Get the Lead Out II
There is an absolutely serious hypothesis that the sharp drop in violent crime in the USA since the early 1990s is due to the EPA and, in particular, to the shift from leaded to unleaded gasoline roughly two decades earlier.
Kevin Drum notes that James Q. Wilson ascribes about half of the total decline to lead exposure. The hypothesis makes sense, lead poisoning too mild to cause other symptoms is very strongly correlated across individuals with violent behavior.
The hypothesis has clear implications for violent crime in other countries, since the USA banned lead early. The UK went unleaded in 1986 12 years after the USA. The USA violent crime peak came in 1995. So the predicted UK crime peak would be in 2007.
In get the lead out I, I thought the USA went unleaded (for new cars) in 1973 so with an arithmetic fail, I predicted 2008 for the UK. Notably, the post is dated 2008 so I had only a very tiny bit of data.
Now they have the number of reported violent crimes for two more years 2008/9 and 2009/0 (I have no idea why they report by a period other than the calender year).
I have a huge ugly pdf with the data I need and much much more.
The bottom line is that the total number of violent crimes was basically identical in 2004/5 2005/6 and 2006/7 then declined about 17% by 2009/10. The predicted peak of 2007 corresponds about as precisely to the data as is conceivable. This really was an out of sample prediction (except I forgot the year of the US requirement that new cars use unleaded but that’s my fault not the fault of the lead hypothesis).
A hypothesis has yielded a correct out of sample prediction about violent crime rates. This isn’t just extrapolating the trend, the prediction was made right at the time of the predicted peak.
The problem is that I have too much data. I have an illegible screen shot, but I will just retype the relevant numbers after the jump.
update: My post with the prediction (and mis remembered date of unleaded in the USA) is here. I would have just edited without admitting it was an update, but Mark Sadowski has already commented on the post (I thought he was more into geld than lead).
The numbers
Total Violence Against the Person Offences
1998/99 502,778
1999/00 581,034
2000/01 800,913
2001/02 850,326
2002/03 845,078
2003/04 967,228
2004/05 1,048,095
2005/06 1,059,583
2006/07 1,046,187
2007/08 961,099
2008/09 903,447
2009/10 871,712
Interesting hypothesis. I’ll have to run the numbers before I acknowledge the correlation. But it makes sense as I am American and was born in 1964 and I constantly feel the urge to break your neck.
I was born in 1960 and exposed to much more lead than you were so watch out.
Amazingly quick comment by the way. I wasn’t entirely sure if I had hit the post button yet.
“(I have no idea why they report by a period other than the calender year).”
Because the government year runs 6 April to 5 April. Budgets, the fiscal year, the tax year, everything.
It used to be Quarter Day (of which there are four of course, being the day on which quarterly rents were payable) and we always used to use the Spring one, March 25th.
Then along came that change from the Julian Calendar to the Gregorian, in 1752, needing an 11 day shuffle in the calendar (and there was another 1 day shuffle later).
Slightly weird, I know, but that’s why it is as it is.
For example, income tax rates (and allowances etc etc) are always quoted as “for the tax year 2009/10, 2010/11” etc.
I would add that 1978 is when lead was also banned in paints. Gradually paint flakes around homes are becoming more and more lead free as the lead paints are removed. I wouls suspect that given that lots of young children put things in their mouth that this may help as much as the lead in the air.
There’s definately a relationship between lead, and violent crime.. Lead bullets, that is.. in the hands of law-abiding, background-checked citizens.. loaded into lawfully concealed handguns..
I’d really like to see the work of Nevin and Jessica Wolpaw Reyes (who claims that half the decline in crime resulted from less lead in the environment, hence in little kids’ heads) combined with the Levitt Roe v Wade data. How much of the decline is attributable to these two effects combined? (It’s clear how they could multiply each other…)
Unless of course there isn’t. Canada controls of access to our much beloved warm guns much more tightly and has historical crime rates much lower than the US.
For those of you who have not spent enough time in New York’s slums and other badlands (Times Square was the a**hole of the world in early ’70s) here are the origins of crime — which led paint or abortion may or or may not add to or subtract from — where it begins.
Five esoteric things in a row on juvenile delinquent boys:
Boys until 18 1/2 are in the emotionally dependent stage — for all practical purposes — as much as if they were 12.
This turns off over a week’s time in my personal observation — pure social instinct thing.
THE CORE: If they perceive nobody cares about them (wrong about half the time) they literally wont care about themselves — no penalty can deter them. Any street temptation at all, they are gone.
This is easy enough to accept with a badly neglected 12 year old — just as crazily true for an 18 year old who has simply been out of control for a long time — every bit as hysterically alienated as I call it.
Unlike the decades of positive socialization it takes to wind down the paranoia underlying heavy heroin or alcohol addiction — only 5 to 6 weeks of normal adult attention (teaching to drive) — slowly brings the kid into the normal supportive orbit (have to kiss his toes and tell him everything he wants to hear first 7=10 days) — but the crime doesn’t slow one bit until a new kid wakes up one day (invasion of the body snatchers day ???).
Gangs:
Very simply the Crips and the Bloods could not whip a decent paying Ronald McDonald.
While reading When Work Disappears by William Julius Wilson and American Project by Sudhir Alladi Venkatesh side by side I noticed that after Wilson’s book ended the project only descended into a completely gang infested hell as the minimum wage dropped in half from LBJ’s peak — as average income doubled.
BOTH OF THE ABOVE — certainly and totally the latter — CAN BE BLAMED ON AMERICA’S AMAZING, INCREDIBLY SQUEEZING LABOR MARKET. A $15 minimum wage would add about about 3% (earlier figuring here *) to the cost of living (not counting other wages pushed up — good), give half the country a raise and send a lot more low end customers (me ;-]) to McDonalds. Needless to say with a labor market where the median wage grew 20% while average income doubled and left a quarter of the workforce below LBJ’s minimum wage by early 2007 (under performing Malthus) many more fatherless homes(massively more in our inner cities) leads to many (massively) more neglected or thinks they are neglected, hysterically alienated kids (for those of you who have not spent enough time in the badlands). * http://ontodayspagelinks.blogspot.com/2008/08/3-cost-of-gdp-output-and-inflation.html
America’s great wage depression also leads to schools that don’t work because nobody can be bothered trying to excel when they know they are not going to get paid anything like adequately when they leave school to finally go to work — according to a professor Martín Sánchez-Jankowski who spent 9 years on the street observing in 5 inner city neighborhoods (slums): read his book Cracks in the Pavement.
So to me crime in America all boils down to what to me almost every other problem in America all boils down to: […]
I certainly recall how foul the air in the larger cities smelled and tasted prior to the clean air amendments. it is a lot easier to walk down the street today!
Comparing crime in Canada, to crime in the U.S. … is like comparing crime in Michigan’s upper peninsula, to crime in Detroit.
Do you believe that crime in Canada would increase, if law-abiding Canadians could own and carry handguns ? I’d bet that it would have the same effect.. Canada’s already low crime-rate, would go even lower.
I’m gonna research this… I’ll bet that crime in Canada’s major cities, is on par with crime in the U.S’. major cities. Canada as a whole, is FAR more rural, than the U.S.
This has already been documented. A long time ago:
“The study, published last week in The New England Journal of Medicine, examined gun and crime records in Seattle and Vancouver, British Columbia. The two cities are similar in population size as well as in geography, climate and history. They share similar rates of education, income and unemployment; their residents even watch the same television shows.
The two differ dramatically, however, in their approach to guns. Vancouver restricts purchases of handguns to legitimate target shooters and collecters, and severely limits their use. Seattle allows virtually unrestricted over-the-counter sales, and requires only a permit to carry a concealed gun on the street. Not surprisingly, gun ownership is more than three times greater in Seattle than in Vancouver.
The researchers analyzed aggravated assaults and homicides according to the weapons used. The results were striking: the rates of crimes committed with knives and other weapons were roughly the same in both cities, but the rates of crimes committed with guns were far greater in Seattle. In fact, the gun crimes appeared to account almost entirely for Seattle’s higher rates of aggravated assaults and homicides.”
http://www.nytimes.com/1988/11/18/opinion/guns-do-kill-people.html
Thanks. The US also has fiscal years, but most other things are done by calender years — basically if you have to have the last years numbers by year end to close accounts then use the fiscal year, but if you can add em up with a delay then why bother. Key rule, it is very difficult and expensive to get anything done in late December.
That Julian to Gregorian bit was a bit late in the UK (Miguel de Cervantes and William Shakespeare died on the same data April 23 1616, but Shakespeare died later than Cervantes since it took the Julian calender had had more leap years).
Thanks for the information. I’d tend to guess that leaded paint is more important than leaded gasoline. Certainly there was always a huge concern about leaded paint back when nobody worried about leaded gasoline. Leaded paint is the cause of sometimes lethal lead poisoning and, fairly often, of poisoning to a level far above that whose only known symptom is violence.
The lead people (I’m an amateur) claim that the US crime wave in the 1920s followed a wave of exposure to leaded paint — use increased until the turn of the last century, then people realized the danger (this was not an effect of regulation just word of mouth warning). It was still very much a problem in the 1970s though.
My original interest in looking at the UK was based on the fact that the UK legalized abortion earlier and banned lead later. There isn’t a Levitt David Osborn effect in the UK data. I think the Roe v Wade hypothesis has less empirical support than the lead hypothesis. Of course, as you note, it’s not a horse race, both can be true.
One unfortunate feature of the EPA reducing crime is that the peak was correlated with three strikes and you’re out and such. It is clear that higher and higher crime caused longer and longer sentences as a response. Then something good but unrelated happened and now people are sure that toughness works. To me it is like welfare reform and the late 90s boom.
The irony of the lefty and effective strategy under Nixon and the right wing extremely costly (especially in suffering of the prisoners) under Clinton adds to the sting.
This comment is consistent with the lead hpothesis. The lead heads don’t deny all of that. But an alienated teenager exposed to lead as a child is more likely to be violent. This claim is based on data. And the time series tends to support it. But mostly, it doesn’t have to all boil down to any one thing.
Now as a practical matter, we can’t ban lead again. I think more could be done to get rid of the last lead paint and to reduce polution from smelters and stuff, but not much.
On the other hand, making a minimum wage job pay a decent salary *after tax* is very simple. First a higher minimum wage by law (the old estimates of large bad effects on employment are *the* classic case of blatant data dredging). Second a higher earned income tax credit especially one also directed at single young men who might go for a bit of crime.
Now as to bringing organized labor back from the dead, well, the best shot was the EFCA and 60 Democratic Senators weren’t enough. I just don’t know what to do.
Indeed. The Gregorian calendar was seen as a Catholic plot (it had been invented by a Pope, after all) and the Protestant countries (esp the Brits) were later to adopt it than the Catholic countries. In some cases, 150, 170 years later.
Like Oliver Hardy used to say to Stan Laural: “If you want something done right, you have to do it yourself.”
I guess I’ll have to get the facts of the crackpot US labor market before the public on my own. Maybe go back to NYC and run for something. Am sure I can get things stirred up.
When people know the minimum wage dropped in half by early 2007 while average income doubled for not other reason than nobody cared — about themselves — the latter being why the median wage grew 20% while average income doubled — and the SECTOR-WIDE labor agreements is the way to start caring about themselves — then, you will know just what to do — support re-balancing the amazing, incredibly squeezing American labor market.
ddrew
you are of course right. but the lead wasn’t helping.
i wonder if lead in the air is the cause of the insanity in Washington. They say it was lead pipes that caused the insanity in Rome.
I would add to your observation about NYC. Culture is something humans create. it takes culture to bring us out of the jungle. There is plenty of violence in places where there is no lead paint.
What we have in America… and it began with slavery… is a destruction of the culture that was so carefully evolved over a million years to make us better than the apes.
Should we bring up the Old Calendrists and New Calendrists in the Russian Orthodox Church?
Ah yes. The Canadians are our moral superiors argument.
Another obvious cause would be the disappearance of child street culture due to television, more restrictive parenting, suburbanization and video games. Kids don’t organize their own play anymore, so they have more trouble picking up criminal past times.
Yet another obvious cause would be the epidemic theory of crime. The rise began with the rise of heroin, followed by the rise of crack cocaine. Epidemologists were predicting that the crime waves would burn themselves out, and now they argue that they have done so.
Personally, I think it was a combination of things including gun control, a return to community policing and stricter sentencing guidelines for repeat offenders. Hell, abortion, birth control, the 90s boom and the overall baby bust probably played a role. Whether they were all part of some larger phenomenon I’ll leave to the historians.
“to make us better than the OTHER apes.”
The lead poisoning from gasoline is mostly from roadside crops or other plants, which people eat. An amusing question is what happened to attacks by viscious rabbits, cows and sheep who chew on the grass.
Also, please read Freakonomics. The time period in question was about 20 years after Roe v. Wade. Levitt makes the case that legalized abortion helped ensure fewer unwanted (and therefore poorly raised) children would be born. They compared crime rates in the five states that had legalized abortion before Roe v. Wade with the rest of the country and saw that the rest of the country’s crime rates started falling 15-20 years after Roe v. Wade. Unlike what Wilson mentions, Levitt has a much more controlled experiment. Don’t flame me on the morality of abortion. I am just repeating Levitt’s findings.
Again, look at Freakonomics. Levitt pretty convincingly takes apart the heroin/crack cocaine theory of crime as a source of the rise and fall of crime levels. As for birth control, most of the laws restricting that went away 5-10 years before Roe v. Wade.