Thursday, 4 April 2019

Gun buyback: Correlation is not causation [updated]

THE LABOUR-LED GOVERNMENT'S compulsory 'gun buyback' legislation is said by the Deputy Prime Minister to cost somewhere around $300 million. That would buy a lot of policemen.

So clearly the Labour-led Government, along with everyone else in Parliament except ACT's David Seymour, has concluded that this compulsory buyback will keep New Zealanders a lot safer than a lot more policemen will.
That doesn't say a lot for our policemen. (Nor does that figure say anything about the realistic cost of the buyback, which is more like billions than millions.)

But what do we know about how much safer the compulsory gun buyback will make us?
The fact is, neither the Labour-led Government, nor anyone else in Parliament really knows, because they haven't and won't have the time to do that research.

What they are really relying on is the alleged popular success of the Australian compulsory gun buyback after the Port Arthur massacre. So it's worth asking just exactly how successful that gun buyback was in making Australians safer. So I went to look for research that did look at how successful that had been in reducing lethal violence. In 2016 Science Direct published 'A systematic review of quantitative evidence about the impacts of Australian legislative reform on firearm homicide,' which concluded:
Australian studies have not found evidence of changes in lethal violence following gun law reform. Empirical findings about Australian gun law reform contradict ‘popular’ views about those laws...
    These [studies] examined various different time periods, and used a range of different statistical analysis methods. No study found statistical evidence of any significant impact of the legislative changes on firearm homicide rates.
That sounds fairly conclusive, right?

In fact, if you look at the rate gun deaths from 1998 to 2014, you would think New Zealand and Australia already had the same restrictions on guns:

Source: GunPolicy.Org

And yet Australia already has these restrictions that New Zealand politicians are now eagerly rushing through.

So what is going on here? Why do those two declining figures (great news, by the way!) seem about the same even though the two country's gun laws are so different? Why does the popular 'knowledge' of the Australian buyback success not tally with the Australian studies that have found no evidence of changes in lethal violence following gun law reform? Why, in summary, do empirical findings about Australian gun law reform contradict ‘popular’ views about those laws.

And what does it say about the gun buyback programme that the decline began nearly a decade before?

The simple answer from the researchers is this: that the rates of gun death were going down in any case. The evidence from those researchers is that the buyback did not cause the decline (which as you can see form the graph above began before 1996) it simply correlates with its continuation.

Correlation is not causation. Perhaps the popular view is so different because that simple lesson is still widely unlearned.

SO WHAT DO THE the results of Australian research on their buyback programme over there tell us we should expect to see as a result of spending around $300 million on the compulsory buyback programme here? "The results," says one, "suggest that the [buyback will] not have any large effects on reducing firearm homicide or suicide rates." In short, concludes another:
Although gun buybacks appear to be a logical and sensible policy that helps to placate the public’s fears, the evidence so far suggests 23 that in the Australian context, the high expenditure incurred to fund the 1996 gun buyback has not translated into any tangible reductions in terms of firearm deaths. 
As I was saying, $300 million would buy an awful lot of policemen. Not to fool ourselves that more policemen would have foiled this or any other shooting -- because a police response time of 36 minutes even of the almost-miraculous 6 minutes* to arrive at a site where unarmed people are being shot is telling proof otherwise -- but if we must have knee-jerk law and rushed spending decisions as a result of this massacre, why not head down that path instead of criminalising otherwise law-abiding gun-owners. (Not forgetting that the murderer himself broke existing gun laws in modifying his weapon before his lethal spree, reminding us that no matter what laws are passed, criminals -- like gangs -- will still ignore them whenever they feel like it.)

AND IF WE ARE serious about avoiding another atrocity like this one -- which is after all the alleged aim of this expensive legislation being rushed through with such unseemly haste -- then another unanswered question seems to present itself, which is this: what would you might most want to have with you when a man with a gun bursts through your door and starts shooting? A policeman to defend you all would be a fine thing to have, but in their absence – and experience from around the world tells us that a policeman can never be there in time to defend us – what you may most want is something to scare the gunman off. Just as this arsehole-with-a-gun was finally frightened away by a very brave man threatening the coward with an EFTPOS machine and with a shotgun the gunman had already discarded.

Because self-defence is still legal in New Zealand, just, under Section 48 of the Crimes Act “If one fears for their life or that of another.” Good law. The very same law that police are covered by when using firearms for their activities. Law however that is not set in stone, and that the police have for some time been wishing to overturn.
Could the outcome of the Mosque shootings have been different if the police upheld the law on self-defence instead of vilifying anyone who uses firearms for self-defence? We can never know.

After the Christchurch atrocity, and the political reaction to it, crime researcher Dr. John Lott asks the obvious question:
Police are extremely important in stopping crime, but the police can’t be there all the time. The police themselves understand that they virtually always arrive on the crime scene after the crime has occurred. And that raises a real question, what should people do when they’re having to confront a criminal by themselves?
Anyone like to have a crack at answering that? Because your politicians haven't. And won't.

But you should keep asking it.

We do know already that this rushed legislation will be followed by other more considered legislation, imposing further restrictions on gun ownership, and considering again a programme of costly gun registration. If there is an agenda, it will become apparent then. We must hope, and remain vigilant, that the agenda does not go from vilification of this legal right to removing it from the books altogether.

Because then where would we and other brave men be when we do need to defend ourselves? 

THIS IS NOT AT all to say that a government has no moral right to regulate weapons. Of course they do. Governments (properly) hold the legal monopoly over the use of force in a given geographic area. That's a fundamental definition of what a government is. Being a primary means of projecting force means that weapons and the regulation thereof must be permanently on their radar. But by what principle should this be done? As philosopher John Mccaskey patiently explains
To see why it is proper for a government to regulate weapons and to understand the principles by which it should, we need to go back to some fundamental principles of moral philosophy, political philosophy, different kinds of rights, and the nature of government... 
    You have a natural right to defend yourself against an attack, using unlimited force if necessary. But it still might rightly be illegal for you to own or carry a gun... 
      Remember, the proper question is not, 'Why can the government restrict my access to guns?' The proper question is, 'What share of its legal monopoly on the use of force should the government share with its citizens?' The proper answer is, 'Whatever is needed for those citizens to protect themselves when the government cannot.'
Those remain a Q+A that this government, and this country, still need to have. Why don't you begin asking and answering it for yourself?

* * * * * 

* The New York Times lays out the probable response timeline which, however rapid, still allowed the murderer to leave the first place of carnage and drive across town to create another:
It is unclear exactly what time the gunman entered Al Noor, which was crowded with worshipers for Friday Prayer. But the police said that they received the first call for help at 1:41 p.m., and that the first officers arrived there six minutes later.
    The video recorded by the gunman, which was livestreamed on Facebook, showed a man trying to tackle him inside the mosque, only to be shot and killed.
    Six minutes after firing his first shot, he drove away. Three minutes later, a siren can be heard on the video as he is driving to the second mosque.
    The siren becomes louder, then fades, suggesting the police and the gunman may have just missed each other, with officers and medical personnel racing toward Al Noor as he was pulling away. 

    About 30 front-line police officers would be on the streets of Christchurch around lunchtime on an average Friday, said Chris Cahill, a detective inspector who is president of a local labor union for police officers.
    When that first panicked call came in, he added, the dispatcher would have sent all of them to Al Noor….

    The police said a special armed tactical unit arrived at Al Noor Mosque four minutes after the first officers, or 10 minutes after the initial emergency call…
    “Any police force in the world — to get to the scene in six minutes, a specialist team there in 10 — that would be a success,” Cahill said.
Patrick Skinner, a former C.I.A. counterterrorism officer now working for an American police department, agreed.
  “I’d say that the police response was rather quick in a tactical sense,” he said, noting that the officers were rushing into a violent situation that was still unfolding — and that had been encouraged by individuals espousing bigotry and hatred.
    Still, it was not fast enough. The officers arrived to a horrific scene, with the dead and wounded outnumbering the city’s usual on-duty police force.



  1. A well researched and thought out article

  2. You have to be careful with the Australian case, the National Firearms Agreement (NFA) had several aspects to it. One of these aspects was a buyback scheme. There is some evidence that the Australian reforms as a whole reduced suicide and homicide rates. Leigh and Neill (2010) say "In 1997, Australia implemented a gun buyback program that reduced the stock of firearms by around one-fifth. Using differences across states in the number of firearms withdrawn, we test whether the reduction in firearms availability affected firearm homicide and suicide rates. We find that the buyback led to a drop in the firearm suicide rates of almost 80 per cent, with no statistically significant effect on non-firearm death rates. The estimated effect on firearm homicides is of similar magnitude, but is less precise. The results are robust to a variety of specification checks, and to instrumenting the state-level buyback rate".

    But they also say "Perhaps a more likely explanation of the strength of the relationship found is that the NFA led states with relatively weak legislation or enforcement relating to sale, ownership and storage of firearms to strengthen their regimes relative to states with initially stronger standards. There is evidence that states with relatively high firearm ownership and therefore high gun buyback rates also had relatively weak regulation prior to 1996. Then, our estimates need to be interpreted as reflecting a combination of both the removal of firearms and the relative strengthening of legislation and enforcement. We might expect to see smaller effects in the case of a buyback that was not accompanied by stricter firearm legislation". Thus there was more going on in the Austrian case than just a buyback and its difficult to know which bits of the reforms drive the results.

    Leigh, Andrew and Christine Neill (2010). "Do Gun Buybacks Save Lives? Evidence from Panel Data", IZA Discussion Paper No. 4995, June.

  3. Just a minor correction. Armed police were on the scene within six minutes of receiving notification and apprehended the offender within 21 minutes.

    Of course the offender still managed to kill fifty unarmed people in that time, so your general point remains sound.

  4. I'd be curious as to what the trends of suicide were in the decades bracketing their gun confiscation to see whether or not those former gun owners simply turned to other means.

  5. I think your statements around police response time aren’t correct, and if anything they performed better than we could reasonably expect. This makes the point even stronger that we can’t rely on them to protect us all the time.

  6. Thanks for the corrections on the timelines, people. I've corrected the post.
    Paul, I'm pretty sure the study I mention above analysed the results of several studies, including that one you mention (but I'll have to look at it again to make sure.). But the main point here remains is that the downward trend was happening in any case, and matches that same downward trend in New Zealand (and also in Canada, the UK and the US) which did not enact any similar scheme.


1. Comments are welcome and encouraged.
2. Comments are moderated. Gibberish, spam & off-topic grandstanding will be removed. Tu quoque will be moderated. Links to bogus news sites (and worse) will be deleted.
3. Read the post before you comment. Challenge facts, but don't simply ignore them.
4. Use a name. If it's important enough to say it, it's important enough to put a name to it.
5. Above all: Act with honour. Say what you mean, and mean what you say.