Beliefs About Public Spending and Efficient Expenditures in Public Procurement Officers

How to cite

Undurraga, Raimundo; Gertler, Paul; Olivares, Marcelo; Celhay, Pablo. 2020. "Beliefs About Public Spending and Efficient Expenditures in Public Procurement Officers." AEA RCT Registry. AEARCTR-0006449. https://doi.org/10.1257/rct.6449-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Raimundo Undurraga* , Paul Gertler , Marcelo Olivares , Pablo Celhay

* Primary investigator

Registry ID
AEARCTR-0006449
DOI
https://doi.org/10.1257/rct.6449-1.0
First registered
Last updated
Status
On going
Trial dates
2020-01-01 – 2021-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/6449

Abstract

We design a series of information interventions aimed at highlighting the benefits and costs of efficiency in public spending, as well as reducing frictions in the search for low cost purchases, all with the ultimate objective of changing the institutional culture of efficiency within Public Services and thus reduce overspending in their public purchases. We implement different interventions depending on the level of responsibility within each treated service, i.e. whether the person is the Director of the PS (top-level or principal) or the person is a ``buyer'' (bottom-level or purchasing agent). The interventions consist of a combination of monthly reports with the performance of efficiency in public purchases at the buyer and service levels with new budget rules designed to punish next-year budget of services that incur in large overspending in the previous year. Central to our analysis is the role of beliefs about the benefits/costs of efficiency in public purchases, and the extent to which the top-bottom alignment of beliefs matters to increase the level of efficiency in public procurement.

Intervention

(see attached PAP for details)

Experimental design

The experiment is run in two phases. In phase 1 we implement the $\gamma$ treatments only (see attached PAP for details), which goes from June to December 2020. Then, from January 2021 onward (Phase 2), we introduce the $\mu$ treatment.

Out of 184 services, in phase 1 (June - December 2020) we randomly assign 61 services to T1 (``Public"; 2,887 buyers), 62 services to T2 (``Private"; 2,916 buyers), and 61 services to the control group (2,570 buyers), and we stratify the randomization by ministry (higher-level organization) to which the service belongs, and for whether the suffered suffered a cut on the assigned budget or not in the 6 months before June 2020. Then, in phase 2 (January - December 2021), T1 and T2 groups are informed about the the New Budget Evaluation Rule to be imposed by DIPRES, while Control group is not informed at all. Finally, within T1 and T2 groups, 1/4 of buyers are randomly assigned a placebo message instead of the report, so that we can test for Howthorne effects.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
For $\gamma$ and $\mu$ treatments, the unit of randomization is the service. For placebo treatment (Howthorne effects), the unit of randomization is the buyer
Clusters
184 services
Observations
8373 buyers
Arms
61 services T1 (2887 buyers) 62 services T2 (2916 buyers) 61 services C (2570 buyers)
Primary outcomes
- Overspending - Beliefs about Overspending (see the attached PAP for details)
Secondary outcomes
(see attached PAP for details)

Cited by

No citation data available for this registration yet.

Keywords