The enduring effects of cohort characteristics on age-specific homicide rates, 1960-1995.

In the past decade, young people in the United States have been two to three times more likely than in the two previous decades to commit homicides, while those 25 years and older have been less likely to commit homicides than were members of their age groups in the earlier time period. These changes in youth homicide rates are associated with two cohort characteristics that are theoretically linked to criminality: relative size of cohorts and the percentage of cohort member born to unwed mothers. These effects persist throughout the life span, are independent of age and historical period, and can explain fluctuations in homicide arrest rates before the recent upturn.