Spatial Frictions in the Garment Labor Market in Bangladesh

How to cite

Cajal, Julia; Kreindler, Gabriel. 2022. "Spatial Frictions in the Garment Labor Market in Bangladesh." AEA RCT Registry. AEARCTR-0009171. https://doi.org/10.1257/rct.9171-1.0

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

Julia Cajal* , Gabriel Kreindler

* Primary investigator

Registry ID
AEARCTR-0009171
DOI
https://doi.org/10.1257/rct.9171-1.0
First registered
Last updated
Status
In development
Trial dates
2022-07-01 – 2022-12-31
Countries
Bangladesh ()
Registry record
http://www.socialscienceregistry.org/trials/9171

Abstract

Garment workers in Dhaka have hyper-local commutes, mostly on foot. Is this a positive sign that Dhaka’s high urban density delivers sufficient attractive job opportunities? Or does this happen due to spatial frictions, either because long travel times and expensive transport make commuting further away difficult, or because information flows about job openings and workplace conditions decay quickly with distance? We run an experiment providing workers and job seekers information on hiring at high-wage factories, and study measures of job application and interest.

Intervention

We will implement an information-sharing experiment embedded in an existing survey. Using weekly waves of data collection, the intervention entails four stages. In stage one, we collect baseline outcome data and obtain consent for this intervention. In stage two, we ask respondents whether their factory (or another factory that they know about) plans to hire in the coming weeks. In stage three (one week later), we will share information on which factories are hiring with other respondents who live near or far from the factory in question. We also share other information about the factory (such as the entry level wage and other non-wage job characteristics) that was previously collected. In stage four, we repeat the survey of stage one, to collect outcomes on job applications and other measures of interest. The intervention cycle is repeated between 2 and 6 times. We will randomly select a group of respondents who will receive information about hiring at (at most) two nearby factories.

Experimental design

Workers are assigned to treatment on the basis of the neighbourhood (cluster) they live and work in. The factories assigned for the treatment and control are randomly selected within hypothetical choice sets satisfying the criteria of the design (distance, hiring status, vacancy attributes).

Design details

Randomization method
Computer based
Randomization unit
At most 150 clusters
Clusters
At most 150
Observations
Approximately 1,300 workers x number of information rounds (2 to 6)
Arms
We expect 50% of the clusters will be treated (unique arm)
Primary outcomes
Beliefs and information; measures of interest and job applications

Cited by

No citation data available for this registration yet.

Keywords