The Impacts of Political Donations on Public Procurement in Colombia

How to cite

Ladino, Juan Felipe; Gulzar, Saad; Riaño, Juan Felipe. 2024. "The Impacts of Political Donations on Public Procurement in Colombia." AEA RCT Registry. AEARCTR-0013709. https://doi.org/10.1257/rct.13709-1.2

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

Juan Felipe Ladino* , Saad Gulzar , Juan Felipe Riaño

* Primary investigator

Registry ID
AEARCTR-0013709
DOI
https://doi.org/10.1257/rct.13709-1.2
First registered
Last updated
Status
Completed
Trial dates
2024-05-13 – 2024-08-31
Countries
Colombia ()
JEL codes
D72, H57, P00
Registry record
http://www.socialscienceregistry.org/trials/13709

Abstract

We explore the relationship between political donations and the allocation of public procurement contracts across multiple local elections in Colombia. By linking the universe of public contractors and the mandatory report of political donations by individuals and firms, our results demonstrate a long-term distortion in the contract allocation process attributable to political donations. We then introduce a novel experimental approach, to study how new mayors strategically react to different warnings about increased scrutiny of contracts given to their donors.

Intervention

Our partners will send warning letters to mayors in 876 municipalities in Colombia using publicly available data on public procurement and political campaign donations. The municipalities will be randomized in different treatment arms that will vary in the type of warning received. The base warnings will have the intention of reporting to mayors the legislation and regulations stablished in Colombia for hiring political campaign donors through public procurement. Variations in the treatment include adding to the warning an attachment with the list of donors reported by the mayor during campaigns, as well as multiple public procurement information about them. Also, a treatment variation consist of mentioning in the warning the possibility of sending the information to the Inspectors General Office or investigative media outlets.

Experimental design

We randomize across municipalities the type of warning elected mayors will receive. The design consist on 11 treatment arms that vary in the type of warning elected mayors in municipalities receive

1. Control group: do not receive letter with warning

2. Placebo letter: receive a placebo warning

3. Letter + no list of donors: receives warnings
4. Letter + full list of donors: 3 + attached list of donors
5. Letter+ partial list of donors: 3 + attached list of donors excluding randomly a donor

6. Letter + government offices + no list of donors: receives warning and report information will be sent to government office
7. Letter + government offices + full list of donors: 6 + attached list of donors
8. Letter + government offices + partial list of donors: 6 + attached list of donors excluding randomly a donor

9. Letter + journalist + no list of donors: receives warning and report information will be sent to media outlets
10. Letter + journalist + full list of donors: 9 + attached list of donors
11. Letter + journalist + partial list of donors: 9 + attached list of donors excluding randomly a donor

Design details

Randomization method
Randomization was done in Stata
Randomization unit
I. Municipal level randomization: municipalities II. Donor level randomization: donors within municipalities
Clusters
876 municipalities
Observations
8,560 donors (roughly 9-10 donor per municipality on average) per month
Arms
1. Control group: 108 municipalities 2. Placebo group: 109 municipalities 3. Treatment group: 659 municipalities
Primary outcomes
Probability of donors receiving a contract, Value of contracts given to donors
Partners
Name: Open Contracting Partnership Partner_type: ngo Url: https://www.open-contracting.org/es/ Name: Anticorruption Institute Partner_type: ngo Url: https://www.estudiosanticorrupcion.org/

Cited by

No citation data available for this registration yet.

Keywords