A serious GenX divide

When Freakonomics was originally published, it used data to show that crime numbers spiked during the early 90s before declining and they posited that it was a function of Roe v Wade. That “unwanted” children would grow up to do crime, and that a reduction in unwanted children led to a future reduction in crime as that cohort was smaller when they reached the crime-doing age.

But in later years, the correlation was found to be wrong. Because it didn’t track to other countries who banned or legalized abortion. There weren’t similar echoes in the crime stats 17-25 years later.

Instead, an incredibly accurate correlation was found between crime stats and when leaded gas was banned in those areas. Different countries banned lead in gasoline at different times. And in basically every country surveyed, charts like the one here are replicated. As car usage climbed, crimes committed 17-25 years later climbed. (More cars, more exposure to the lead fumes). Then when leaded gas was banned, a decline begins 17-25 years later. Repeated over and over throughout the world. Lead exposure in very young children leads to impulse control issues in young adulthood. Lead exposure is bad at any age. But it’s worst for toddlers.

[-]

wyohman@reddit

I think you've missed the follow up that Levitt did a few years ago that showed the link between abortion and crime to be stronger (he also includes crack, more police, increased incarceration). I also read the paper published by Jessica Reyes. She also mentioned the Levitt study when she was on Freakonomics.

I think that all of or some of these can be true. No matter what, the negative effects of lead have enough science behind them to consider minimizing lead exposure regardless.

[-]

Kershiser22@reddit

We don't know for sure that they are unrelated.

[-]

wyohman@reddit

Just consider the volume of leaded gas and leaded paint used 50 years ago to the volume of leaded AVGAS used today.

This is a simple correlation been lead levels in the blood of children over the years v. violent crime. It's an interesting graph that shows a correlation. This correlation was be relevant but it's hard to day without reading the underlying study.

[-]

BWWFC@reddit

avgas is minuscule, unless living/school is near. graphic sucks for "info transmission" though.

[-]

wyohman@reddit

This is likely someone pulling a graph out of something that provided the appropriate context. I ignore almost all graphs because of this.

I read this paper a few years ago and it is quite interesting.

[-]

Lampwick@reddit

avgas is minuscule, unless living by or going to school near

Even then, AVGAS is infinitesimal compared to millions of cars using tetraethyl leaded gasoline from the 20s to the 70s.

[-]

SXTY82@reddit

Thank you. I just spent a few minutes trying to figure out what this graph was representing. It seems to show violent crime rising from 66' to 90' if you only follow the blue line. Corrected it a bit because it was driving me nuts too.

[-]

Magnum-3000@reddit

This is comparing two things with similar curves and sliding one timescale 20years so the curves line up in order to show correlation….and try to claim delayed causation. It doesn’t make sense to you because it is stupid.

[-]

Gavin_Tremlor@reddit

Sometimes it is though.

[-]

Large-Raise9643@reddit

I am not saying it’s bad data, I am saying don’t jump to conclusions.

[-]

BWWFC@reddit

how to "lie" with statistics and graphics lol

[-]

Numerous_Bad1961@reddit

You should read “Murderland.”

[-]

DarenRidgeway@reddit

Ah yes. It also corrolates to the rise of video games... so the more realistic the simulated violence got, the less actual violence took place. Sounds like we need simulators for all crime as realistic as possible then... it's good for society... right?

You can literally do this crap all day.

[-]

Frosty-Escape-4497@reddit

Yes, there are a few divides between the early Xers and the late ones.

Early Xers are not as financially sound as the younger ones.

[-]

deludedinformer@reddit

Correlation does not imply causation

[-]

smapdiagesix@reddit

Okay, so I'm having a little bit of a grumpy morning.

THIS DOES NOT MEAN WHAT YOU THINK IT DOES, AND USING IT AS THE AUTOMATIC GAINSAY OF SCIENCEY STUFF YOU DON'T LIKE IS SUPER DUMB AND LAZY.

Of course correlation implies causation. That's why we look for causation by using... correlation. Double-blind controlled trials like you see in medical testing are still, at the end, correlative. Being in the treatment group is CORRELATED WITH better/faster outcomes.

The thing is, and this is where the saying comes from, the world is a messy place and lots of things are correlated with each other just by stupid fucking happenstance. So. You can't just look at a correlation and say you've found causation; you need a fucking theory. Then, you need to pull out observable implications from your theory that aren't the original correlation that inspired you. Then go look at all those correlations implied by those observable implications.

In the case of lead and crime, this isn't rocket science. If it were the lead, then you'd see places that reduced their lead first also reduce their crime first. If it were the lead, you'd expect to see places that never had as much lead see a slower, smaller rise in crime rates and slower, smaller reduction after lead was banned than you would in places where kids had high blood levels. If it were the lead, you'd expect at the individual level to see high blood levels as a child correlated with more violent behavior as a young adult and expect it to be correlated with some kind of observable brain deficit, all ceteris paribus. And I'm too fucking lazy to go look this up but ISTR that these associations mostly check out.

[-]

deludedinformer@reddit

What about the theory posited in Freakonomics around abortion being legalized in the USA

[-]

smapdiagesix@reddit

It can be los dos. [homerfindinga$20]Complex phenomena arising from the interactions of millions of humans over decades can have multiple partial causes![/homerfindinga$20]

[-]

MhojoRisin@reddit

Correlation implies causation even if it doesn’t prove causation.

[-]

_Brandobaris_@reddit

Hey, they’ve got their one thing to argue against things they can’t understand, don’t take that away! /s

[-]

Jordangander@reddit

Graph doesn’t even remain internally consistent to show that, each bar represents completely different years.

[-]

pwaltman1972@reddit

This excellent response explains why the years are offset:
https://www.reddit.com/r/GenX/comments/1sk7r1q/comment/ofxdra6/

[-]

bigtime_porgrammer@reddit

This one isn't even internally consistent, so it doesn't even pass the correlation test very well. Why are they so aligned and then diverge so much?

It also doesn't pass the common sense one very well either. Why would the violent crimes curve track the preschooler lead level curve so tightly on the way up? Are preschoolers doing all the criming? "Hands in the air, pal, this is a stick up! Now give me all your applesauce."

[-]

marle217@reddit

The blue line and the green line are on scales 23 years apart. Lead causes developmental delays and other problems, so years later kids who had high lead exposure are more likely to be involved with crimes.

They don't line up exactly because lead isn't the only cause for violent crime.

[-]

GenX-ModTeam@reddit

Misleading Content - Misinformation, disinformation, anti-science, pseudoscience, or any other misleading posts/commentary will be removed.

[-]

Blurghblagh@reddit

Was shocked to recently find out leaded petrol was still being used until 2021, the last country to end its sale being Algeria.

[-]

activelyresting@reddit

I'm a younger Gen X, but my parents owned a petrol station when I was growing up, so I definitely got way more than my fair share of lead exposure. I was getting pocket money filling cars, checking oil, washing windshields, bleeding brakes etc from about age 8

[-]

Message_10@reddit

But how are you when it comes to violent crime at 100,000 people

[-]

activelyresting@reddit

Isn't there some law protecting people from answering questions that might incriminate themselves?

[-]

charliefoxtrot9@reddit

I'm fine at 99,999. But one more...?

[-]

Message_10@reddit

Alright well just take it easy, you're doing great. Hold off on that last one and you're OK ;)

[-]

Magnum-3000@reddit

Murders increase each summer. I’ve cream consumption increases in summer. It’s clear ice cream causes the murder rate to go up.

[-]

LanguidLandscape@reddit

Man… I’d KILL for an ice cream right about now!

[-]

grin_ferno@reddit

RFKjr says we should all use more lead! We should listen because as he's bragged before, he used to snort cocaine off toilet seats!

[-]

merlot2K1@reddit

I got dumber looking at this stupid graph trying to make a correlation between different data sets. I bet we can find some horsepower dyno graphs that would fit in there just fine.

[-]

Edan1990@reddit

Per this data shouldn’t there be a 10-20 year lag between lead levels in children and violent crime? Or is it implying that toddlers full of lead instantly become violent criminals?

[-]

SXTY82@reddit

The years are just misaligned. 1970 lead starts to be removed. 1990, crime starts to drop.

[-]

Magnum-3000@reddit

It’s clear from this data that children growing up caused a reduction in lead across the board. Don’t you see the correlation?

[-]

worrymon@reddit

It's a 23 year lag. The green is lead. The green's axis is birth year. The blue is crime. The blue's axis is crime year. The blue number is 23 years after the green number above it.

[-]

shotsallover@reddit

This such a poorly made chart.

It’s not like pixels are expensive.

[-]

worrymon@reddit

I'm not defending the chart, just telling the person how to read it.

Personally, I think the GenX divide is big GI Joe/small GI Joe...

[-]

SXTY82@reddit

Chart is a mess.

I 'normalized' it by pulling the data off of it and realigning it in Excel. I cut the crime per 100K people to 10k people to get the scales to more closely align so that you can really see what was happening. I aligned the years as well. No data was recorded as zero. Zero is 'Empty Set' not zero on this chart. You can see that lead levels dropped in the 70s and 20 years later, crime dropped as well. Clearly a generation which indicates that lead was in fact a contributing factor to the crime rates. Doesn't prove it, indicates it.

[-]

blackfarms@reddit

The decline in violence also correlates to declining Testosterone levels.

[-]

JJQuantum@reddit

You guys went to preschool?

[-]

alicecuriouser@reddit

Does anyone remember this commercial? It's been embedded in my brain, I had to find it recently to show my daughter the kind of horror we were just casually fed.

[-]

Numerous_Bad1961@reddit

This was misinformation from the Lead Industries Association (oil and gas industry, smelters, mining, pigment manufacturing). They knew they were polluting our air, soil and water. They used this to blame children and low income families to distract everyone else from the toxins they were dumping into the environment.

How the lead industry misled the public about its toxic problem for decades using racial bias

[-]

Significant-Dance-43@reddit

Please tell me you found this on the Spurious Correlations website. Correlation is NOT causation folks.

This is some real Boomer shit you’ve posted. Here let me do some other stupid spurious correlations.

[-]

Significant-Dance-43@reddit

[-]

Significant-Dance-43@reddit

No love for Vermont. It’s being downvoted. No… not Vermont!

[-]

Significant-Dance-43@reddit

Dear God… Vermont. Why do they hate you?

[-]

smapdiagesix@reddit

Okay, so I'm having a little bit of a grumpy morning.

THIS DOES NOT MEAN WHAT YOU THINK IT DOES, AND USING IT AS THE AUTOMATIC GAINSAY OF SCIENCEY STUFF YOU DON'T LIKE IS SUPER DUMB AND LAZY.

Of course correlation implies causation. That's why we look for causation by using... correlation. Double-blind controlled trials like you see in medical testing are still, at the end, correlative. Being in the treatment group is CORRELATED WITH better/faster outcomes.

The thing is, and this is where the saying comes from, the world is a messy place and lots of things are correlated with each other just by stupid fucking happenstance. So. You can't just look at a correlation and say you've found causation; you need a fucking theory. Then, you need to pull out observable implications from your theory that aren't the original correlation that inspired you. Then go look at all those correlations implied by those observable implications.

In the case of lead and crime, this isn't rocket science. If it were the lead, then you'd see places that reduced their lead first also reduce their crime first. If it were the lead, you'd expect to see places that never had as much lead see a slower, smaller rise in crime rates and slower, smaller reduction after lead was banned than you would in places where kids had high blood levels. If it were the lead, you'd expect at the individual level to see high blood levels as a child correlated with more violent behavior as a young adult and expect it to be correlated with some kind of observable brain deficit, all ceteris paribus. And I'm too fucking lazy to go look this up but ISTR that these associations mostly check out.

[-]

pwaltman1972@reddit

thank you for the excellent response

[-]

GK8888@reddit

Not a fair comparison. You are using the same x-axis scale for both data sets.

[-]

Majik_Sheff@reddit

OP's graph has the same scale on X, just shifted 23 years. Unless you were expecting violent infants.

[-]

LoanDebtCollector@reddit

violent infants

Stephen King book or a edgy punk group?

[-]

shadowknight2112@reddit

Yes

[-]

alchebyte@reddit

hard mode spurious

[-]

knowlessman@reddit

https://www.sciencedirect.com/science/article/pii/S0166046222000667

[-]

nietheo@reddit

https://scienceblog.cincinnatichildrens.org/long-term-study-documents-link-between-adult-crime-and-brain-damage-from-childhood-lead-exposure/

[-]

Majik_Sheff@reddit

Lead and other heavy metals have been demonstrated time and time again to cause developmental delays/ permanent reduction of reasoning and impulse control. Especially in young children. Ya know, the things that keep healthy people from behaving in destructive and antisocial ways?

Being an industry apologist and advocating for the destruction of our environment because "the economy" is the real boomer shit.

[-]

You_Must_Chill@reddit

You're dismissing the possible relationship... because it looks like their might be a relationship?

[-]

WatchStoredInAss@reddit

Nice, /thread.

[-]

Significant-Dance-43@reddit

[-]

Jordangander@reddit

Ok, so you are comparing two things, but overlapping them at different points in time,

This is a garbage chart no direct connections.

[-]

AdditionalTip865@reddit

The idea is that the people committing crimes as young adults were affected developmentally by lead poisoning as children. So the delay is the amount of time required to grow up to peak criming age.

I've always thought that as a single cause for the crime wave that went from the late 60s to the mid-90s, this hypothesis is a bit too reductive. There were a lot of things going on in that era. But I have no trouble believing it was a contributor.

[-]

pwaltman1972@reddit

Excellent response. I've had the same thought as well (about it being too reductive), e.g. how does this square with the theory from the book "Freakanomics" that posits that legalized abortion in the US lead to the reduction in crimes (and other anti-social behavior such as teen pregnancies).

Personally, I'd be curious about comparing these curves for other countries, assuming that there are/were countries that allowed/used leaded gas and paint, e.g. comparing France versus Ireland (assuming both of those countries allowed the use of leaded gas & paint).

[-]

AdditionalTip865@reddit

If I recall correctly, the lead-crime people claim that the correlation holds up in other countries over that same time period, which is interesting if true. I think the rise and fall in violent crime was NOT just in the US. But also, there's been a much more recent rise in Latin America and it may not fit into this scheme.

The US has generally high levels of violent crime for a rich country, though not high for a poor developing one (it might be possible to model the US as two countries nailed together). For a while it seemed like the post-1995 drop had stalled out and even reversed during the COVID pandemic, but last I heard, it'd dropped again post-pandemic to the lowest levels ever.

[-]

hocfutuis@reddit

There's a book called Murderland, by Caroline Fraser, which looks into this link. It's a pretty interesting read if you like true crime.

[-]

Jordangander@reddit

A contributor to early death, sure. But a contributor in any meaningful way to street crime? If that was the case, civilizations like Rome would have never been able to exist.

[-]

_Brandobaris_@reddit

Actually there is a lot of thought that the Roman use of lead pipes contributed to its demise. When the Huns sacked Rome in 410 CE historians still can’t see how they lost as the Huns didn’t go quickly and Rome never responded.

Unfortunately we can’t test their blood but the existence of lead pipes certainly put lead in their blood and bodies. And we do know what happens with developing brains when exposed to lead.

[-]

Skelastomybag@reddit

Jesus christ I didn't notice that! It's two different timelines! That's beyond ridiculous.

[-]

KennyGeigh@reddit

The problem is that Beanie Babies and Blockbuster video sales also have similar arcs, but have not been factored in. OP needs more time with ChatGPT.

[-]

Teaspoonbill@reddit

Good to know that witnessing the collapse of society isn’t the only thing we have in common with the people who lived during the end of the Roman Empire.

[-]

bliceroquququq@reddit

I'll regret asking this, but in what way is society collapsing?

[-]

RedditSkippy@reddit

I assume that the dip in lead levels in the early 1940s was caused by less driving during the war years.

The interesting thing is that the lead levels and crime rates diverged in the very late 70s. Is that because younger generations were seeing the effects of lead regulation but there were still a lot of adults out there who didn’t benefit? I would be curious to see this cart carry into the 1990s.

[-]

OGCelaris@reddit

It could be more due to the income inequality gap. The poorerpeople are the more crime. Tackling that problem takes effort and fundamental changes to our way of life so it's unpopular with politicians who just want a slogan and a bandaid solution.

[-]

_Brandobaris_@reddit

Lead remains an important munitions metal not necessarily the bullet part. So likely more the war efforts demand on lead.

[-]

_Brandobaris_@reddit

Here is detailed discussions on both overlapping theories.

https://en.wikipedia.org/wiki/Legalized_abortion_and_crime_effect

https://en.wikipedia.org/wiki/Lead–crime_hypothesis

Remarkable overlap of these two theories seems like both play their parts in crime reduction.

[-]

AdAggravating8273@reddit

I believe the abortion overlap. I read Freakanomics years ago and it laid it out pretty well.

[-]

_Brandobaris_@reddit

As does the lead which is supported from non-US data as well. I don’t think either is the sole contributor but the overlap is remarkable.

[-]

pwaltman1972@reddit

Man, the amount of pushback on this theory is certainly somthing.

For those unfamiliar, the gist is that lead - a proven and well-known cause of intellectual disability, even in small quantities - was everywhere in the US while we (and our parents' generations) were growing up.

Even in *super* small quantities, it might be enough to shift our collective IQs a few points down from what they would have otherwise been, while decreasing our impulse control. For the overwhelming majority of us this hasn't had a huge impact, but for those of us with lower IQs and less impulse control, it could have enough of an impact to increase their likelihood of criminal behavior. Once in the system, that would just be magnified. The effect was likely greater in urban areas because of the greater concentration of cars, pollution, etc.

It's one of those effects where the impact doesn't show up until someone hits their teenage years, which is why the dates of the two curves are offset by 18 years.

[-]

OptimusWang@reddit

You must be an older X’er, they’re offset by 23 years ;)

[-]

pwaltman1972@reddit

Yeah, yeah, yeah, lol. Just fixed it

[-]

OptimusWang@reddit

Haha sorry, I normally wouldn’t point something like that out any more than I would point out a spelling mistake, but you kinda set yourself up for the joke 😂

[-]

_Brandobaris_@reddit

Good points. I’d add it is bioaccumulate which means when injected in some their our bodies don’t know what to do with it and can’t get rid of it naturally.

[-]

Vylnce@reddit

For anyone who misses having high lead levels, rest assured that you can still obtain them by making poor choices around shooting sports hobbies. Spend a lot of time shooting indoors and/or start casting your own bullets and you too can have the same lead levels as those kids who were eating paint chips.

[-]

Six_Pack_Attack@reddit

We are seeing the effects of lead all over the comments.

[-]

Majik_Sheff@reddit

It's painful.

[-]

ONROSREPUS@reddit

So this isn't from eating your #2 pencils? /s

[-]

KennyGeigh@reddit

Hmm...what a brilliant correlation. However, you failed to factor in the rise and fall of Blockbuster video, which also follows the same arc. And everyone knows watching action movies makes you violent. Please re-evaluate and report back.

[-]

CrazyFaithlessness63@reddit

What happened in 1979 (or 2002) to cause them to diverge so much? Something new that leads to increased crime rates in young adults, more broader demographic committing crimes?

[-]

bigtime_porgrammer@reddit

Incidentally, that drop in crime in the 1990s was famously theorized to have been a result of fewer unwanted pregnancies in the wake of Roe v. Wade - https://en.wikipedia.org/wiki/Legalized_abortion_and_crime_effect

[-]

AdditionalTip865@reddit

The Freakonomics hypothesis. Yeah, I doubt that's the One True Cause too, but I suspect the crime boom was the result of a complicated combination of causes and all of these things might have contributed. The Vietnam War, white flight and flawed urban planning, big changes in economic and social policy, all these were big frickin' deals.

[-]

Old_Use7058@reddit

I know! imagine if we killed violent criminals at the rate we killed innocent babies. That number would plummet

[-]

pwaltman1972@reddit

No one kills babies, but nice try

[-]

WindowFruitPlate@reddit

Abortion is murder

Those are human lives and nothing you say will change that very base fact

[-]

pwaltman1972@reddit

No. Spare us the BS. You don't care about those "lives." If you did, you and your movement would be actively working to better fund foster care (demanding more money & passing laws), and would be making it easier for people to adopt those kids (universal healthcare, universal childcare, etc). But in the half century that your movement has existed, you never have - because you don't actually give a shit. So, again, spare us the bullshit. You're just LARPing like you actually care.

[-]

SamePhotographs@reddit

Best of luck that zygote living outside of my body. I do not consent to being it's host. I'm already here. My health and wellbeing is more important.

[-]

jipsee1973@reddit

If that were true, it would be called 'murder'. But it's not. Because it isn't.

[-]

DeltaBlues82@reddit

Murder is murder. Abortion is abortion. Abortion isn’t murder.

Words mean things.

[-]