Manuel Lleonart Anguix

Abstract

This paper explains why many organizations deliberately duplicate employee oversight—even for routine, verifiable tasks—and when doing so is efficient. I study internal monitoring when supervisors themselves can be bribed. The firm jointly chooses the architecture of oversight (how many supervisors per worker and how their spans of control overlap) and the incentives needed to keep supervisors honest. Overlap—assigning two independent supervisors to the same worker—makes it harder to hide wrongdoing and, crucially, lowers the wage required to maintain honesty; the resulting wage savings can outweigh the extra hiring cost. The analysis shows that redundancy is optimal when baseline verifiability is low or detection decays steeply with span; improvements in monitoring technology or stronger external enforcement shift the optimum toward wider spans with less overlap. The model predicts sorting: weak-verifiability environments sustain narrower spans and more redundancy, while strong-verifiability environments sustain wider spans and greater independence. By linking organizational structure to the credibility of enforcement, the paper reframes overlapping oversight not as waste but as a rational governance choice.