Do Guns Make It Easier Or Harder To Commit Crimes? Donohue Versus Lott.

In the wake of a massive, unprecedented social media campaign by a bunch of high school kids, all of a sudden the gun industry finds itself facing a storm of protests over whether or not its products should be made or sold. Well, maybe not all of its products, but certainly the products whose use continues to produce enough multiple killings and injuries to ignite a debate about whether such guns should be around at all.

lott2Behind the argument about owning high-capacity, semi-automatic weapons is another debate which has been going on for nearly 20 years about what I call the social utility of guns, namely, do guns make us more or less safe, or to put it another way, do guns protect us from crime or increase crime?

This debate got started in 1998 with the publication of John Lott’s book, More Guns Less Crime, the title of which says what the book is all about. One of the early reviews of this book was by an academic, John Donohue, who also collaborated with Steven Levitt on a controversial study linking legal abortions to the post-1990’s decline in crime.

Over the years, Lott’s book has become something of a Holy Grail to the gun-rights movement, Donohue’s multliple critiques of this book serving in the same fashion for the gun violence prevention crowd, a.k.a, the GVP.

What I have always found interesting in this debate is the degree to which the criticisms of both Lott and Donohue flow directly from where the critics stand on the issue of guns. I have yet to read a single critique of Lott’s book by anyone who considers themselves to be a proponent of guns. Ditto, I have never found a single critique of Donohue’s work emanating from anyone who supports more controls over guns.  In other words, what we have here is an academic argument in which neither side can find a single, critical word to utter about the work with whose conclusions they agree.

This isn’t an academic debate. Frankly, it’s  nothing more than the same, old, tired and hackneyed argument about guns that has been going on for more than twenty years. It’s not driven by evidence-based work, it’s driven by emotions and advocacy that both sides always make.

What you can download here is a detailed paper I have posted on SSRN.  It is not an attempt to prove that Lott is correct and Donohue isn’t, nor the other way around. It is also not an attempt to come up with yet another statistical model which can be used to provide yet another regression analysis linking crime rates to guns.

Frankly, I have absolutely no interest in proving either Donohue or Lott to be correct. My interest is simply to take this long-standing, academic argument and look at it from the only perspective that really counts, and that perspective happens to be what I know about guns.  Which is something that neither Donohue nor Lott know very much at all.


4 thoughts on “Do Guns Make It Easier Or Harder To Commit Crimes? Donohue Versus Lott.

  1. I’m unpacking boxes from the move to Santa Fe right now and hanging speaker mounts but look forward to reading that paper, Mike. Seems to me a lot of what passes for scholarship in the Great Gun Debate is more along the lines of hypothesis validation (confirmation bias) rather than hypothesis testing which, as my advisor told us on multiple occasions, means that we all have to be our pet hypothesis’ own worst enemy.

    Most of which misses the most important point, i.e., if there were far fewer guns, there would be far fewer bullets flying around. That’s not to take anything away from thoughtful and responsible gun ownership. Its just that with guns ubiquitous, those “wrong hands” will always find a way to be armed courtesy of the ubiquitous nature of those little bangers.

  2. “I have yet to read a single critique of Lott’s book by anyone who considers themselves to be a proponent of guns. ”

    I consider myself a proponent of guns. Mr. Lott;s book is bilge, He should have stuck to economics, in which he has a doctorate. We need more good economists.

    I read his book, and was left underwhelmed. It was not entirely clear what he did, but he appeared to use a multi-variate linear regression. Are the data terms independent? What was his statistical model? Are the results significant? Is he comparing chalk and cheese? Does correlation mean causation?

    Having read the book I felt somewhat cheated.

Leave a Reply