My general stance toward government-provided information is what I consider “healthy skepticism.” If the state is making a factual claim, the first question I ask is, “Would manipulating this information expand the government’s power?”
That question is necessary because politicians have an incentive to choose information they share or to frame shared information in a way that enhances their own authority and job security.
I mention this to give some context to new data about the crime rate in 2017, released this week by the Brennan Center for Justice using “crime data [collected] directly from local police departments in America’s 30 largest cities.” As The Washington Post summarizes, “2017 is on pace to have the second-lowest violent crime rate of any year since 1990.” The lowest year, 2014, is also on the recent end of that scale.
This information, which was assembled by the Brennan Center but comes from government sources, is consistent with other government-provided information about public safety in recent years: Crime is down and police are safer than ever.
You don’t have to take my word for it. These conclusions are based on stats law enforcement agencies themselves are reporting.
For example, these striking graphs, created by the conservative American Enterprise Institute to show how dramatically safe police work has become compared to decades past, are based on data from the Officer Down Memorial Page, a law enforcement-run tally of police deaths in the line of duty. Information from the FBI and the National Law Enforcement Officers Memorial Fund (NLEOMF) says the same thing: For decades, American police have been getting safer. A police officer today is less than half as likely to suffer a felony killing while on duty as compared to the 1970s.
Likewise, as Charles C.W. Cooke noted at the conservative National Review citing government statistics, violent crime “is actually on the way down” compared to the supposed good old days. We are objectively much safer than we used to be. Crime still happens, of course, but things are getting better.
A couple years back, I talked to Christopher Preble, a scholar at the libertarian Cato Institute, about how governments and politicians use data to their own advantage for a column about terrorism over at The Week. “[T]he incentive structure is such that political figures and public officials can really only be held to account for bad things happening,” Preble explained:
Thus politicians tend to exaggerate threats so they can say “I told you so” if the threat comes true and “Look, I kept us safe” if it doesn’t. Unfortunately, this incentive structure means a great deal of blood and treasure is spent fighting threats that aren’t very threatening. Even more fundamentally problematic, Preble argued, “We are terrorized. We are frightened as a society because we are focused on the still unlikely, but understandably frightening, prospect of terrorism.” This acceptance of terror plays into terrorists’ central goal of forcing policy change via the demands of a fearful population.
Terrorism isn’t the only policy area where this dynamic is at work. It is generally in governments’ interest to hype threats — to make the general public afraid — so that we will accept or even demand expansions of state power we would not otherwise approve.
And that skepticism about government-provided data is exactly why I trust these new crime numbers: They work against the government’s incentives.
In court, a jury is usually more likely to trust a witness if they’re providing testimony against their own interests. The logic being that this person wouldn’t share information that is embarrassing, legally risky, or financially costly to themselves were they not so certain of its truth.
In the same way, it is very much in the government’s interest to hype crime rates and attacks on police officers as a way to suggest our country is dangerous and we need to expand government power for our own safety. Authoritarianism is fueled by public fear.
So when the data is so clear that crime rates are dropping and police work is less dangerous that even the government has to admit it, you know it’s true, because that’s the last thing the government would want to admit.
It’s hard to bamboozle the public into signing away liberty for security if we know we’re already historically safe.