Monday, October 12, 2015

October 12, 2015 - The Dreadful Gun Control Debate

Last Friday, President Obama traveled to Oregon to meet with family members and victims of the recent Umpqua Community College shooting. So the gun control debate is going to continue to dominate the news for the foreseeable future, and today it will reluctantly dominate The Daily Face Palm as well.

I say reluctantly because the quality of political discourse on this subject is rather abysmal. Both sides of the debate argue as if their position is so obviously correct that they don't even need to bother with attempts at persuasion. So they don't. The left points to a massacre and the high number of gun deaths in the US; the right says, "Look at Chicago!"* and guns don't kill people... and I'll spare you the rest of that sentence. That is a slight oversimplification of course, but you get the gist.

*Aside: Chicago has a high murder rate and relatively strict gun laws, which means it's perfect if you're lazy and looking to make a shallow argument against gun control. One city with a correlation that could be driven by 100 different factors is meaningless. A much more substantial point for those wanting to push the "gun control doesn't work" argument can be found here, which actually relies on a sample size large enough to be relevant (shockingly, it's larger than one). More on this later.

In light of this unsavory state of affairs, I would like to introduce four points that will hopefully inform the conversation:
  • Almost everyone believes in some level of gun control.
  • We do not know whether more gun control would work to reduce violence or not.
  • Gun control laws create gun-related crimes, some of which have no victims.
  • Gun control laws are likely to have a disproportionate impact on minorities.

I've summarized each of these briefly below.

Almost everyone believes in some level of gun control.

This is important to acknowledge because it is actually a rare point of common ground on this issue. For most people, there is no absolute pure principle they're fighting for. They may pretend that they want no restrictions on gun ownership, but most do favor some restrictions on private ownership of weapons generally. If you meet someone who claims otherwise, ask them to make a case for why private individuals should be allowed to own scud missiles and anti-tank weapons. I suggest to you that the conversation will end in your favor shortly. Most people don't like the idea of a privately run military force (even if our government currently funds several to fight abroad).

Now that everyone is on the same page that some restrictions are acceptable, that's a much more promising starting point to discuss what restrictions we should in fact have.

We do not know whether more gun control would work to reduce violence or not.

The effectiveness of gun control is generally just assumed by both sides of the debate to be whatever suits their argument. Liberals, broadly speaking, assume it would definitely work, and conservatives frequently assume it would make matters worse. And naturally, neither side is really being honest here.

Ultimately, it is an empirical question whether intensive gun control legislation would reduce violent gun deaths. It may be true, but it is not inherently true. There are no laws of physics at work here; we're dealing with human behavior and politics. And there is good reason to be skeptical of its effectiveness; after all, the government passes lots of laws aimed at a particular purpose that it never achieves. Remember when the courageous war on drugs completely ended illegal drug use in America? Nope, me neither.

So what does the data show? And the answer depends on how you look at them. The pro-gun control side will often tout results from Australia (which passed intense gun control regulation in response to a mass shooting in 1996) or show the general correlation between gun deaths and gun ownership across states in the US or something along those lines. If you look at those resources, however, you'll notice that they're focusing on gun deaths (suicides or homicides), so in some ways their results are unsurprising. Fewer guns are liable to lead to fewer gun deaths, fair enough. But do they lead to fewer overall violent deaths? Eugene Volokh at The Washington Post points out that gun deaths aren't really the statistic we care about. We primarily want to know if reducing access to guns will decrease the incidence of violent homicides. And just as with other goods in an economy, restricting access to guns is liable to create a substitution effect whereby murderers who would use firearms switch to another weapon that can achieve a similar purpose. And Volokh accurately notes that from the point of view of the victim, it's hard to see how being stabbed is much better than being shot.

At the end of the day, when Volokh reruns the numbers focusing on homicides, he finds essentially no correlation. Also, since Volokh uses an odd variable for the x-axis called the Brady Score (a rating of how restrictive gun control laws are), we should note that his findings are replicated when just using basic gun ownership rates as the independent variable, as shown here.

So how do we square these opposing results? One side finds a significant correlation, the other comes up with nothing. The reason appears to be that correlations found between gun deaths and gun ownership are being driven significantly by gun suicides; indeed suicides actually account for the majority of gun deaths. And while reducing gun suicides may be a worthy goal, we should recognize that this is not the primary purpose of gun control. Further, if that was our goal, it would likely require different measures, such as an outright ban on handguns or very significant restrictions on acquiring one. But so far, that's not what's being considered. Accordingly, it doesn't make sense to include suicides when trying to determine the effectiveness of gun control because that distorts the conclusion. For this reason, Volokh's analysis makes more sense; in the US, gun control doesn't appear to make things significantly better or worse.

 Gun control laws create gun-related crimes, some of which have no victims.

This may seem like a bizarre claim, but it is true. Let's consider illegal possession of a firearm for our purposes. Illegal possession of a firearm may occur because the firearm in question is banned (like assault rifles in some places), the proper permits haven't been acquired (concealed carry permits for example), or the person with the firearm isn't allowed to have one (usually because they are an ex-felon). And frequently, if there's a crime for illegal possession, there was an illegal sale that preceded it.

But all of these circumstances above amount to victimless crimes. After all, the buyer and seller engaged in a voluntary transaction which they presumably found beneficial. And unless and until the buyer commits a crime with the weapon, no one has been harmed. In this way, gun possession crimes are very similar to drug possession crimes.

Gun control laws are likely to have a disproportionate impact on minorities.

Realizing that some gun control crimes are victimless is important because victimless crimes inherently require a different sort of policing. Unlike other offenses, neither of the parties involved in a gun sale are going to report the transaction as a crime to police. This means law enforcement agencies have to try to proactively seek out crimes that are happening, just as the DEA does in the war on drugs. And herein lies the real problem. Proactive policing is usually discriminatory policing; given limited resources, the law enforcement agency has to decide who to target. And all too often, they target minorities.

I'll let Radley Balko at The Washington Post pick up the story here. In this article, he highlights some examples of discriminatory policing in gun control already, and suggests that this needs to be a part of the gun control debate too. It's depressing, but well worth your time to read.

Summing Up

So, where does all of that leave us? Well, for me, it suggests that certain things really could make sense. Following from the first point that most people don't want their neighbor to have a rocket launcher, banning or restricting access to high-magazine rounds is a plausible option that both sides could find common ground on. As long as the ban was isolated to certain weapons or ammunition, you could also diminish the likelihood that you'd sweep up truly peaceful offenders--if somebody actually has a rocket launcher, it seems unlikely it's just for hunting or self-defense.

On the other hand, Balko's analysis strongly indicates that we should be wary of the unintended consequences of legislation that is more sweeping in nature. Gun violence may be a problem, but so is discriminatory policing. And before we advocate for major gun control reform, we need to acknowledge that hidden cost upfront.


  1. Very interesting. I would've assumed you'd have seen some drop in violent deaths simply due to the low effort and low risk to the user a gun affords. If you go in for the stab, you have a decent chance of getting attacked yourself due to proximity and it seems like it would take a lot more conviction to try to take someone out in such a primal, personal way.

    (As a mini-sidebar, folks against our drone program often argue the detached nature of the operations as a reason for why they should be discontinued relative to direct action don't they? This finding, on a much smaller scale of course, really refutes that whole idea.)

    Massacre prevention is a pretty worthy goal on it's own though. You can't shoot up a theater if you don't have a weapon capable of mass killing. And you do have a real chance, especially in a group situation, to overpower someone wielding a knife with minimal injury to the group as a whole, at least relative to a gun.

    In general we have these shootings happen so often anymore that it just seems like something to expect every month or so. It's amazing how much fatigue has set in. Remember Sandy Hook? That was just last year and you had Bob Costas getting yelled at for doing PSA's on Sunday Night Football halftime segments, David Gregory getting shit-canned from Meet the Press for bringing banned guns to the show.

    Now the most recent Oregon massacre? I ran into 1 reddit post about it and that's it. Anecdotal for sure, but just how it seems. It really feels like the NRA has completely won on this issue unfortunately.

  2. Yeah, I would think that too. But I guess if you're committed to attacking someone, method is no obstacle.

    I think people find drone warfare grotesque for this reason, but I don't think that many people make it a main argument. I think people lament it just because it's lowered the cost of war and thus like anything else, we get more of it. The reason to oppose drones, is much more the whole rule of law thing and the fact that it's effectively our version of terrorism. There's a great documentary by Scahill that talks about how kids in droned countries now fear blue skies because that's when drones might launch weapons (since then they can sort of see something).

    See, but it's not even clear that massacre prevention is achieved by gun control. Check the link after the Volokh piece. They found no correlation between gun ownership and mass shootings across states either. Sure, Australia did it, but you literally could not do what Australia did. I mean it's obviously unconstitutional, but more than that, I think you'd have Waco / Branch Davidian situation all over the place if a mandatory buyback was introduced.

    Sad we're becoming numb, I guess, but it also reduces the incentive for people to do it for notoriety, which is a possible upside. Also, we've been broadly immune to people dying for no damn reason for a while, maybe now we're just extending it to some American victims as well.