Beware.

Photographer: Scott Olson/Getty Images

Maybe Crime's Not Rising, But If Voters Think It Is ...

Megan McArdle is a Bloomberg View columnist. She wrote for the Daily Beast, Newsweek, the Atlantic and the Economist and founded the blog Asymmetrical Information. She is the author of "“The Up Side of Down: Why Failing Well Is the Key to Success.”
Read More.
( Updated
)
a | A

At the Republican Convention in July, Donald Trump intoned that “Decades of progress made in bringing down crime are now being reversed by this administration’s rollback of criminal enforcement.” The media rushed to state, equally firmly, that he was wrong. And then this week brought the announcement that violent crime in the U.S. had indeed risen significantly last year, by almost 4 percent. Murder was up even more.

Was Trump right?

First, the data. Violent crime rose 3.9 percent in 2015. Murders were up 10.8 percent, rape 5.1 percent, aggravated assault 4.6 percent. Economic crimes like robbery, burglary and auto theft either decreased or stayed close to flat.

That’s certainly troubling. Any increase in violent crime is, to state the obvious, bad. But America has experienced a stunning decline in crime over the last 20 years. Since it was unlikely that a country of 300 million people was ever going to reach the point where we had no murders, no rapes, no aggravated assaults, we always had to expect that at some point, that decline was going to level off. And when it did, even if it basically stayed level, some years it would rise.

Hmm. "Level"? You may be thinking: "I do not think that word means what you think it means." But we do not live in a textbook model where lines are perfectly straight and humans behave like probability distributions. In the real world, level means “small variations around a flat trend line.”

In some years, people will commit a few more crimes than you’d expect, because an unusual number of places have a local gang war, or simply because individual criminals, for reasons we’ll never know, decided to be a little more active. In other years, things will be a little more peaceful than usual. For example, murders in 2004 were probably a little below the average tendency at the time, 2007 a little higher. But those variations were not much noticed, because the overall direction was down.

As the line flattens out, however, those variations around the average tendency to commit crimes will get noticed, because they’re now the only thing that’s changing. If the trend line ends up truly flat, then all the year-over-year changes will, in retrospect, be dismissed as random noise. And while obviously it would be nice if the line had kept going down until it reached zero, if this is the endpoint where it stabilizes, that will still be a remarkable achievement, with murder down by a third since 1996.

So even a substantial one-year movement in the numbers is not reason to freak out and declare that the Obama administration has let everything go to hell in a handbasket. For one thing, very little crime policy happens at the national level, meaning Trump was really really wrong when he referred to "this administration’s rollback of criminal enforcement."

For another, this could be simply a blip. At the moment, violent crime remains extremely low. We have not even gone back to the relative disorder of 2012. Crime is just a little higher than its record lows in 2014. Or, to be precise, it was. The data we're talking about is from 2015. By now the crime rate could be well below the flat trend line. Or soaring.

So … at what point should we start to worry? Only if the jump from 2014 to 2015 ultimately turns out to be the beginning of a new, upward trend line. At what point can we go back to dismissing Trump's entire claim? Only when we see evidence that the trend line remains flat or headed downward.

For now, the most responsible reaction to the 2015 crime data is neither “the sky is falling” not “Move along, nothing to see here.” Vigilance for 2016 and beyond is appropriate.

And it’s likely this will play a role in the coming election. Whatever the underlying reality, people are more concerned about crime than they were a few years back, and as politicians discovered in the 1960s, the public is very sensitive to any perceived increase in public disorder. I suspect that those perceptions have more to do with the riots that have filled our television screens in the last two years than they do with a significant increase in the average American's personal danger of being victimized. But whatever the cause, politicians will have to contend with the effects.

(Corrects when the crime statistics were released in first paragraph.)
  1. Obviously murder is the most serious violent crime, and also the hardest for police to finesse out of the crime stats by discouraging reporting or reclassifying major crimes as less serious ones.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Megan McArdle at mmcardle3@bloomberg.net

To contact the editor responsible for this story:
Philip Gray at philipgray@bloomberg.net