x

Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
Bureaucrats vs. Politicians: A Field Experiment on Political Oversight over Bureaucrats
Last registered on March 07, 2019

Pre-Trial

Trial Information
General Information
Title
Bureaucrats vs. Politicians: A Field Experiment on Political Oversight over Bureaucrats
RCT ID
AEARCTR-0000402
Initial registration date
August 31, 2015
Last updated
March 07, 2019 11:40 PM EST
Location(s)
Region
Primary Investigator
Affiliation
Harvard University
Other Primary Investigator(s)
Additional Trial Information
Status
Completed
Start date
2014-05-22
End date
2016-12-31
Secondary IDs
Abstract
Local bureaucrats are crucial for service delivery. In many developing countries, they operate in settings with weak oversight institutions and high levels of information asymmetries, further contributing to suboptimal levels of public services. One important, yet understudied, mechanism to hold them accountable is political oversight. The effects of political oversight over bureaucrats on the quality of service delivery are ambiguous. On the one hand, a growing body of empirical evidence finds that in particular in settings of weak institutions elected officials are often more accountable than their appointed counterparts, suggesting that increased political oversight may have positive effects. On the other hand, the literature points to several reasons why bureaucratic insulation may be beneficial, including pandering and patronage by politicians. My dissertation grapples with this ambiguity theoretically and empirically. Building on delegation theory, I formulate an original theory on the conditions under which we should expect an increase in political oversight to improve the quality of service delivery. I test the derived hypotheses through a field experiment, which I implement in collaboration with the Ugandan Ministry of Finance and ODI in 260 local governments in Uganda, by increasing the de facto oversight local politicians have over their bureaucratic counterparts, and, in a second treatment arm, their electoral incentives to use it to improve service delivery.
External Link(s)
Registration Citation
Citation
Raffler, Pia. 2019. "Bureaucrats vs. Politicians: A Field Experiment on Political Oversight over Bureaucrats." AEA RCT Registry. March 07. https://doi.org/10.1257/rct.402-7.0.
Former Citation
Raffler, Pia. 2019. "Bureaucrats vs. Politicians: A Field Experiment on Political Oversight over Bureaucrats." AEA RCT Registry. March 07. http://www.socialscienceregistry.org/trials/402/history/42706.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
Interventions
Intervention(s)
The intervention is designed to increase the oversight capacity of local elected representatives over their bureaucratic counterparts. The core of the intervention is the dissemination of highly disaggregate financial information. Due to a reform in 2010, Ugandan local bureaucrats are required to report location and project specific information on local government budget allocations and expenditures to the Ministry of Finance on a quarterly basis. These figures, as well as details on the actual amounts received by local governments, are largely unknown to elected representatives and the general public.
In the first treatment arm, financial information relevant to the specific constituency is provided to elected representatives in non-technical hard copy reports on a quarterly basis. The first dissemination is accompanied by a day-long workshop for local politicians of randomly selected local governments. The curriculum of the workshop is designed to enable participants to interpret the financial information, clarify their oversight mandate, and to increase their monitoring skills. In the second treatment arm, local opinion leaders are invited to participate in the workshop and to receive the financial information together with their elected representatives.
Intervention Start Date
2014-08-04
Intervention End Date
2015-12-15
Primary Outcomes
Primary Outcomes (end points)
* Knowledge about local government budgets and expenditures
* Perceived accountability pressure
* Monitoring effort
* Transfer rates of low/high performing bureaucrats
* Quality of local service provision

For a full list of outcome variables, please see the preanalysis plan.
Primary Outcomes (explanation)
Please see the preanalysis plan.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Local governments (subcounties) in the sampled districts will be randomly assigned to one of three conditions:
1) Local politicians and bureaucrats are invited to attend the training workshop and receive financial information on a quarterly basis.
2) Local politicians, bureaucrats and local opinion leaders are invited to attend the training workshop and receive financial information on a quarterly basis.
3) Status quo.
Experimental Design Details
Subcounties will be randomly assigned to one of three treatment conditions: (1) Information on location-specific budget allocations and reported expenditures will be provided to subcounty councilors through hard copies and a call center. At the beginning of the interventions, councilors will receive a day-long training on how to interpret the information. (2) The same as above, but in addition to councilors also local opinion leaders -- defined as people who are known to speak up on behalf of the community, while not holding any formal public office -- will also be provided with information and training. (3) A comparison group. Randomization will be stratified on characteristics of bureaucrats and politicians as measured during the baseline. I will construct an index of 'quality' consisting of measures of effort, honesty, and public-mindedness. Subcounties will be stratified on two variables: (1) above/below median 'quality' of bureaucrats, (2) above/below median 'quality' of politicians.
Randomization Method
Randomization done in office by a computer
Randomization Unit
Subcounty
Was the treatment clustered?
Yes
Experiment Characteristics
Sample size: planned number of clusters
260
Sample size: planned number of observations
2950 individual respondents
Sample size (or number of clusters) by treatment arms
75 subcounties councilors provided with budget information, 75 subcounties councilors and local opinion leaders provided with budget information, 110 subcounties control.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Assuming rho = 0.1, power = 0.8 0.16 sd for pooled analysis 0.19 sd for heterogeneous analysis Assuming rho = 0.1, power = 0.9 0.18 sd for pooled analysis 0.22 sd for heterogeneous analysis
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Uganda National Council for Science and Technology (NARC)
IRB Approval Date
2014-05-16
IRB Approval Number
ARC 153
IRB Name
Yale University
IRB Approval Date
2014-06-05
IRB Approval Number
1404013737
Analysis Plan

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Yes
Data Collection Completion Date
Final Sample Size: Number of Clusters (Unit of Randomization)
Was attrition correlated with treatment status?
No
Final Sample Size: Total Number of Observations
Final Sample Size (or Number of Clusters) by Treatment Arms
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)
REPORTS & OTHER MATERIALS