The Performance of Performance-Based Contracting in Human Services: A Quasi-Experiment

Abstract

Performance-based contracting (PBC) is becoming increasingly attractive to public human service agencies. By linking contract compensation to contractors’ performance, PBC is expected to foster quality services, to improve outcomes, and to reduce government monitoring. However, empirical evidence on the effectiveness of PBC remains both limited and mixed. Based on a case study of the Indiana vocational rehabilitation program, this article employs a quasi-experimental design to evaluate the effectiveness of PBC for individual employment outcomes over the 2004–09 period, with the Michigan vocational rehabilitation program as a control. Using propensity score matching and difference-in-differences regressions to control for the imbalances between the two states, this article finds that PBC significantly improves employment results and reduces time to employment, two measured performance indicators. However, the impact of PBC on unmeasured employment quality, as indicated by working hours and wages, is trivial. This article further suggests relational contracting as a supplemental mechanism for PBC in human service provision.