What we’ve been reading on incentives in public procurement

This page in:
Woman and man in construction hats standing face to face, looking and pointing at construction in the background Woman and man in construction hats standing face to face, looking and pointing at construction in the background

This blog is part of a series of three pieces on public procurement, with a common theme on using empirical methods and experiments in public procurement research.

 

Public procurement is often plagued with cases of corruption, collusion, and other uncompetitive practices that cause misallocation of public funds , and the OECD reports that 10-30% of the value of (publicly-funded) construction projects may be lost through corruption and mismanagement.  Corruption risks arise from conflict of interest in decision-making among different actors (e.g. procurement officers, firms, auditors) involved in public procurement and are shaped by contexts and customs, regulations and procedures which impact the costs and benefits of engaging in corruption.

Building on our previous literature review on public procurement, in this blog we discuss some empirical evidence on  the incentives associated with: (i) contract value thresholds, (ii) audit processes; (iii) effectiveness of courts, and (iv) vendor rating systems.


The introduction of contract value thresholds may generate distortions in decision-making.

While contract value thresholds are often used to balance the need to incentivize more competition for higher value contracts and guarantee more flexibility for smaller contracts, they also create incentives for the manipulation of contract values. For example, Palguta & Pertold (2017)  demonstrate that in the Czech Republic, the introduction in 2016 of contract thresholds under which procuring entities could preselect potential contractors, led to an increase in bunching of procurement contracts right below these legislative thresholds. These manipulations of contract values may mask malpractices, and indeed in the case of the Czech Republic, contracts just below the thresholds are more likely to be awarded to firms with anonymous ownership structure (see Figure), i.e. firms whose legal form allows for hiding rent-seeking behavior.

 

Line graph showing anonymous contractors in construction works

Source: Palguta & Pertold (2017).

 

The audit process of public procurement can create unintended side-effects.

If properly designed, audits can be an effective tool to dis-incentivize direct contracting and promote open auctions. However, as demonstrated by Gerardino et al. (2017), the design and targeting of audits can distort incentives of procurement officers. The authors investigate the impact of the audit selection process in Chile, using public procurement data during 2011-2012. Under the existing audit protocol in that period, open auctions underwent more than twice as many checks as direct contracting. The authors find that, given this protocol, procurement officers shift towards direct contracting methods and reduced the use of open auctions, especially in procuring entities that experienced more audits and therefore had more opportunities to learn about this targeting design. While these results do not indicate that audits have negative consequences per se, they demonstrate that the design and targeting of audits can be important when it comes to incentivizing the choice of procurement methods.


Public procurement is closely interconnected with the effectiveness and efficiency of the judicial system.

The quality and duration of judicial processes impacts the probability that procurement irregularities will be detected, investigated, and prosecuted and it has implications on the timing of procurement and contract execution. Coviello et al. (2017) use a dataset on public works collected by the Italian Public Procurement Authority for the period 2000-2006 to empirically investigate the effects of court inefficiency on public work performance. It is expected that, where courts are inefficient, procuring entities tend to shy away from the litigation process of enforcing a penalty on late deliveries and indeed, the authors find that inefficient courts are associated with longer delays on the delivery of public works, and a higher share of payments postponed after delivery.


The use of past performance indicators may incentivize firms to perform well during contract execution without hindering entry of new firms in procurement markets.

Reputation mechanisms play a critical role in generating cooperative and socially optimal equilibria in settings with repeated interactions between actors. As demonstrated by Decarolis et al. (2016), these mechanisms could also be used in public procurement. The authors use a unique experiment run by a large utility company in Italy which introduced a vendor rating system scoring its suppliers’ past performance and linking it to the award of future contracts. They find that the introduction of this rating system improved average performance from 25% to 90% of the audited parameters. Despite potential concerns that reputation mechanisms may favor existing suppliers and limit competition, the study finds that contract prices did not significantly change prices overall, improvements in performance were long-lasting and resulted in higher service quality by the utility.


A key takeaway from this research is that procurement regulations, practices and management can incentivize the actors involved in the procurement process. Procurement regulations and practices may distort incentives and have unintended adverse effects on procurement outcomes, if not accurately designed, piloted, and evaluated . Studying the incentives of procurement officers, firms and organizations can be helpful during the design stage to examine the trade-offs of employing specific practices or regulations and to tailor “good practice” reforms to specific country contexts. Impact evaluations can be helpful during the piloting stage of new regulations or policies to assess, before full-scale implementation, whether they are generating the expected results and to correct course of action if necessary.


Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000