Related: https://www.journals.uchicago.edu/doi/10.1086/715713, estimates the annual cost of crime in the US as ~$5 trillion including transfers from victims to criminals and ~$3 trillion net of transfers, or ~$16,000 per American including transfers / ~$10,000 net of transfers.
Good to have some sharp stats on this. Yes crime is bad and the culture that minimises, excuses or even glamourises it is worse in a sense.
One note that doesn't change your conclusion but is worth a thought: US soft power is ginormous, of benefits hard to fathom and that go way beyond the economic. And inner city culture has hugely, disproportionately contributed to this power. Rebels from Tehran to Moscow, Dakar or Beijing rap their message to hip hop - willingly or not idolizing some subset of US culture. Not saying it makes the crime culture worth it, but no analysis of its economic drag is complete without at least some consideration of what the US culture and soft power has gained from it.
Might this just be a proxy for population density? Crime needs victims; more dense means more victims means more crimes. More dense also means more jobs, and it makes sense that blacks would migrate to places with jobs. Richer people (often whiter) can afford to relocate to avoid crime, hence white flight. Don’t whites commit 50-70% of crimes (maybe you think arrests data is biased) and also whites aren’t more likely to be victimized by other whites? It seems wrong to attribute all of this to the presence of blacks. Maybe you’re saying the influx of blacks made whites more precarious and also drove them to increased criminality, but… that seems pretty tenuous. The stuff about other races not seeing similar effects is interesting but I wonder if there are temporal factors here; I didn’t think black and asian migrations happened at the same times in major cities but can look.
Density is not a predictor of crime. Nor is poverty or lack of education. Look at crime in cities during the Great Depression or crime levels in densely packed Tokyo.
Overlay crime rates on a map of the world. Same story.
This would be more convincing if it explored quasi random increases of young men who are much more predisposed to violent crime than any particular racial group
The is a huge difference between a large city and a large metro area. There are only 12 cities that have more than 1 million residents in the U.S. but there are 56 metro areas with more than one million people. One should average crime over the metro area.
What if we turbocharged the agglomeration effect like Los Alamos during the Manhattan Project — moved all the smartest people to very liveable fully automated luxury communism zones, and the less productive groups are paid to relocate to “productivity cities” with surveillance and company towns run by businesses where manual labor is still necessary
Confused. I don’t get how crime in a city I want no part of living in costs me. Also confused because crime rates have dropped so dramatically over the past two generations (didn’t it peak like 36 years ago?)
I’ve lived in the US my entire 67 years and the only time as an adult anyone in my family experienced anything other than credit card fraud was a purse snatching from my wife in the 90s.
If the city is more economically productive its residents pay more taxes and generate more economic growth. The taxes and economic growth benefit other parts of the state and country.
Geoffrey west has work that show there are also agglomeration effects for crime in large cities too. I wonder if that was included in the model what would happen, if we expect murders to be higher and the elasticity metric, it may be even worse if adjusting for that.
Are there studies on how much homeless encampments (and related things that would fall under 'disorder' rather than 'crime') hurt agglomeration?
There aren't, but there should be. Seems like a topic Charles Lehman would be interested in.
Related: https://www.journals.uchicago.edu/doi/10.1086/715713, estimates the annual cost of crime in the US as ~$5 trillion including transfers from victims to criminals and ~$3 trillion net of transfers, or ~$16,000 per American including transfers / ~$10,000 net of transfers.
Unpaywalled version: https://www.researchgate.net/publication/358417739_The_Aggregate_Cost_of_Crime_in_the_United_States.
Cited in the replication package!
Good to have some sharp stats on this. Yes crime is bad and the culture that minimises, excuses or even glamourises it is worse in a sense.
One note that doesn't change your conclusion but is worth a thought: US soft power is ginormous, of benefits hard to fathom and that go way beyond the economic. And inner city culture has hugely, disproportionately contributed to this power. Rebels from Tehran to Moscow, Dakar or Beijing rap their message to hip hop - willingly or not idolizing some subset of US culture. Not saying it makes the crime culture worth it, but no analysis of its economic drag is complete without at least some consideration of what the US culture and soft power has gained from it.
Hi Cremieux,
Can you look at the recent xylitol findings and examine their accuracy?
They're saying it's linked to 57% increase in heart attack risk. I love my artificial sweeteners!
That's absurd. Link?
Found it: https://x.com/cremieuxrecueil/status/2098261215906591122
Thank you for being the goat Cremieux.
Speaking of, idk if you’ve ever done just a whole essay on artificial sweeteners but that would be awesome.
Might this just be a proxy for population density? Crime needs victims; more dense means more victims means more crimes. More dense also means more jobs, and it makes sense that blacks would migrate to places with jobs. Richer people (often whiter) can afford to relocate to avoid crime, hence white flight. Don’t whites commit 50-70% of crimes (maybe you think arrests data is biased) and also whites aren’t more likely to be victimized by other whites? It seems wrong to attribute all of this to the presence of blacks. Maybe you’re saying the influx of blacks made whites more precarious and also drove them to increased criminality, but… that seems pretty tenuous. The stuff about other races not seeing similar effects is interesting but I wonder if there are temporal factors here; I didn’t think black and asian migrations happened at the same times in major cities but can look.
Agglomeration? It's about density, and the Great Migration reduced said density.
13/50
Based on everything we know, that "50" is low!
Density is not a predictor of crime. Nor is poverty or lack of education. Look at crime in cities during the Great Depression or crime levels in densely packed Tokyo.
Overlay crime rates on a map of the world. Same story.
This would be more convincing if it explored quasi random increases of young men who are much more predisposed to violent crime than any particular racial group
The is a huge difference between a large city and a large metro area. There are only 12 cities that have more than 1 million residents in the U.S. but there are 56 metro areas with more than one million people. One should average crime over the metro area.
If crime changes when people's opportunities change, how confident can you be that the problem is the population rather than the conditions?
What if we turbocharged the agglomeration effect like Los Alamos during the Manhattan Project — moved all the smartest people to very liveable fully automated luxury communism zones, and the less productive groups are paid to relocate to “productivity cities” with surveillance and company towns run by businesses where manual labor is still necessary
Confused. I don’t get how crime in a city I want no part of living in costs me. Also confused because crime rates have dropped so dramatically over the past two generations (didn’t it peak like 36 years ago?)
I’ve lived in the US my entire 67 years and the only time as an adult anyone in my family experienced anything other than credit card fraud was a purse snatching from my wife in the 90s.
If the city is more economically productive its residents pay more taxes and generate more economic growth. The taxes and economic growth benefit other parts of the state and country.
One should not value white flight or even longer commutes at zero.
Geoffrey west has work that show there are also agglomeration effects for crime in large cities too. I wonder if that was included in the model what would happen, if we expect murders to be higher and the elasticity metric, it may be even worse if adjusting for that.
Check the links in the article to see this answered!