Hacker News Reader

Comment by JumpCrisscross | Hacker News Reader

It's lead.

Lead concentration in America "rapidly increased in the 1950s and then declined in the 1980s" [1]. There is a non-linear discontinuity among kids born in the mid 80s, with linear improvements through to those born in the late 2000s [2].

Arrest rates for violent crimes are highest from 15 to 29 years old (particularly 17 to 23-year olds) [3]. They're particularly low for adults after 50 years old.

We're around 40 years from the last of the high-lead children. 17 years ago is the late 2000s.

[1] https://www.sciencedirect.com/science/article/abs/pii/S10406...

[2] https://ehp.niehs.nih.gov/doi/10.1289/EHP7932

[3] https://kagi.com/assistant/d2c6fdd5-73dd-4952-ae40-1f36aef1e...

View on HN

It is insane to just confidently assert that the only factor in the decrease in crime is Lead. Treating an insanely nuanced issue as an absolute doesn't make your argument more compelling, it is actually kind of baffling.

Why bother stopping at crime rates with that confidence?

The 1st recorded cases of fatty liver disease and T2D in children were in the 1980’s are have continued growing since - lead must have been protecting children’s health.

Testosterone has been on a sharp decline during this same time period - lead must promote healthy testosterone production.

Debt of all kinds, from the national debt, to household debt, to student loans debt has increased exponentially and consistently with lead removal - lead must promote financial literacy.

If you do the same comparison of the rates of leaded gasoline during childhood to adulthood crime rates across different countries which have different histories of leaded gasoline usage, you notice that the correlation persists. While of course correlation does not imply causation, it's a link that's fairly well-established in literature, it's not a spurious correlation, and we know that lead has concrete neurological effects, so it's plausible from a pharmacological basis.

Since 1970 testosterone has declined 1% per year and it’s well established higher testosterone is linked to impulsive and violent criminal behavior and in countries like the US crime rate is at a 50 year low correlating with this decline starting 1970.

There are many factors that correlate and potentially contribute to a reduction in incarceration rates.

There are estimated 1.8-1.9M incarcerated. Since 1980 to the present there are well over 1M violent crimes (rape, murder, aggregated assault, robbery) per year. Let’s look at another factor that might contribute to falling incarceration rates that tend to explain this discrepancy in incarceration vs total crimes…conviction rates:

Murder: ~57.4% in 1950 vs. ~27.2% in 2023—a ~2.1x difference.

Rape: ~17.3% in 1950 vs. ~2.3% in 2023—a ~7.5x difference.

Aggravated Assualt: ~19.7% in 1950 vs. ~15.9% in 2023—a ~1.2x difference.

The neurological effects of lead don’t tend to explain away falling police clearances nor convictions.

Where are these conviction rate statistics from? What are they measuring? (is it reporting of crime to a conviction on that crime?)

There have been a lot of studies that show the correlation with lead up and down and varied by lead in different cities countries with different phaseout timelines.

Kevin drum and Rick Bevin both did a ton to lay this out systematically.

As leaving drum has noted, Lead is NOT the only contributor to crime, but it was the cause of the largest variations for most of the 20th century.

But it's so satisfying to one's ego that a single cause is the issue. All complexity of societal changes in the last 50 years can be outmanuevered. Simplification is sexy.

It’s satisfying to know that we’ve eliminated a major environmental toxin with so many awful effects. It doesn’t mean that lead explains everything, but it is a lot better than the “we built enough prisons to lock up all the bad guys, maybe we should build more” alternative hypothesis/proposal I’ve heard.

There was a crime decline in many rich countries from the 1990s as well.

https://en.wikipedia.org/wiki/Crime_drop#Decline_since_the_e...

Maybe they were doing similar things with lead or something else is a big factor. Perhaps the rise of ever more cheap entertainment for young males who are most likely to commit crime. That's a global thing.

Yes, leaded gasoline was being banned in many rich countries at about the same time, and there's a positive correlation between the year it was banned and the year that violent street crime began to decline.

So reducing lead exposure immediately changes your brain to do less crime?

Kids that grew up huffing leaded exhaust are more bad decisions inclined than they would otherwise be. It's not just crime. The most heavily leaded cohort in the US is also known for drunkly crashing their muscle cars and wasting their youth smoking pot in a commune.

Bad decisions like these get less common with age, partly because of consequences (jail, death, etc), partly because getting up to no good requires free time, ambition and freedom, all of which are in shorter supply with age and the resultant responsibilities competing for every individual's supply of these resources.

So if the replacement cohort of people who are coming into prime crime age decline to participate at the same rates the crime rate goes down.

No, there's an offset of about 18 years, if I remember correctly?

I see, so since a large majority of crime is done by young people, peaking between 15-25, they are basically comparing a whole new generation of kids who didn't have developmental brain issues vs their elders.

Were the older people who grew up with lead exposure also experiencing higher rates of impulsive crime in the late >1990s relative to the new and prior generations? That would help eliminate the major differences in economics/culture/politics of their upbringing (for ex: mass flight of families moving to the suburbs to raise their young kids after the 1970s crime wave scared them away).

That's an interesting question, and I don't know the answer.

I think lead is nasty stuff, but if it was the single cause of high crime, surely we'd see a similar effect in other domains, like a rebound effect on IQs (another thing lead was blamed for)?

Instead the Flynn Effect seems to have been strongest during the era of high lead, and it's tailing-off now.

I'm not convinced these tests measure what they claim to. Even assuming they do, IQ scores offer little practical value.

The human body and mind are always adapting, however subtly, to changing environments. So I wonder -- are IQ tests assessing abilities that may no longer be optimal today?

Homer likely had an exceptional memory, as did many ancient Greeks that participated in oral traditions. But how relevant is memorizing epics in the modern world?

The only reasonable conclusion is that lead causes crime by making people smarter.

or maybe intelligence doesn't correlate with likeliness to commit crime?

plenty of criminals are intelligent.

below study claims test score variances are mostly related too declarative knowledge side note, i wonder how internet had an effect on iq scores.

https://www.sciencedirect.com/science/article/abs/pii/S01602...

Lack of boredom is also a factor.

Social media and modern games are keeping them occupied.

People also have fewer possessions worth stealing and trying to hock? It's not like TVs and radios cost that much anymore. People wear less jewelry. Though this is not a significant factor, it might be worth putting on the list still.

The most valuable things on a person these days (credit cards, phone) are also incredibly easy to lock down and make worthless. Many of the things like jewelry, are also now rendered essentially worthless because a lot of jewelry now is cheaply sourced; pawning off crap from fast fashion is not going to be worth it.

I was thinking that as I was getting ready to sell my house. I'm not a particularly materialistic person to start with, but there are hardly any physical objects in my home that I value that much besides (a) some photo albums/pictures and yearbooks - and for newer generations these are mostly digital I guess, (b) my violin and (c) my espresso machine and grinder. I guess you could throw my cellphone in there as well - easy to replace but would be a PITA, like losing my wallet. It'd be a pain to replace all my furniture and other stuff but I certainly don't feel any attachment to those things.

I feel you. I’m selling my house and I joke that I’ll give someone a better deal if they just take everything in it as part of the sale. A suitcase for my clothes, my computer, and some physical mementos is all I need to keep. Even the clothes are optional, but I don’t feel like buying a new wardrobe.

My coffee grinder may have been on my list, but I moved countries and the power is incompatible hah.

Bicycles and tools seem to be the main things still stolen. They are often left unattended locked to poles or in the back of cars which can be easily broken in to, and can be immediately flipped for a lot of money.

Right, there has been a huge reduction in home burglaries over the past several decades. The only stuff really worth stealing anymore is cash, drugs, and firearms.

https://doi.org/10.1057/s41284-021-00284-4

That's funny to see. Sometimes I get stressed about the lack of security around my house, but I'll stop and think, if someone broke in what would this hypothetical thief actually steal anyway?

Also, TVs have gotten way larger on the screen size, making them harder to transport in a hurry, and are often screwed to the wall.

I was wondering about this the other day. Do people even steal car radios/amps/subs anymore? When I was a kid in the 90s, having your car radio stolen was typical.

The more modern equivalent has long been the catalytic converter. I don't know how well legislative efforts to crack down on the resale of used catalytic converters has gone though.

https://news.ycombinator.com/item?id=44365162

Let's add an example to illustrate the difference:

Let's say that there is a correlation between the number of flights between London and New York, and the prices of sulfur. The correlation is near perfect.

When your neocortex is working, you ignore it. You can't create any plausible scenario how this could work (it doesn't exist within your latent space) so you don't learn anything from it, it doesn't even register in your brain as anything worthy of notice.

But everybody with the cerebellum only absolutely does learn it. And completely for real, not just as some fun factoid, but as a fact that they know the same way you know that airplanes have wings, and everybody knows it, only you don't.

Then, one day out of nowhere people start buying sulfur. Your questions are met with laughter and mockery "dude, everybody's buying sulfur, are you autistic?". And you don't know, because you haven't even learned the pseudo facts that everybody else bases their reasoning on.

This is only a made up example, but this is exactly how it works.

No it's not. Not entirely anyway.

One thing I've learned in my decades on this planet is that just about never is one explanation for a human condition mostly correct. Lead is a convenient technical explanation that underestimates the impact of upbringing and community.

It doesn't explain a lot of factors of juvenile delinquency that existed for generations before lead service lines or leaded gasoline.

Industry and highways and other high sources of lead pollution were built in the areas with higher juvenile delinquency. Not in rich, privileged areas. I think you can also correlate the rise in violent crime to amount of lead contamination in the soil, some articles claiming down to the city block level.

Which order did these things happen in?

Maybe industry and highways increase lead exposure which leads to crime, or maybe areas already high in crime are cheaper so that's where industry and highways go?

Specifically violent crime. Poor areas had more violent crime. These areas got dumped with more lead pollution. The rise percentage wise correlates almost perfectly with lead levels, with poor (and now polluted) areas rising more than well off areas. And then the drop, where after lead was removed from petrol you get a significant drop, but still less of a drop in areas more lead soil contamination, such of those blocks next to a highway. We know the developmental effects on young brains from lead; it is why we banned it.

Seriously? https://en.m.wikipedia.org/wiki/Correlation_does_not_imply_c...

I'll do you one better. https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis

And abortion access.

Probably not. That played out in the last wave of crime reduction.

Can we blame lead for the US’ electoral landscape too?

I don't think it shifts the red blue much which is probably what you're getting at.

I think it absolutely affects the quality of politicians we get though. The best that a given generation can offer is probably lower if that generation huffed a lot of lead gas. So as they age out and younger people hit peak career and fill those roles things will probably improve a bit.

I was mostly thinking that the politicians all look so incompetent — with some exceptions of course — that if it wasn't real, it'd be a very unfunny joke exactly because of how much caricature-like and exaggerated personalities the politicians have.

You could, if you wanted to misdiagnose the problem.

You'd have more success blaming COVID inflation and the general public's poor education in economics and lack of understanding why eggs were $3.50/dozen. (Today they are $6.00/dozen)

No. Much of the American electoral landscape is still shaped by the systemic remnants of slavery, reconstruction and segregation, and the post-Trump landscape by the cultural trauma of having elected a black president. Although I'm sure all of the lead poisoning didn't help.

i don't know why there are two copies of this comment now, I didn't post it twice.

> Can we blame lead for the US’ electoral landscape too?

More of a pet theory, but voters born between 1950 and 1980, boomers and Gen X, have had a well-documented set of policy preferences.

What if I told you voters born between nnnn-yyyy had a set of policy preferences?

There’s supposedly a cycle of attitude between generations. If your parents are X, you want to be Y. If your parents are Y, you want to be Z. If your parents are Z, you want to be X

Boomers were essentially statistically indistinguishable from Millennials in the 2024 presidential election: https://www.businessinsider.com/how-generations-voted-trump-...

No. You can’t blame lead. There is zero justification for making the average person less responsible for their own worldview and choices in leadership.

Well, that’s the first time I’ve heard anyone explicitly say they don’t want to understand causal factors because it would reduce the ability to tell people they should bootstrap themselves.

What exactly are you claiming?

Your points say old people have more lead, but then you say young people are more violent. That doesn't square with the articles point that incarceration rates are falling.

Reply on news.ycombinator.com