Services Timeline Revision RCT

How to cite

Barnwal, Prabhat; Dar, Aaditya; Mahajan, Aprajit. 2022. "Services Timeline Revision RCT." AEA RCT Registry. AEARCTR-0009621. https://doi.org/10.1257/rct.9621-1.0

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

Prabhat Barnwal* , Aaditya Dar , Aprajit Mahajan

* Primary investigator

Registry ID
AEARCTR-0009621
DOI
https://doi.org/10.1257/rct.9621-1.0
First registered
Last updated
Status
In development
Trial dates
2021-01-01 – 2023-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/9621

Abstract

Service provision by governments is often delayed in developing countries. On one hand, the government's capacity is constrained by lack of resources. On the other hand, it is often argued that government officers do not perform optimally with given resources. To address these issues, the state government in India. The existing Act mandates bureaucratic officials to deliver the services within a specified duration, a violation of which can incur a penalty. Using a randomized controlled trial, we evaluate an intervention where the time provided by the pre-existing Act to government officials to deliver citizen services is shortened. We measure officials' performance using administrative data and citizen surveys.

Intervention

The government provides many services to citizens, from which we shortlisted high-volume services for our study. Such services include Residence Certificates, Income Certificates, Caste Certificates, Correction in Birth Certificates, etc.
In this intervention, the main treatment is to reduce the mandated timelines for government officials to deliver the services to citizens. Citizen applications in the control group services will continue to operate with the status quo deadlines.

Experimental design

Selection of services: We began by selecting top 80 services (by volume of application issued in the time period January-December 2021) as processed in the government’s e-services portal. After deliberations with the government, we then dropped services that were not exclusively processed on the said portal (some could be offline as well). Further we dropped services that had mandated deadlines of less than 5 days. From amongst the new list, we selected the top 36 services (by volume of applications issued in the time period January-December 2021).
Randomization: Our final sample consists of 36 services that are delivered by 6 different government departments. We randomize these services into treatment and control stratified by departments.
Intervention: For a given service in the treatment group, we select the number of days it takes for 75% (or 60%, depending on which one is larger and is less than the designated deadline) of applications to get processed as their official deadline.

Design details

Randomization method
Randomization is done in office using the statistical software package Stata
Randomization unit
Citizen Services
Clusters
36 services
Observations
~650,000 applications (3 month estimate based on monthly average volume of application processed in January-December 2021)
Arms
18 services control, 18 services treatment
Primary outcomes
Process Days: the number of business days taken to process citizen applications (administrative data).
Secondary outcomes
Citizen Satisfaction: We will measure citizen satisfaction using primary surveys.

Cited by

No citation data available for this registration yet.

Keywords