Using international crime data the committee has attempted to compare per capita homicide rates and rates of firearm homicides in the United States with those in other countries. While we recognize that the measurement of these events is not entirely consistent, these data do provide rough but useful comparisons.

International Comparisons

Table 3-1 displays the data on homicides, firearm-related homicides, and firearm availability for 36 countries. Krug et al. (1998) collected these data by surveying ministries of health or national statistical centers in each of these countries. Review of these data indicate that while the United States does not have the highest rate of homicide or firearm-related homicide, it does have the highest rates for these among industrialized democracies. Homicide rates in the United States are two to four times higher than they are in countries that are economically and politically similar to it. Higher rates are found in developing countries and those with political instability. The same is true for firearm-related homicides, but the differences are even greater. The firearm-related homicide rate in the United States is more like that of Argentina, Mexico, and Northern Ireland than England or Canada. While certainly not the highest homicide or firearm-related homicide rate in the world, these rates in the United States are in the upper quartile in each case.

Some researchers have used data like those summarized above to assess the relationship between firearm-related homicides and firearms availability. For the most part this research focuses on industrialized nations and uses various proxies for the measure of firearms availability. While the vast majority of these studies conclude that homicides and availability are closely associated (Lester, 1990; Killias, 1993a, 1993b; Hemenway and Miller, 2000), the methodological problems in this research (measurement of key variables is of questionable validity, the use of nation-states as the unit of analysis may mask subnational variability, and models tested are poorly specified) do not encourage us to place much weight on this research. However, as noted earlier, the level of nongun homicide is much higher in the United States than it is in other countries. A high level of violence may be a cause of a high level of firearms availability instead of the other way around. Further work with better measures and more complete samples might be useful; for now this literature can be considered suggestive but not conclusive.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement