Auditing processes are intended to simply monitor compliance with existing rules. However, they may by themselves sometimes create unintended impacts and incentives through the specific protocol by which audits are executed. This paper investigates the effects of 2011 and 2012 audits in Chile on public entities’ subsequent public procurement practices. While the national procurement legislation tries to promote the use of more transparent and competitive auctions rather than the more discretionary direct contracts for selection of public procurement suppliers, the audit protocol of the national comptroller agency mechanically leads to more scrutiny and a higher probability of further investigation for auctions than for direct contracts. This can create a disincentive for the use of auctions. Using a regression discontinuity design based on a scoring rule of the national auditing agency, we find that indeed being audited leads public entities to decrease their use of auctions by about 35 to 55 percentage points and increase the use of direct contracts correspondingly. In order to further test the underlying mechanism, we develop a new econometric approach to conduct subgroup analysis in regression discontinuity designs while holding other observables constant.