Liars, damn liars, and gun nuts.

...or how figures lie, but liars sometimes don't figure too well.

Someone yesterday tried to prove to me that gun bans don't reduce murder rates. They sent this link...

...and something didn't look right. I've been looking at noisy data for nigh on 40 years, trying to see signal in the murk, and that spike in 2002 seemed really odd.

Now I have family in the UK; I had more in 2002. I subscribe to the London Times. I go back and forward, though not as often as I'd like. And I was aware of nothing that might have caused a large number of homicides in 2002. So I did a little digging.

And, remarkably, the spike in 2002 is the work of one man, a man you've probably never heard of, named Dr. Harold Shipman. Shipman was a general practitioner in the UK. He became a GP near Manchester (where I was born and lived the first eight years of my life). And then he started murdering people. He killed mostly older women, in order to steal their jewelry, or after having them write him into their will. He was arrested in 1998, and convicted of 15 murders in 1999. Following this, an official inquiry reported in 2002 that he was responsible for at least 215 murders. (The final estimate was at least 250, but some could not be proven).

15 of the homicides appear in the 1999 statistics, 28 in the 2000 statistics, and 172 in the statistics for 2002. But the actual murders occurred between 1975 and 1998. So let's now take the murder statistics for England and Wales, subtract Shipman's contributions for 1999,2000 and 2002, and share them out evenly from 1975 - 1998, about 16 per year. I've also divided total number of homicides by the population of England and Wales, and multiplied by 100,000, to get a standard rate per 100,000 per year.

The second, most stringent Firearms Act went into effect in early 1998, and parliament estimated that surrender of prohibited firearms was essentially complete in January 2000. Of course, homicide statistics lag a little behind murders, because the death has to be discovered and ruled a homicide. That being said, the murder rate was clearly increasing through the 1990s, peaked in 2001, and thereafter fell dramatically. It is not unreasonable to attribute the rise up to 2001 as due to an ongoing trend, and the dramatic fall by a factor of almost 2 after 2001 to the handgun ban. So the data prove the opposite of what was claimed.

This isn't unique, by the way. You find the same shite all over the net. Sometimes the axes are drawn a little more honestly. On the other hand, the Mises Institute also lied about Ireland. The spike in 1974, following the 1972 confiscation, was actually due to the Dublin-Monaghan bombings, a terrorist act with which I am far too intimately acquainted (I was close enough to be knocked over by one of the car bombs). And the long term trend in Ireland is probably an aftermath of the IRA campaign; while the IRA's guns were supposed to be decommissioned, a lot found their way into the hands of criminal gangs.