Drawing from uniform crime reports since 1933 and vital statistics since 1910, I document the evolution of black-white homicide disparities in the United States since the early 20th century. Homicide Arrests The Uniform Crime Reporting program began in 1929 to compile national crime statistics, and since 1933, arrests by race have been reported. An early example is the following 1934 report showing per-capita arrest rates by race for various crimes:
Very nice presentation. Especially, the stats on correlation of criminality with victimization odds. How can anyone not see what is going on with sky-high black murder rates. It's the same dynamic as an endemic mafia war. It's not soldiers on civilians, it's soldier on soldier casualties. Some might cynically describe it as "the trash taking itself out."
I have a theory that you could model crime offender/victimization rates mathematically by a formula that takes account of the exponential rise in risk rates when the "criminally inclined" interact in the same environment. In other words, a criminal interacting with a law abiding person might have X% risk of offending/being victimized when he is in an otherwise law abiding community. But if he interacts with another criminal, BOTH criminals would shoot up to, say, 5X% risk in their interaction with one another. Thus when more criminals are added to the network/community/environment, the rate of crime increases geometrically with the increase of the percentage of criminals in the relevant environment. In other words, crime rates could probably be largely predicted as a function of the DENSITY of criminals in the relevant environment (defining the relevant "interaction" area or group would be an issue but you could probably get by with using census tracts or zip codes). Another way to put the hypothesis is that there is a "network effect" for criminality.
The power effect of criminal-density on crime rates could be reverse engineered from the basic crime stats by location.
Anyway, I'd humbly suggest this hypothesis might be interesting to test, in case you wanted to crunch the numbers to see if it holds water.
Anytime anyone says "crime is down compared to the crack wars of 1990" I ask if that is the relevant date to compare to.
I think most people as a comparison have in their minds the kind of public order we had up until the late 60s and beyond (the racial riots can mark the beginning of this).
Another problem with using the stats is that attempted murders become murders about 80% less then they used to before the trauma care innovations of the Vietnam War saved a lot of gunshot victims.
In any even I think the level of "acceptable crime" in the black community is just too high for most whites. When it reachers "1990 crack war" levels even blacks will start demanding broken windows policing, but even historically low points in black crime are just way above that whites will accept.
I love this sort of clear presentation of historical data. I advance the fiscal sink hypothesis about crime patterns in the US here.
https://lorenzofromoz.substack.com/p/race-and-other-annoyances
Very nice presentation. Especially, the stats on correlation of criminality with victimization odds. How can anyone not see what is going on with sky-high black murder rates. It's the same dynamic as an endemic mafia war. It's not soldiers on civilians, it's soldier on soldier casualties. Some might cynically describe it as "the trash taking itself out."
I have a theory that you could model crime offender/victimization rates mathematically by a formula that takes account of the exponential rise in risk rates when the "criminally inclined" interact in the same environment. In other words, a criminal interacting with a law abiding person might have X% risk of offending/being victimized when he is in an otherwise law abiding community. But if he interacts with another criminal, BOTH criminals would shoot up to, say, 5X% risk in their interaction with one another. Thus when more criminals are added to the network/community/environment, the rate of crime increases geometrically with the increase of the percentage of criminals in the relevant environment. In other words, crime rates could probably be largely predicted as a function of the DENSITY of criminals in the relevant environment (defining the relevant "interaction" area or group would be an issue but you could probably get by with using census tracts or zip codes). Another way to put the hypothesis is that there is a "network effect" for criminality.
The power effect of criminal-density on crime rates could be reverse engineered from the basic crime stats by location.
Anyway, I'd humbly suggest this hypothesis might be interesting to test, in case you wanted to crunch the numbers to see if it holds water.
Anytime anyone says "crime is down compared to the crack wars of 1990" I ask if that is the relevant date to compare to.
I think most people as a comparison have in their minds the kind of public order we had up until the late 60s and beyond (the racial riots can mark the beginning of this).
Another problem with using the stats is that attempted murders become murders about 80% less then they used to before the trauma care innovations of the Vietnam War saved a lot of gunshot victims.
In any even I think the level of "acceptable crime" in the black community is just too high for most whites. When it reachers "1990 crack war" levels even blacks will start demanding broken windows policing, but even historically low points in black crime are just way above that whites will accept.