Comparing the Impacts of Industrial Jobs and Self-Employment in Ethiopia

How to cite

Blattman, Christopher; Stefan Dercon Oxford University, Centre for the Study of African Economies, Department of Economics and Blavatnik School of Government. 2016. "Comparing the Impacts of Industrial Jobs and Self-Employment in Ethiopia." AEA RCT Registry. AEARCTR-0001642. https://doi.org/10.1257/rct.1642-1.0

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

Christopher Blattman* , Stefan Dercon Oxford University, Centre for the Study of African Economies, Department of Economics and Blavatnik School of Government

* Primary investigator

Registry ID
AEARCTR-0001642
DOI
https://doi.org/10.1257/rct.1642-1.0
First registered
Last updated
Status
Completed
Trial dates
2010-01-01 – 2013-12-31
Countries
Ethiopia (Africa)
JEL codes
J24, O14, F16, J81, O17
Registry record
http://www.socialscienceregistry.org/trials/1642

Abstract

As low-income countries industrialize, workers choose between informal self-employment and low-skill manufacturing. What do workers trade off, and what are the long run impacts of this occupational choice? Self-employment is thought to be volatile and risky, but to provide autonomy and flexibility. Industrial firms are criticized for poor wages and working conditions, but they could offer steady hours among other advantages. We worked with five Ethiopian industrial firms to randomize entry-level applicants to one of three treatment arms: an industrial job offer; a control group; or an "entrepreneurship" program of $300 plus business training. We followed the sample over a year. Industrial jobs offered more hours than the control group’s informal opportunities, but had little impact on incomes due to lower wages. Most applicants quit the sector quickly, finding industrial jobs unpleasant and risky. Indeed, serious health problems rose one percentage point for every month of industrial work. Applicants seem to understand the risks, but took the industrial work temporarily while searching for better work. Meanwhile, the entrepreneurship program stimulated self-employment, raised earnings by 33%, provided steady work hours, and halved the likelihood of taking an industrial job in future. Overall, when the barriers to self-employment were relieved, applicants appear to have preferred entrepreneurial to industrial labor.

Intervention

Low-skilled workers applying to entry-level jobs at one of five firms in Ethiopia were randomly assigned to one of the following interventions:

1) Job offer treatment: the worker receives a job offer
2) Control group: the worker does not receive a job offer
3) Entrepreneurship program treatment: The worker was offered five days of business training and planning followed by $300 grant (about $1030 in purchasing power parity, or PPP). This program was intended to relieve key constraints on their informal employment opportunities.

Experimental design

We follow a panel of nearly 1000 applicants to entry-level jobs in five industrial firms in different parts of Ethiopia. Jobs were advertised locally using flyers. Applicants were self-selected to have at least some interest in an industrial job, and screened by firms to have some secondary schooling. We randomly assigned roughly a third of the applicants to a job offer, a third to a control group, and a third to an entrepreneurship program intended to relieve key constraints on their informal employment opportunities: five days of business training and planning followed by $300 grant (about $1030 in purchasing power parity, or PPP). Five firms participated in the experiment: a water bottling plant, a vegetable farm, a flower farm, a shoe manufacturer, and a textile and garment factory.

After treatment assignment, the firms posted the names of people receiving the job offer at the factory site and the research team contacted all those assigned to the job or grant. Job offers began within a few days. We gave each firm a list of unsuccessful applicants and asked the firm not to hire them for at least 1–2 months.

Design details

Randomization method
Randomization to treatment group was done by computer and was stratified by gender.
Randomization unit
low-skilled worker
Clusters
No clusters
Observations
947 low-skilled job applicants
Arms
304 job offer treatment, 285 entrepreneurship program treatment, 358 control
Primary outcomes
employment in formal sector/industrial labor, employment in informal sector, earnings, hours worked per week, physical health, mental health, employee turnover

Papers from this trial

  1. Blattman, Christopher, and Stefan Dercon. "Occupational choice in early industrializing societies: Experimental evidence on the income and health effects of industrial and entrepreneurial work." NBER Working Paper, September 2016. http://www.nber.org/papers/w22683

Cited by

No citation data available for this registration yet.

Keywords