Running Out of Inmates? Cohort Non-Replacement and the Approaching End of Mass Incarceration

• In Event: Frontiers in Criminal Justice Policy Methods: Discrimination, Incarceration Dynamics, and Consumer Choice

Abstract

For the past decade, most policy debate about mass incarceration has started from the same premise: absent an aggressive, sustained political push for reform, the system will not shrink. The 2010s bipartisan reform consensus collapsed during the COVID-era crime spike, tough-on-crime politics have returned, and states that were once rolling back sentencing enhancements are reinstating them. Most criminologists and policy analysts have concluded that decarceration is stalled, or reversing. And yet the prisons are emptying anyway. Prison admission rates for people aged 18-30 have fallen 53% since 2000. This paper argues that what looks like a stall in reform is masking a quiet, large, and accelerating decline being driven by a force indifferent to politics: cohort non-replacement. Drawing on the Age-Period-Cohort (APC) framework, we show that the crime boom of the 1980s and 1990s swept up a particular birth cohort — late Baby Boomers, Generation X, and older Millennials — who came of age as punitive sentencing peaked and remained entangled with the system long after the crime boom ended. Those cohorts are now aging out. Younger Millennials and Generation Z have entered adulthood with extraordinarily low rates of offending and arrest, and are simply not replacing them. Using data from the National Corrections Reporting Program, we project that prison admission rates will fall an additional 37–58% within two decades, returning to levels last seen in the 1970s. The projected decline for Black men is even steeper: we estimate the Black-White disparity in admission rates will fall from 7.9 in 2000 to roughly 2.0 by 2036. This dramatic decline in Black imprisonment — what we call the "orphan fact" of American criminal justice, because it fits neither conservative nor progressive narratives — deserves far more attention than it has received. The paper closes with a policy and research agenda for a world the field has not yet seriously planned for. Much of what we know about crime, punishment, and collateral consequences was learned from crime-boom-era populations; those findings may not generalize. States currently investing in prison construction based on outdated projections are likely to regret it. Risk assessment tools trained on high-crime cohorts may mismeasure risk in low-crime ones. Evidence-based programs with cost-benefit ratios estimated in the 1990s may no longer pencil out. The institutional machinery built during the buildup, meanwhile, shows clear signs of fiscal hysteresis — spending and capacity that persist long after the problem they were built to address has receded. The end of mass incarceration is visible on the horizon. The question is whether policy and research are ready for it.

Author

• Shawn D Bushway, University at Albany
  Presenting Author