Association for Public Policy Analysis & Management

After thirty years of increase, crime dropped rapidly in the 1990s and continues to trend downward. We have no good explanation for why. Current criminal justice policies, which attempt to change incentives and opportunities for current offenders, account for some of the decline but not most of it.

An alternative explanation centers on birth cohorts: Recent generations (particularly those born after 1980) may have been successively less criminally inclined than their elders. This is hard to prove because generational or “cohort” effects are intertwined with age and period effects. There is no commonly accepted general solution to the age-period-cohort problem.

I developed a specific solution, tailored to the crime problem. Economic and social factors commonly associated with crime rates can be used to identify trends in U.S. period effects from 1980-2015. This allows identification of age and cohort trends. (An alternative identification strategy, based on comparative age distributions among OECD countries, produced almost identical results.) Criminal activity appeared to decline steadily between the 1916 and 1945 birth cohorts. It increased among Baby Boomers and Gen X, then dropped rapidly among Millennials, born after 1985. Period effects were mostly responsible for the late 1980s crack boom and the 1990s crime drop, but age and cohort effects were primarily responsible for crime reductions during the early 1980s and after 2000. Overall, cohort effects had a larger effect on crime rates than period effects.

The importance of cohort effects suggests that policies aimed at reducing delinquency among young children may be more effective in the long run than the current focus on incapacitation, deterrence, and opportunity reduction.