Australian gun control

Australian homicide rates

This is a quick and dirty response to the Young Turks video about Mark Hamill‘s Twitter comments about gun control in the US and in Australia:

First off – let me say that I think there SHOULD be gun control. I am not a friend of the NRA (see, I’m not even linking to their website). I also am a big fan of The Young Turks. And Mark Hamill for that matter.

So when I was watching the video (I’ll embed at the end), I was pleased to see that they have all these statistics about how Australian National Firearms Agreement in 1996 brought down suicide and homicide numbers – yay! They quote this sort of thing:

The average firearm suicide rate in Australia in the seven years after the bill declined by 57% compared with the seven years prior. The average firearm homicide rate went down by about 42%

But, I couldn’t help but notice they kept talking about it bringing down ‘firearm-related’ deaths… not just deaths. This got me thinking – did it actually prevent death or just gun related death?

Here’s a quick and dirty, back of the envelope look at this. The Turks mention suicide and homicide, so we’ll look at both of those.

Suicide

Looking at Australian government suicide statistics, we can find a table for 1994-2004 – this straddles the gun control legislation (1996) usefully. Also check out this meta-analysis of Australian suicide rates and methods 1988-2007.

Australian suicide rates
Australian suicide rates for male and female between 1994 and 2004, indicating a peak in 1997 and 1998

What the data shows is that there was a peak in suicides in 1997 and 1998. This spike happens AFTER the mass giving up of guns in 1996-1997. To quote the authors of that meta-analysis paper:

This peak was associated with a near doubling of the rate of hanging, and occurred despite a 60% fall in firearm suicides between 1988 and 1998.

So what the data is showing is that if you remove the guns, you remove firearms suicides (as the Young Turks were saying), but you do not remove the will for suicide. Suicide is a bigger problem than guns.

Homicide

The Young Turks quote a load of firearm-related homicide stats but they also point to a decrease in overall homicides during the time that the firearm-buyback was in place:

1996 and 1997, the two years in which the NFA was actually implemented, saw the largest percentage declines in the homicide rate in any two-year period in Australia between 1915 and 2004

The problem here (and I wish there was no problem!) is that the Australian Government homicide statistics for 1993-2007 show that there was a bounce-back in homicides to even higher levels in 1999:

Australian homicide rates
Australian homicide rates for 1993 to 2007, showing record dip in 1998 and spike to higher levels in 1999

Now, the ever-charming Cenk Uygur almost acknowledges this after making that quote, he says “I know, coincidence, coincidence…” (see video around 7m23s). He then goes on to back up his point by quoting another statistic that is purely related to firearms deaths. Sadly, this is not valid. Yes, firearms homicides go down when you have gun control, but the stats from Australia indicate that homicides are not significantly affected.

Mass shootings and accidental death

What these stats show is that humans are very inventive, resourceful and determined. Removing firearms does stop people from using firearms for suicide and homicide but it doesn’t necessarily reduce the total number of either of those things – where there’s a will, there’s a way. Intentional deaths are hard to stop and these intentions stem from sources that are not affected by methods available.

Two areas where this might not be the case would be accidental death, where there is no will or intention behind the tragedy, and mass murder where guns simply make the objective so much more achievable.

In one paper that looks specifically at the effect of the Australian 1996 gun laws using data from 1979 to 2003, they observe that there were 13 mass shootings in Australia between 1979 and 1996 but absolutely none between the gun reforms of 1996 and their latest data of 2006. However, this again is somewhat selective.

For those authors, a mass shooting must actually achieve the deaths of at least 5 people. But in 2002, the Monash University Shooting occurred with 2 deaths and 5 injuries – not making it into their stats but clearly being an attempt at the same sort of atrocity. A quick look on Wikipedia for Australian Mass Murders  indicates that subsequent to that paper’s data, there have been several although predominantly from arson and knife or hammer attacks (humans are resourceful) but including the Hectorville Siege where 3 people were killed and 3 injured in another mass shooting.

What can be said is that there have been decidedly fewer mass shootings and that the numbers killed in each event have been fewer since the gun control laws. Personally, I’d say that was a decent argument for gun control.

I think we’d all expect that accidental death would drop after gun control, as said, there is no will for an accident so there’s no chance of human inventiveness getting around the system… however I’m going to have to take my own medicine here and show you that the Australian data for 1979-2003 indicates an INCREASE in accidental death. To quote the authors of the paper,

The finding that there was a significant increase in unintentional (accidental) firearm deaths after the new gun laws is perplexing

No shit! There may be ways to explain this away such as the number of accidental firearm deaths is relatively small so perhaps this was a statistical glitch, or perhaps most accidental firearm deaths occur when hunting or cleaning with legitimate firearms and therefore would not be controlled by the laws… but at the end of the day, this post is about going with the data – wherever it leads.

 What the stats might indicate

You may have noticed that I’ve said a couple of times that the roots behind suicide and homicide stretch deeper than merely having access to the means of carrying them out. Although I do agree that gun control would help reduce mass shootings (in frequency and rate of fatalities), and I hope that data will show reduction in accidental death – if America really wants to reduce its homicide and suicide rates, here’s what I think the data shows.

The suicide rate in Australia has dropped considerably since its peak in 1998. Something else that’s happened in Australia since then is a massive growth in its economy.

Australia GDP
Australian GDP from 1998 to 2013

Here, I’m trying to point to a possible link between financial stability and a reduced tendency to suicide or murder. For example, the global financial crisis has been linked to an increase in suicides and, indeed, the notion of bankers jumping out of Wall Street windows has long been associated with the crash of 1929, where record numbers of suicides (23000) were recorded in that year.

Now GDP doesn’t necessarily translate to a happy society but Australia has generally a relatively low Gini coefficient indicating that when its GDP grows, it’s spread around ‘fairly’ well. What I’m about to suggest is that where income stability might be linked to suicide, income inequality might be linked to homicide.

Bringing it back to the US, we can find the stats for homicide rate per state, and also the Gini coefficient (income inequality indicator) per state (for  2010). The data in the table below produces a Pearson correlation of 0.57 with a P-value of 1.77e-05. This shows there is a shared trend (in 2010) for homicide and income inequality.

US StateMurder RateGini Coefficient
Alabama5.70.472
Alaska4.30.422
Arizona6.40.455
Arkansas4.60.458
California4.80.471
Colorado2.60.457
Connecticut3.70.486
Delaware5.70.44
Florida5.20.474
Georgia5.70.468
Hawaii1.80.433
Idaho1.40.433
Illinois5.50.465
Indiana4.10.44
Iowa1.20.427
Kansas3.40.445
Kentucky4.30.466
Louisiana110.475
Maine1.80.437
Maryland7.40.443
Massachusetts3.30.475
Michigan5.90.451
Minnesota1.80.44
Mississippi6.90.468
Missouri70.455
Montana2.50.435
Nebraska30.432
Nevada5.80.448
New Hampshire10.425
New Jersey4.20.464
New Mexico6.80.464
New York4.50.499
North Carolina50.464
North Dakota1.50.433
Ohio4.20.452
Oklahoma5.20.454
Oregon2.50.449
Pennsylvania5.10.461
Rhode Island2.80.467
South Carolina5.70.461
South Dakota2.80.442
Tennessee5.60.468
Texas4.90.469
Utah1.90.419
Vermont1.10.444
Virginia4.70.459
Washington2.30.441
West Virginia3.10.451
Wisconsin2.70.43
Wyoming1.40.423
US state gini coefficients and homicide rates
US income inequality correlating with homicide rate per state in 2010

Now on a post regarding statistical accuracy in political debate, I’m not going to say that correlation is the same as causation! This is just a quick and dirty calculation using the easy to find data and doesn’t prove anything. I’m just proposing that the discussion ought to include a look at inequality in the US.

Conclusion

I think the Young Turks have been guilty of misrepresenting the data from Australia. I’d hate to think it was intentional, but Cenk Uygur does indicate his knowledge of the shaky ground he’s on when he says  ‘I know, coincidences, coincidences’.

The problem with these things is that although they advance your argument in the short term, they are very easy for the opposition (e.g. the NRA) to debunk and therefore bring doubt upon every other argument you make.

The data probably does indicate that you can achieve a reduction in mass shootings through gun control. This alone should be a decent argument for gun control in the US. I want to reiterate again and again, that I’m FOR gun control in all countries.

I’d be hopeful that some data might show that there is reduced accidental death following gun control – but the Australian data didn’t show it.

The key issue for me is that where human will is concerned, these tragedies will happen. Therefore the key discussion needs to be on reducing human will for violence and suicide. That argument needs to address not just the mechanisms (guns) but also the societal problems that lead to one human losing empathy and gaining hate for other humans (in the case of murder) or coming to the decision to end their own life.

Financial issues are a major problem in people’s lives and income inequality may be an indicator of societal issues that lead to increased crime. The data shown here indicates that the US may need to tackle income inequality if it wants to reduce homicide.

The Young Turks Video

Leave a Reply

Your email address will not be published. Required fields are marked *