used to kill an individual in a particular year is 1 in 10,000. The typical gun buy-back program yields less than 1,000 guns. Even ignoring the first two points made above (the guns turned in are unlikely to be used by criminals and may be replaced by purchases of new guns), one would expect a reduction of less than one-tenth of one homicide per year in response to such a gun buy-back program. The program might be cost-effective if those were the correct parameters, but the small scale makes it highly unlikely that its effects would be detected.
In light of the weakness in the theory underlying gun buy-backs, it is not surprising that research evaluations of U.S. efforts have consistently failed to document any link between such programs and reductions in gun violence (Callahan et al., 1994; Police Executive Research Forum, 1996; Rosenfeld, 1996).
Outside the United States there have been a small number of buy-backs of much larger quantities of weapons, in response to high-profile mass murders with firearms. Following a killing of 35 persons in Tasmania in 1996 by a lone gunman, the Australian government prohibited certain categories of long guns and provided funds to buy back all such weapons in private hands (Reuter and Mouzos, 2003). A total of 640,000 weapons were handed in to the government (at an average price of approximately $350), constituting about 20 percent of the estimated stock of weapons. The weapons subject to the buy-back, however, accounted for a modest share of all homicides or violent crimes more generally prior to the buy-back. Unsurprisingly, Reuter and Mouzos (2003) were unable to find evidence of a substantial decline in rates for these crimes. They noted that in the six years following the buy-back, there were no mass murders with firearms and fewer mass murders than in the previous period; these are both weak tests given the small numbers of such incidents annually.
In 1994, Congress enacted the Violent Crime Control and Law Enforcement Act, which banned the importation and manufacture of certain military-style semiautomatic “assault” weapons and ammunition magazines capable of holding more than 10 rounds (National Institute of Justice, 1997). Assault weapons and large-capacity magazines manufactured before the effective date of the ban were grandfathered and thus legal to own and transfer. These guns are believed to be particularly dangerous because they facilitate the rapid firing of high numbers of shots. While assault weapons and large-capacity magazines are used only in a modest fraction of gun crimes, the premise of the ban was that a decrease in their use may reduce gunshot victimization, particularly victimizations involving multiple wounds or multiple victims (Roth and Koper, 1997).