The cost of ideas

Well, this is interesting. From U.S. Murder Rate for 2018 Is on Track for a Big Drop by Jeff Asher.

Its a limited study. "Murders are down so far in a sample of large American cities, typically a good indicator for national numbers." They are properly noting the limitations to the study. None-the-less it is a canary in the coal mine.

What is interesting is how circumspect the reporting is.

Murder rose 23 percent nationally between 2014 and 2016 before leveling off in 2017. Major increases in murder in Chicago and Baltimore received much of the national attention, but the increase occurred throughout the country.

In the cities in which data is available, murder has been down about 7 percent on average this year relative to the same point in 2017.

Is there something that occurred in 2014 which served as a catalyst to this surprising 23% rise between 2014 and 2016? And it was a surprise given that it followed a nearly fifteen year long run of nearly consistent year-on-year murder rate decline?

Tracking the change in murder nationally is far easier than explaining why it’s happening. There is still no consensus on why murder rose nationally in 2015 and 2016, though various theories have been proposed, including simple randomness. Similarly, a projected drop in murder in 2018 would not have an obvious cause. Employment of smarter technologies, expanded community intervention programs, and even colder weather could help explain year-to-year local changes.

Nowhere in the reporting is there any mention of the Ferguson Effect hypothesis. The mainstream media and the critical theory wing of academia are dismissive of the hypothesis. Heather Mac Donald has probably been among the most vocal and articulate proponent of the hypothesis.

I suspect there is good merit to the Ferguson Effect hypothesis in the first couple of years but also suspect that the further we get from Ferguson and the more people get exasperated with crime, the more police departments will return to at least some modified form of proactive policing.

So we had three years of unexpected increases in murder rates from the time of Ferguson and the abandonment of proactive policing in many big city jurisdictions. Since 2017, with a return of a commitment to strong support of police and criticism of policies of reduced proactive policing, we are now seeing a return to the trend of declining murder rates. Sure it might be coincidental. But it is exactly what the Ferguson Effect proponents predicted.

It is too early to tell just how supportive this is of the Ferguson Effect hypothesis, but those dates sure are suggestive.

And this isn't an academic exercise. In 2017, there were 17,284 murders in the US. A decline of 7%, if it eventuates, means that 1,210 people will have survived 2018 because of a return to more effective policing. Another way of looking at it is that since 2013, the last year of a declining murder rate, there have been nearly 7,500 lost lives in excess of what we would have expected if only we had sustained the rate of 2013.

I am not sure what freedoms or other benefits we might have obtained from abandoning proactive policing (if that was the cause of the increase) but I find it hard to imagine that it was worth 7,500 lives.