In the Anti-Reactionary FAQ, Scott Alexander argues that rising crime statistics are an illusion.

See this graph:

Seems pretty damning, right? Well, Alexander makes a few points that the graph may be biased, though it seems doubtful to me that they account for the total increase:

Question number one: what does this graph mean by “indictable offenses”? This very broad term introduces no fewer than three dangerous biases. First, we have reporting bias – the more police there are and the more active there are, the more crimes get heard about and reported. Second, we have definition bias within individual crimes – for example, larceny in Britain fell by two thirds in 1855, but this was because Parliament passed a law raising the minimum amount of property that had to be larcened for it to count. Third, we have broader definition bias in what is or isn’t a crime – how much of that rise around 1970 was the “indictable offense” of people smoking marijuana, something that was previously neither illegal nor widely available?

Criminologists’ recommended way around this problem is to look at murder. The murder rate tends to track the crime rate in general. Murder isn’t as subject to reporting bias – if someone is killed, the police are going to want to hear about it no matter how understaffed they are. And murder is less subject to changes in definition – dead is dead.

So let’s add the homicide rate to the above chart:

Hm, so homicide in the United Kingdom is simply staying the same… violent crime has skyrocketed, but it’s gone back down a bit since 2000 (probably due to increased policing acting as a kneejerk immune response to the horrifying violence), so that’s nice.

How about America? Alexander implied I was concealing the American data to try and boost my point, so now is my chance to come clean:

Well, it looks like the homicide rate has definitely gone up since a hundred years ago, but has actually gotten better in the last two decades. Right?

Wrong.

Medical advances over the past 40 years have masked the epidemic of violence by lowering the murder rate by more than five times.

According to new research, doctors are saving the lives of thousands of victims of attack who four decades ago would have died and become murder statistics.

Although the study is based on US data, the researchers say the principle applies to other countries too: “There is reason to expect a similar trend overall in Britain,” said Dr Anthony Harris, the lead author of the study.

In the research he and a team from Massachusetts University and Harvard Medical School found that technological developments had helped to significantly depress today’s murder rates, converting homicides into aggravated assaults.

“Without this technology, we estimate there would be no less than 50 000 and as many as 115000 homicides annually instead of an actual 15 000 to 20000,” they say in a report of the study in the journal Homicide Studies (2002;6:128-66).

The team looked at data going back to 1960 on murder, manslaughter, assault, and other crimes. It merged these data with health statistics and information on county level medical resources and facilities, including trauma centres, population, and geographic size. The researchers then worked out a lethality score based on the ratio of murders to murders and aggravated assaults.

They found that while the murder rate had changed little from a 1931 baseline figure, assaults had increased. The aggravated assault rate was, by 1997, almost 750% higher than the baseline figure.

Aggravated assault is up 750% since 1931, and the murder rate, if it weren’t for better medicine, would be between 7.5 and 28.75 times greater today than in 1900! So much for falling crime.