Disentangling the relationship between serious disorder problems and the use of supermax prisons

Abstract

Research Summary

Supermax prisons exist, according to the policy's driving logic, to address serious disorder problems by isolating people who cause serious disorder. Criticisms of supermax, however, suggest that its use is driven by factors aside from serious disorder problems and that transfers to supermax have a degree of arbitrariness. The current paper uses 10 years of data on one state prison system to disentangle the relationship between “disorder” and supermax utilization by examining (1) the association between a facility's level of serious disorder and supermax transfers (prison-level analysis), (2) the association between a person's level of serious rule breaking and oppositely, “nuisance” behaviors, and the odds of a supermax transfer (individual-level analysis), and (3) whether the influence of a person's behavioral record on the odds of supermax depends upon contextual factors (prison-by-individual interaction). The results suggest that a facility's level of serious disorder does strongly correspond with its reliance on supermax and that indicators of serious rule breaking, and not minor forms of it, are strongly associated with a person's odds of experiencing a supermax transfer. However, a person's misconduct record has a weaker association with the odds of a supermax transfer in facilities that are more disorderly.

Policy Implications

The findings highlight the importance of prison context for dictating who experiences supermax and raises questions about whether prison systems have a uniform “bar” for supermax transfers. This in turn suggests a need for prison systems to reevaluate individual pathways to supermax and consider whether there are consistent metrics used for supermax placements. More broadly, the finding that context shapes who is transferred to supermax raises questions about the fundamental policy logic and suggests a need to investigate other solutions to disorder problems.