I’m afraid I was not convinced by Jason Riley’s WSJ column asserting that more police officers is the key to reducing crime:
The Democratic reaction to President Trump’s federal crackdown on crime was as predictable as the dozens of shootings in Chicago over the holiday weekend, which left at least seven dead.
Chicago’s violent-crime spikes during the warmer months have become too commonplace to shock us as they should. CBS News reported that during Independence Day weekend five people were killed and 36 were injured. Over Labor Day weekend in 2023 and 2024, a total of 13 were killed and 67 wounded in shootings around the city.
When Chicago Mayor Brandon Johnson was asked repeatedly in a television interview last week whether putting more police officers on the street would help curb crime, he called that approach “antiquated” and tried to change the subject: “I believe the city of Chicago and cities across America would be safer if we actually had, you know, affordable housing.”
I will readily stipulate that Mayor Johnson is a buffoon and one could do much worse than doing the opposite of anything he suggests. Mr. Riley goes on to quote a number of studies supporting his view that more police = less crime. I genuinely wish that columnists would actually cite the studies they quote. It would be helpful.
I did a little quick research of populations, police officers, and other factors. I limited my research to the ten largest cities and the homicide rate as an epitome of violent crime. The results are a little hard to read—click on the table for a larger version:
Since it’s a bit difficult to discern any patterns from the table, I created a scatterplot:
That’s not dispositive but I think it’s highly suggestive: there is no straightline relationship between the number of police officers and the homicide rate. A scatterplot of the area of the cities is even more interesting:
I would also make the claim that factoring demographics into the findings would be revealing.
I would summarize my findings that
- There is no strong relationship between the number of police officers per 100K population and the homicide rate.
- There is a stronger relationship between population density and homicide rate with New York City being the great outlier. New Yorkers are a breed apart.
- Otherwise the causes of a high homicide rate are multifactorial including population density, demographics, and how law enforcement is performed including police officers, states attorneys, and judges.
Of all the studies and meta-analyses I have seen if the number of police has an effect, it is likely fairly small. I can see a large bump in numbers having some effect for a short while. I would rather the number of police a city has be determined by that city. They are the ones most likely to have a feel for the culture of their city and what the effects and tradeoffs for having more or less police would entail.
Steve
That’s just about exactly what I think.
I also think that if the likelihood of punishment rises it will deter crime but that is more closely related to what the police do than their number. It also depends on prosecutors and judges.
I was a intrigued by the New York City outlier and the possible relationship in density and homicide rate; so I looked for more data points to plot. Referencing the densest cities in the US on wikipedia, I found San Francisco, Boston, Miami have densities between Chicago and New York; but they all have very low homicide rates in 2024, all between 4 to 6 per 100K.
If you add those to the map, there doesn’t look to be a clear linear relationship between density and homicides, and New York isn’t a great outlier.
Homicide rate is at best a crude first order approximation of the rate of violent crime. It’s quite possible there’s no relation between density and violent crime. I was merely commenting on what I saw in the scatterplot.
As I’ve said before it’s hard to make intelligent generalizations about violent crime without taking demographics into account.
Much, if not most violent crime is impulsive. I am skeptical about numbers of police or government policy affecting that other maybe overall economic growth, I think culture is more influential. However for non-impulsive crime I think it’s not just surety of punishment but that it occur relatively quickly. If your gang commits a planned crime and punishment isn’t until 5 years later it doesnt seem that related especially to the younger members and it’s just a cost of doing business.
Steve
Agree completely.
Flooding the zone with police officers definitely helps – as long as the zone stays flooded. But flood waters recede, and rats can scurry to higher ground. It was getting worse in my neighborhood (Pine Hills, on the outskirts of Orlando) about seven or eight years ago, and the local Authorities decided to Make A Statement. More deputies would be pushed into Pine Hills at night and one the weekends. The day before this started there were four fatal shootings! ?
Also, the Bad Guys just started conducting their business during banker hours. This led to another fatal shooting more or less just up the street from me one morning when my daughter was out front playing with a neighbors grandchildren. FUN FOR THE WHOLE FAMILY!
More police won’t help, but getting the bad guys off the streets does. If only someone would try this.