in education trends over the same years that the juvenile and young adult arrest rates have plummeted is that the high school dropout rate has fallen to an all time record low and college enrollment rates have risen to an all time record high. So that by itself doesn't prove the association, but it's exactly what we would expect to see. But the other relationship, the other relationship that is really stunning and people have a hard time wrapping their head around is that in my 2000 study, some research had a number of studies that already shown a link between IQ and criminal behavior and incarceration rates.
But another study had shown that there was also a clear link with unwed birth rates. And at the same time that the crime rate rose, the unwed birth rates and particularly teenage unwed birth rates really went up from the 60s all the way through the early 1990s. And my study in 2000 tracked I looked at both unwed birth rates and abortion rates and found that it was exactly the same pattern as crime and that it had risen. And in fact, the time lag fit with which?
With the age group you were looking at that the young the birth rate for girls under the age of 15 seemed to map at a 14 year lag for girls 15 to 17 HapMap map that a 17 year lag for 18 and 19. It mapped at a 20 year lag. And we've seen a massive, stunning decline in in teenage birth rates in general, unwed birth rates in particular, and and in abortion rates. And it's the same pattern if you look across age groups that you see in the crime and incarceration rates, that the over 40 is the only group where unwed birth rates or abortion rates are still increasing even slightly, that you've seen the massive decline in the younger age groups.
And I'm currently working. I've got the data, but I haven't published anything on a study that shows exactly the same association, the same trends in unwed birth and abortion rates in Britain and in Canada linked to their their lead exposure trends.
There's one thing I wanted to ask you about in your book. You talk about how people back in the eighteen hundreds were already worried about the effects of lead exposure. I guess it was higher lead exposure than than just, you know, inhaling lead dust near a highway or like kids working in lead mines.
But so I was wondering first, can you give an example of of concerns raised, you know, back in the 19th century about lead? But then second, if we knew back then that lead has had these negative effects on people's brains, why did we suddenly start using it widely and in the mid 20th century?
Well, there isn't a good answer to that question, it was a horrible mistake, obviously, and it's. It's a cautionary tale because people today think, well, you know, I said ninety nine point eight percent of children had blood lead levels above five in the late 70s. So how can you be at all worried about any percentage of kids being above five? Well, it can still be affecting some part of the population. And in the eighteen hundreds, what happened is that they actually would refer to the lead trades and lead poisoning in the encyclopedia was called a disease of occupations, and they were especially aware of effects on painters and typesetters had horrible exposure.
And there were other professions in the early decades of the Industrial Revolution where light exposure was widely recognized as a serious occupational hazard for a variety of workers. And we were talking about people dying. We weren't talking about losing a few IQ points. There was no understanding at that time of how this might be affecting very young children and they would be affected both by industrial emissions. There was increasing concern about the paint lead levels, and France actually started reducing their paint lead levels in the mid eighteen hundreds or actually around 1840.
And I found crime data from the eighteen hundreds that showed a steep decline in French crime over the eighteen hundreds. And a similar shift in arrest rates by age in Britain to what we're seeing now, that the term juvenile delinquency actually was first used in the early eighteen hundreds at the dawn of the industrial revolution, about 10 to 20 years after there had been explosive growth in the number of patents for lead paint and the production of led to be used in lead paint.
And then over the eighteen hundreds, Britain caught up with France phasing out the lead in paint. And they saw not only a significant decline in British crime or the latest hundreds, they also saw a shift from juvenile delinquents to much older offenders. That's just recently been documented in a study I cite in Lucifer. So you're seeing this pattern in hindsight, going back all the way to eighteen hundred. But when they first added led to gasoline, there were people, prominent public health advocates who were trying to block this.
They had been working to reduce blood in pain and trying to make people more aware of the hazards to workers from lead poisoning. And there were people who when they were discussing this. Possibility, we're saying if this gets to a point where there are a lot of cars and this is being spewed everywhere, it just seems like you could have some very profound, pervasive effects. And, of course, those warnings turned out to be right. And to give you an idea of how.
Little regard they had for the warnings that they were facing at that time. The tetraethyl lead manufacturer, when they first started producing this to reduce engine knock in in with the gasoline led additive. They had they called it loony gas at the plant where it was made, a number of workers died, and so a lot of others were taken away in straightjackets. Oh, my God. And it was pulled off of the market lead. Additives were pulled off the market for gasoline for I think it was a year or two.
And then there was, you know, corporate pressure. And the attorney the the federal government was involved and they announced, OK, well, it was just they didn't have the proper ventilation in the plant. It's not going to be a problem in the gasoline and leaded gasoline was back and that was the beginning of the long rise in their blood levels.
Are there any specific villains that we can you know, I want someone to be mad at, like a person who made this decision there?
There are a lot of good candidates that and I know that there have been other books that are written on that subject, but I haven't really focused on that. So I'd rather not call them up by name. The five that one of the one of the leading scientists working for the oil companies and OK, tell them the manufacturer of the of the lead additive. He was they tried to get him involved in a defense of lead paint companies, and he refused to to help them defend that case because he said that he had seen evidence of children in the 1940s when they first recognized the link between lead poisoning from paint chip ingestion and other sources of light ingestion in children and the risk of what is now called intellectual disability that was discovered in the early 1940s.
And the same scientist who was saying, yes, I've seen this effect myself. It's a serious risk. I can't defend the paint companies. I think he was the same scientist urging the use of blood in gasoline so people can be blind. Well, I hope I have that story correct that I should have emphasized again before I got into it that this is not where I focus my research. I, I'd rather stick to the root of the exposures we still have.
There are a lot of other people out there who have sued the paint companies and I wish them well, but that's not where I can probably do the most good. Understood.
How has the reception to the lead hypothesis been? And and also, what do you think it would take to get sort of mainstream consensus on board and taking action in response?
For many years, people it was just largely ignored and there was no I got a very good article in the Baltimore newspaper when my twin study was published. I thought, sure, it would generate another call and some more interest and it never did. The Washington Post did a great article after my 2007 study was published. Oh, yeah. They actually they actually had a front page article in The Post two years later about how crime was continuing to decline and no one could explain why.
And you're over there waving or telling you, hey, I actually I called the reporter who wrote the story on my research two years earlier and said if he'd seen this, he said, yeah, I talked to them before they published. I sent my earlier study. They just decided to go know why.
And I've had more luck in the last few years, thanks to Kevin on it, who did a cover story, and Kevin not only wrote the cover story on this, but as a daily blogger, he's continued to follow up on the and and directed people to all of the different research. But he's also posted a blog post on how they had tried to interest The New York Times or the L.A. Times or other newspapers into. You know, showing this relationship that he would do a shortened version of the article he did for Mother Jones or update things, and for some reason they're not interested.
It's just rather frustrating. And it relates, as I say, not only to the lead hypothesis, but to clear effects of this that you would think people would be acknowledging. If we're having the debate we're having today about mass incarceration, someone should be paying attention to the fact that we're already seeing a 70 percent decline in the 18 and 19 year old incarceration rate and that that is going to roll through the prison system over time. I mean, even simple capital expenditure decisions, I forget if it was Alabama or Mississippi last in or last couple of years, have been debating a massive spending program to replace aging prisons.
And it would be financed over a 30 year period of time with 30 year bonds. And no one has looked at what's happening by age group in prisons to know that they weren't. There's no way they're going to need that much prison capacity in 30 years.
I wonder if so you would think at least those facts would be just this from the self-interested financial perspective.
So I wonder if it would be helpful to make some concrete predictions over for, you know, the next five to 10 years, say that are like, you know, probabilistic predictions about a bunch of things that like if you can demonstrate, like strong predictive power of this hypothesis that like in five to 10 years, that might I mean, especially if you could get people to agree ahead of time to like. Yes, if those predictions come true, then are like some percentage of those predictions come true, then like we will consider that like strong evidence for the latter hypothesis.
Well, the funny thing is it's still the same prediction, and I can't I can't tell if there is some baseline crime rate level. Yeah. That would exist with zero light exposure. I'm starting to think that there might not be really you can the most. I mean, there is crime. The fallout is to zero.
Like the truth is, lead paint started spreading in Europe in the seventeen hundreds. I don't think we have data on anything. Huh. In history, without lead exposure and but there was violence at least, right, like I mean, Steven Pinker's talked about rates of violence well before we had led paint. We might not have called it crime back then, but, yeah, I don't know, I in or certain places in the world, although. This is getting into material I haven't published, but.
I've actually come across a couple of historical things about, you know, a period of accentuated warfare that couldn't be explained among Native Americans in the southwest United States, and for some reason, there was horrifically violent warfare that didn't seem to compare with anything at other times or in other parts of the country.
And. I've come across the evidence that Native American tribes in that area were they discovered, led Cirillo's mountains and there were incredibly high lead content levels in these ceramics that they ate and drank from that interest. So it's it's I keep seeing it over and over again. And I don't think we've, you know, in an era of data and this is particularly relevant to IQ because the very first IQ tests were created at the beginning of the 20th century, around nineteen hundred.
The Binay test in France was in 1994, and that was one of the first and most significant advances in IQ testing. And of course, France had already dealt with what they had understood was a dangerous increase in leading paint, you know, starting in 100 years before that test. Well, now people look and they talk about how does lead exposure affect IQ? I'm starting to think we should be asking a more profound question. What does IQ even measure in the absence of any lead exposure?
I'm not sure we know. I know for certain that these studies that have shown an incredibly strong correlation between IQ and education attainment and unwed birth rates and criminal behavior. Well, as you see, particularly among juveniles, those rates plummeting and over twenty five years establishing a trend line that, as I said before, shows them getting close to zero in twenty twenty five. The way that they calculate IQ is always relative in relative terms, so. Some percentage of the population will always be under 75 and some percentage will be between 90 and 110, et cetera, but it's not going to mean what it used to mean, that there are areas in the population that, well, you might you can still come up with.
This person has a very low score in relative terms, but in in absolute life terms, how much does that affect their potential? I don't think anywhere near as much as it did in the 1960s or 70s when what you were really measuring was the biggest part of what you were measuring with IQ tests was how much how severe was that early childhood lead exposure for this child?
Does that model fit with the apparent hereditary ness of like, you know, significant hereditary ness of IQ?
Well, one of the things I've pointed out is that there are two trends that have been well documented in IQ. One is that this is mostly done on siblings that are separated and especially twins. That's the ideal data. If they are separated very early in life. And then you you find that the genetic siblings that that never knew each other after the age of one, their IQ tend to be closer than either either of them are with their adopted rights that grew up in the same household.
But when you understand the impact of lead, exposure is very substantial before birth through maternal blood, and that the peak period of ingestion is when children are learning to crawl around the age of six months. And I've looked at some of these adoption studies. It's very often children who were adopted after six months around the age of one or older.
Oh, interesting. So I just so people are assuming, well, they have nothing in common from their birth life. Well, no, they they crawled on the same floor just in the same lead dust and shared the same maternal blood blood. And furthermore they could inherit the same biological vulnerability to lead exposure. Right.
So it's modifiers. Yeah.
You might have, you might but that would be irrelevant if we eliminated lot exposure. Interesting. So it's an intriguing. The other trend, though, is that a guy by the name of Flyn actually discovered comparing IQ test norms over many decades. And I think more than two dozen countries around the world has shown that IQ scores have been rising substantially for more than a century everywhere in the world. Right. And the way they calculate IQ obscures that rise because every new IQ test, your IQ is calculated relative to the other people in the norm and sample for that IQ test.
But when they have a new IQ test, they also give people the older IQ test to prove that it's consistent that the new IQ test is scoring people as having high IQ. If they had high IQ on the old test and low IQ if they had low IQ on the old test. What Flynn discovered is that when you look at these Naum comparisons, as they're called, the new IQ sample always or almost always scores above 100 on the old test on average and 100 by definition was the average for the people who were the norm sample for that test 20 or 30 years earlier.
So that means IQ is actually increased. And in my 2000 study, I pointed out that a lot of the data that Flynn had was consistent with the steep declines in lead paint exposure that would have occurred in Europe starting in the mid to late eighteen hundreds and in the United States starting around nineteen hundred. So I think there are really and people have never understood how to have never really been able to explain the side. And they have different speculative ideas about that, like maybe our education system is teaching analytical skills more or something like that.
Yeah. Yeah. But that that is interesting.
One other question that I meant to ask earlier but forgot is, do we have. Have there been any attempts to exploit, like, pseudo random variation in looking at these correlations? Like obviously we can do a real randomized controlled trial where we assign some children to grow up lead and others not. But but sometimes economists get very clever about exploiting these kind of natural experiments. Well, one of the one of I think the most striking natural experiments is the fact that I mentioned earlier that the the US increase in lead emissions after World War Two was much greater and much earlier than it was in Britain and other places that were struggling to recover from the devastation of that war and had nowhere near the number of automobiles and nowhere near the gasoline consumption that we had in the 1950s.
And then the fact that the phase out of luddin gasoline occurred at different times, sometimes 10 or more years apart in different countries. Hmm. And then you see that you've got the same time lag explaining the rise and fall in crime in every one of the countries that I've looked at and and predicting another dozen years of steep declines in those countries. But it's the lag related to the rise and fall of lood within that country. So that's that's a pretty striking natural experiment that no one planned on that confirms the relationship overall.
And the other thing is that the other thing is the change by age that I think is is the most striking that we're still seeing such steep declines, ongoing declines in juvenile restraints. Cool.
Well, that's probably a good place to wrap up. But before I let you go, Rick, I wanted to ask you to nominate a book or article or website or something like that that that either had some significant influence on your thinking or that you consider to be like a great representative of of your field, like a really well conducted experiment or, you know, well argued or something like that.
What have you got?
Well, I think this is not really about lead poisoning directly, but the there's a book called The Rising Curve, and it was edited by a very well respected academic, by the name of Oelrich Nicer, who passed away a few years ago. And when the book The Bell Curve came out, which was extremely controversial, and it came out in the early 90s and talked about how IQ affected crime and unwed births and education attainment. And, you know, he cited they cited the research showing IQ was inherited.
Ulric Neisser was the person that the American Psychological Association chose to be the chair of the committee that produced a paper called Intelligence Knowns and Unknowns to try and deal with the the extreme controversy created by the bell curve in the course of writing that paper. He found that the research on the Flynn effect and related trends on the narrowing of of black and white achievement score differences in school were two really fascinating subjects that he wanted to look at more. And James Flynn was the author of one of the chapters of that book.
He also had other distinguished academics authoring. And this is of The Rising Curve. So after the publication of The Bell Curve, The Rising Curve was published, got nowhere near as much attention as the bell curve did. But I think is a really interesting piece of background to provide perspective on other things that you might have heard about IQ. And particularly having noted that and I've been lucky enough, both James Flynn and Alrick Nicer had traded emails with me when I was first doing this work around town and encouraging me because the mystery had not been solved and they thought I might be on to something.
So I would strongly recommend the rising curve as a good example of a really rigorous and principled academic thought, including an acknowledgement of we can't really explain this yet. You know, in the case of James Flynn in particular. Yeah, Flynn had documented gets more credit than they actually call rising Tykerb trend. The Flynn effect now. And right. Even though he was credited with discovering it, he makes it clear in the rise curve that he doesn't think that any of the other explanation, any of the explanations for it are satisfactory, that there's something else missing.
I'm convinced that it's led, but.
Well, that's that sounds like a great recommendation in parts personally, because I. I have a weakness for book titles that reference earlier book titles as a response to that.
Like like there is a pair of books I forgot to the authors, but one was titled How the Mind Works. Then the second one was titled The Mind Doesn't Work That Way, I think. So now we have the bell curve in the right.
That was actually the problem to my left. That was the inspiration for the title of Lucifer Kerbs because.
Oh, really? So you have a trampoline in the first chapter that there's actually a relationship between the lead contamination in the atmosphere and the origin of the word Lucifer. And what I'm suggesting. Right. What I'm suggesting is that both the the IQ bell curve and the rise and fall of violent crime that I have, I have both of those on the cover. Those are both impacts of lead exposure and it's following on the bell curve and the rising curve.
So now you've got the word Lucifer.
Yeah, that's perfect. Tell me about the word Lucifer. Where does that come from?
Well, it's just a fascinating tidbit that I picked up that I realized I could use that the word Lucifer is commonly thought to be another name for Satan or the devil, but. Right. But biblical experts say this is actually a mistake of of of ancient biblical translation that the word Lucifer refers to Venus as the morning star and in the Latin version of the Bible that Lucifer or however the Latin word was phrased, was used in many references to Jesus Christ when that were then translated into the English Bible as the rising star or the morning star in references to Christ, there was an Old Testament reference to a Babylonian king who likened himself to the Babylonian God of of Venus.
And it was a ridiculing commentary to call him Lucifer. And for some reason, when it was the Bible is translated into English, I guess the King James version, that was the only place of the word Lucifer was kept. And I pointed out that the reason that Venus is so bright in the sky, the reason the morning star is so bright, we now know from space expeditions, is because of its incredibly dense, incredibly toxic atmosphere that reflects light.
And the heat on the planet is so high that it actually melts, led on the surface and that leads settles on the mountains as a glistening reflective shield. So I point I pointed out as a new perspective on crime that Lucifer isn't actually the name of the devil and doesn't even represent anything inherently evil. It just refers to the the the impact of an extremely contaminated, toxic led environment. So that's why I. I love some some good etymology, epidemiology, public health.
Well, thank you so much for coming on the show. This is fascinating. We'll link to the rising curve and the curve, as well as to a couple of your papers on the topic for us. Thank you.
This concludes another episode of Rationally Speaking. Join us next time for more explorations on the borderlands between reason and nonsense.