Academic journal article Contemporary Economic Policy

The Australian Firearms Buyback and Its Effect on Gun Deaths

Academic journal article Contemporary Economic Policy

The Australian Firearms Buyback and Its Effect on Gun Deaths

Article excerpt

I. INTRODUCTION

Australia's 1996-1997 National Firearms Agreement (NFA), which involved the buy-back and destruction of over 600,000 guns within a few months, is one of the most massive government adjustments to gun control regulations in the developed world in recent history. The extent of the buyback and accompanying swift nationwide change in the firearm regulatory environment following the enactment of the NFA (prohibition on certain types of firearms, registration requirements, etc.) makes this policy a natural experiment in which the results of a methodical evaluation would be extremely interesting. (1) Indeed, this call has been heeded by several quantitative researchers (see Baker and McPhedran, 2007a; Chapman et al., 2006; Neill and Leigh, 2008; Reuter and Mouzos, 2003). Yet, more than a decade later, despite considerable research using homicide statistics from the same source--the Australian Bureau of Statistics (ABS)--there is still considerable rhetorical debate regarding the impact of the NFA on firearm homicide. In a typical time series design, violence rates are analyzed using time series regression methods to see if there is a significant downward shift in homicides around the time a new gun law goes into effect. Much of the debate surrounding the effects of the NFA has revolved around interpretations of the results of quantitative analyses and their accompanying statistical tests. This paper aims to contribute to this debate by a reanaly-sis of the same data used by previous researchers, using an alternative time series approach based on unknown structural breaks. As shown in Piehl et al. (2003), such tests for unknown structural breaks provide a useful framework for estimating potential treatment effects in an evaluation framework. In this paper, by employing a battery of structural break tests and analyzing the available data in a rigorous fashion, an attempt is made to resolve the debate surrounding the effects of the NFA.

A noteworthy feature of these studies that analyze the effects of the NFA is that they are all based on aggregate statistics for the whole of Australia. Unfortunately, the consistency of the approach taken by each state and territory government in Australia in the enactment of uniform firearm legislation has meant that an evaluation can only be conducted for the whole country and not for individual states or counties. In other words, unlike most quasiexperiments analyzed in the evaluation literature, there is no appropriate comparison group as the entire population in Australia was targeted. (2) This makes the use of difference-in-difference or matching estimators that explicitly rely on comparison groups inappropriate. It also makes approaches often used in the U.S. gun control literature that are based on state-level variation in gun laws and which implicitly rely on comparison groups infeasible. (3) Hence, from a policy perspective, although the NFA is a clean example of a natural experiment, from a statistical perspective, evaluation of the reform is not a straightforward exercise.

The main aim of this paper was to analyze the effect of the NFA on reducing firearm related deaths in Australia, with an emphasis on deaths by homicide and suicide. In addition, following the previous literature, we also examine homicides and suicides by methods other than a firearm to allow for the possibility of method substitution effects.

Several features of the Australian gun buy-back make it of interest to policy makers in the U.S. and other developed countries where gun control is an important issue on the political agenda. First, it has been argued that many gun laws fail because they are local, allowing guns from contiguous counties or states with more lenient laws to "leak" into the stricter jurisdictions. As Australia is geographically isolated with no domestic supply of the prohibited firearms, a nationwide implementation of the gun buy back implies that leakage would not be a serious issue. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.