From the New York Times
Whether Crime Is Up or Down Depends on Data Being UsedBy TIMOTHY WILLIAMS SEPT. 27, 2016The news from the F.B.I. crime data was alarming: The murder rate rose sharply last year, driven by jumps in several major cities.Four urban areas — Baltimore [Freddie Gray], Chicago [Laquan McDonald], Milwaukee [Dontre Hamilton] and Washington — accounted for about a fifth of the increase in homicides in 2015. Those cities, however, make up only about 1 percent of the nation’s population.
For the cities with the 3 highest increases in homicide in 2015, I inserted the names of the black males shot by cops that spurred Black Lives Matter protests that led to big increases in murder.
But whether crime is up or down depends on what data is being looked at — and who is doing the looking.The F.B.I. data showed that violent crime rose about 4 percent last year from 2014, and homicides increased 10.8 percent. Yet crime over all fell in 2015 for the 14th consecutive year.
Homicide is one of the rare crimes, along with kidnapping, that definitely gets a police response and statistical enumeration. Are people in Chicago at present bothering to call the cops over their bikes getting stolen? In 1986 right outside the Loop I got the back window of my hatchback smashed and my bike stolen. I don’t recall bothering to report it.
Property crime and armed robbery has been typically falling for the last 40 years or so. Individual physical pieces of wealth tend to be worth less than in the past. ATM machines have allowed me to take out a maximum of $300 for the last third of a century. Car stereos cost about, I don’t know, $300 back then and about $300 today, so they get stolen less today.
Plus, there has been massive target hardening over the decades.
Why should crime go down … unless we are screwing up?
Because crime is a bad thing, like infant mortality or airliner crashes. We invest a lot in reducing it. If we’re not reducing it, we’re probably doing something obviously wrong, like encouraging blacks to be angry at whites.
And the total number of homicides last year was fewer than 20 years ago even as the country’s population increased, criminologists said. There were 19,645 homicides in 1996 in a nation of 265 million; in 2015, there were 15,696 in a population of 321 million.
Look, homicides should be falling perhaps 5% per year or things are getting worse.
What that data means, criminologists and police officials said, is that the decline in homicides has been so significant in the last quarter century that sudden increases in the number of killings in just a few cities can skew the entire national picture, even as the country has one of its safest periods on record.“It isn’t a national trend, it’s a city trend, and it’s not even a city trend, but a problem in certain neighborhoods,” said Richard A. Berk, a professor of statistics and criminology at the University of Pennsylvania. “Certainly, people around the country should not be worried. People in Chicago shouldn’t be worried. But people in certain neighborhoods might be.”Criminologists and police officials point out that homicides do not usually disrupt entire cities. Instead, they occur in particular neighborhoods — and on the same blocks — leaving much of the rest of the city relatively untouched.
So, when you go to vote, don’t worry about America becoming an increasingly homicidal nation. It’s not who we are. It’s just those people who live on particular blocks.
Explanations for the increase in homicides in certain American cities are largely guesswork. Criminologists acknowledge that the required analysis has not been done in the neighborhoods where killings are occurring — or even an agreement of what such a study should include — to arrive at any but the broadest conclusions.“Why is there a rise? No one can possibly know that,” said Robert Smith, director of the Fair Punishment Project at Harvard Law School. “But crime statistics are most accurate when they are block by block, because even a neighborhood is too large.”Still, attempts at explanations are numerous: battles over drug dealing turf; the dissolution of once powerful street gangs, resulting in violent crews that disband as quickly as they form; petty disputes that turn deadly because of the ready availability of guns; and a deepening economic and social isolation of the nation’s poorest.
Or maybe it has something to do with … Black Lives Matter?
I wouldn’t rule out some drug market effect, but I haven’t been able to find one in about a year of looking.
If you look at the history of homicide rates, there were three periods of upswing:
- The Sixties, which ran from about 1964-1975, which was a combination of multiple drugs and political liberalism that reduced sentencing, discouraged cops, and riled up minorities and white hotheads.
- The brief powder cocaine spike around 1980.
- The big crack cocaine from about 1988-1994.
It could be that the last two years are caused by fighting over some drug, but, personally, I’m not down with the streets so I don’t know what it would be. The other huge possibility is it’s just a reprise of Sixties Liberalism.
Crime experts say that studying a single year of crime data — instead of a decade or 20-year periods — reveals little about crime, especially if a particular year runs counter to the current long-term trend of falling crime.
It appears to be at least a two year trend running from Ferguson in August 2014 at least through the first half of 2016.
“Crime rates are well known to be volatile, and year-to-year variation does not really tell you much,” Dr. Berk said. “The reasons are that crime rates are driven by many factors at the neighborhood and even interpersonal level that we do not measure and about which we have little understanding.”Inimai M. Chettiar, director of the Justice Program at the Brennan Center for Justice, a nonpartisan public policy group, said that the nation might have entered a period in which some types of crime, particularly homicides, would regularly seesaw from year-to-year in certain cities, but neither rise nor drop significantly.“There have been predictions that as crime decreases, we will see more up and down movements, but that crime itself will stay relatively steady,” she said.The patchwork that has emerged from the steep crime reductions over the past generation has meant that Baltimore, Milwaukee and Washington — where murders increased in 2015 — have had fewer homicides this year. Baltimore is on pace to have its murder rate drop 9.7 percent this year, and Washington by 12.7 percent, according to a study by the Brennan Center.
Or Freddie Gray didn’t get killed in 2016 in that region, he got killed in 2015.Slate
’s article is a little better:
Should the Media Downplay the New Murder Spike?Depending on your politics, the FBI’s new stats can look scary or meaningless. Here’s a more honest reading.By Leon Neyfakh
But the author can’t gin up the courage to mention Black Lives Matter, either.