Working Paper: NBER ID: w23978
Authors: Maria Paula Gerardino; Stephan Litschig; Dina Pomeranz
Abstract: Public sector audits are a key element of state capacity. However, we find that they can create unintended distortions. Regression discontinuity analysis from Chile shows that audits lowered the use of auctions for public procurement, reduced supplier competition, and increased the likelihood of incumbent, small, and local firms winning contracts. Looking inside the black box of the audit process reveals that relative to comparable direct contracts, auctions underwent more than twice as many checks and led to twice as many detected infractions. Procurement officers perceive the consequences of such detected infractions as severe. These findings show that standard audit protocols can mechanically discourage the use of more regulated, complex and transparent procedures that involve more auditable steps.
Keywords: Public Procurement; Audits; Regression Discontinuity; Supplier Competition
JEL Codes: D73; H57; O38
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Shift away from transparent auctions (D44) | Reduction of supplier competition (L14) |
| Shift away from transparent auctions (D44) | Contracts awarded to incumbent small and local firms (L14) |
| Audits (M42) | Checks and detected infractions for auctions (D44) |
| Checks and detected infractions for auctions (D44) | Discouragement of the use of auctions (D44) |
| Shift away from transparent auctions (D44) | Price increases (E30) |
| Audits (M42) | Shift away from transparent auctions (D44) |
| Audits (M42) | Increase in direct contracting (L14) |