There is great variation in estimates of the portion of incarcerated individuals who have mental illness. This wide variation is one of the main findings of the analysis by Prins (1) of mental illness rates in U.S. state prisons. Since at least the 1930s, social observers such as Penrose (2) have tracked the prevalence of mental illness in jails and prisons and have attempted to identify underlying clinical and social trends. In 2002, the first author and his colleagues questioned the presumption that the large number of incarcerated people with mental illness was attributable to failures in the assessment and treatment of mental illness and concluded that there is no reliable evidence that directing resources toward psychiatric treatment would have a significant impact on incarceration rates (3). In this commentary, we argue that this conclusion remains true.