Since 1991 the homicide rate of the United States declined by more than 40%. Such a dramatic change in the crime rate of any country, particularly one of the size of the United States, is highly unusual. Numerous studies have proposed explanations for this event, yet experts agree that the causes of the homicide decline are still a mystery.

Recent comparative research found that many countries worldwide experienced very similar homicide declines as the United States’, suggesting that the homicide decline was actually an international event. This finding has several implications for the study of crime trends. In particular, it shifts the search of causes from domestic policies, to shared international phenomena.

This study tests whether changes in the relative size of countries’ youth populations, an event that is occurring globally, explain international homicide trends since 1960, including the international homicide decline of more recent decades.

While strong theories exist predicting a relationship between age composition and homicide trends, empirical studies on the topic have consistently found a null association between the two variables. This dissertation contextualizes that literature, discussing how its shortcomings may have artificially created a contradiction between the expected and the observed effect of age composition on homicide trends.

To investigate this topic, this study makes use of a novel dataset on international homicides from the United Nations Office on Drugs and Crime, spanning the previous seven decades. These data were combined to other sources to provide evidence that changes in the homicide rates of countries are largely driven by the size of their youth population, and that the international homicide decline has been a consequence of a global process of population aging. Moreover, by showing that the effect of age composition is most visible at the safest countries of the world – in the absence of competing criminogenic forces driving the homicide trend – this study also explains why the most violent countries are failing to accrue the safety benefits of the aging of their populations, and are not themselves experiencing homicide declines.