Voluntary Audits: Experimental Evidence on a New Form of Bureaucratic Monitoring

• In Event: Bureaucrats and Identity Bias

Abstract

Bureaucratic under-performance is a major challenge in many developing countries. While state audit institutions are advocated as a solution to waste and corruption, they have a mixed record of success. Traditional audits may fail to elicit greater bureaucratic effort or may displace corruption as officials shift spending away from audited expenses. Is it possible to motivate bureaucrats to be honest and exert effort in their work without relying on punitive auditing systems? We hypothesize that voluntary audits that provide bureaucrats with professional and social recognition might be an effective alternative to traditional audit systems. If street-level bureaucrats care about recognition for a job well done, as well as their reputations among colleagues and the broader community, voluntary audits may be an effective tool for eliciting effort and honest performance.


We test this hypothesis in the context of the performance of school principals in their administration of a critical school free lunch program in a poor province in northeast Argentina. We partner with the Tribunal de Cuentas in Chaco, Argentina, to test whether a new system of voluntary audits improves the performance of school principals in their administration of the provincial free meal program. Our field experiment intervenes in the oversight of a highly decentralized school meal program, in which principals are responsible for many aspects of school-level implementation (including menu planning and food preparation). We randomly assign principals to receive an invitation to a voluntary audit or place them in the control group that is subject to a small possibility of status quo mandatory audits. The invitation emphasizes the pro-social nature of the program, and those who accept the invitation receive an in-person audit administered by the provincial auditing body in the coming months. That visit is accompanied by a letter documenting their participation and decision to volunteer. We expect that principals who receive the voluntary audit invitation will exert more effort in their administration of the program, as measured through the number of days meals are served and student attendance. We conduct an endline survey of school principals, qualitative interviews, and analyze administrative data in order to assess the effects of voluntary audits and examine potential mechanisms driving audits’ effects.

Authors

• Ana Lorena De La O Torres, Yale University

• Lucas I. Gonzalez, CONICET/UCA-UNSAM

• Rebecca Weitz-Shapiro, Brown University