Queueing paper reframes SNAP's Holmes ruling as staffing failure

Nearly half of SNAP application denials in Missouri were procedural — applicants who could not get through the required call-center interview, not people found ineligible — according to the record in Holmes v. Knodell, in which a judge ruled the system failures violated procedural due process. A preprint on arXiv by Andrew Daw, Chloe Pache and Angela Zhou turns that finding into a staffing-math problem.

The authors fit a queueing model to call-center data disclosed in the court documents, with parameters for the behaviors that distinguish benefits lines from a typical help desk: redials and abandonment, which let backlogs feed on themselves. Their claim is that the standard operations-research recipe gets the answer wrong for this regime. "Standard queueing guidance from Erlang-A that does not address endogenous congestion fundamentally understaffs, which could lead to persistent shortfalls in practice," they write.

The paper also tries to separate this kind of failure from the algorithmic-rule failures that benefits delivery is more usually critiqued for. "Emergent arbitrariness arises from interactions between system dynamics and access demand, rather than from an explicit algorithmic rule," the abstract states, which the authors argue makes diagnosis and repair "inherently system-level." They pitch the model as a tool for ex-ante evaluation of access systems and as evidence about whether applicants are "afforded a meaningful opportunity to be served at scale."