Title: Characteristics and Determinants of Global Homicide Crime Waves, 1946 to 1998
Name: Gary LaFree
University of Maryland
Department of Criminology and Criminal Justice/Democracy Collaborative
Year: 1999
Type: Research Grant

Difficulties in comparing political and legal systems have long hampered our efforts to estimate cross-national violent crime rates. Adding the requirement that such studies examine trends over time makes the task even more complex. In this research project, we used World Health Organization records to collect the most comprehensive set of annual homicide victimization data that has ever been assembled. In the main part of our analysis, it was possible to examine annual data for some forty nations for a fifty-year time period. This unique data base has provided answers to a variety of questions about trends in homicide across nations.

First, we were able to use these data to determine how common violent crime "booms" and "busts" have been since World War II. We defined crime booms as crime rates that increase rapidly and exhibit a positive sustained change in direction. Based on our definition, we found that 12 of the nations in our sample of 34 nations (or 35.5%) had experienced crime booms since World War II. These findings contradict both those who argue that booms in violent crime have been rare or nonexistent in the postwar period and also those that argue that violent crime booms have been nearly universal. In support of modernization perspectives, we also found that 70% of the industrializing nations in our sample qualified as having crime booms, but less than 21% of industrialized nations did.

Second, we were also able to use the data to examine whether violent crime rates for the nations of the world are getting more similar or less similar over time. Some researchers have argued that crime rates for all nations of the world will gradually converge as individual nations go through similar globalization processes while others contend that instead there will be growing divergence between crime rates in industrializing nations and the industrial elite but convergence in crime rates within both groups. We use econometric methods to test for convergence and divergence in homicide victimization rates for 35 nations from 1956 to 1998. Our results provide little support for the argument that there has been a growing rift in homicide rates between industrializing and highly industrialized nations. However, we also find that convergence in national homicide rates is far from universal. In fact, nearly all of the cases of homicide convergence we could identify took place among nations of the European Union. Thus, it appears that modern examples of crime convergence are limited mostly to nations that are geographically and culturally proximate. A paper reporting these results is currently under review.

And finally, we are in the process of using the global homicide data base to examine a series of related questions: Is there evidence that male and female rates of homicide victimization are getting more similar over time? How has the global transition to democracy affected homicide victimization rates? And do nations with higher levels of social capital experience lower rates of homicide victimization?

Bibliography: 2002. Declining Violent Crime Rates in the 1990s: Predicting Crime Booms and Busts. Criminology 25:145-168.