Estimating the Returns to Offers of International Migration and Skills in the United Arab Emirates
Last registered on July 20, 2016

Pre-Trial

Trial Information
General Information
Title
Estimating the Returns to Offers of International Migration and Skills in the United Arab Emirates
RCT ID
AEARCTR-0001294
Initial registration date
July 20, 2016
Last updated
July 20, 2016 9:42 AM EDT
Location(s)
Region
Region
Primary Investigator
Affiliation
Columbia University
Other Primary Investigator(s)
PI Affiliation
NYU
PI Affiliation
Wharton
Additional Trial Information
Status
On going
Start date
2015-05-20
End date
2017-05-01
Secondary IDs
Abstract
The proposed research is a survey of potential workers in India who are seeking employment in the UAE. The UAE
Ministry of Labor is working with a recruitment firm in India to enhance the general skills of workers while they are still
in India as well as randomizing a small share of eligible workers out of job eligibility _at that recruitment center_. The training program is designed to be directly linked to skills in construction jobs in the UAE. The research
involves a survey that will be given to this pool of workers. The survey is based on questions from the NSS survey
conducted by the Indian Government (first conducted in 1950), and on the Yale EGC-CMF Tamil Nadu Panel Survey.
The survey collects basic (voluntary) demographic data, as well as information on current and future expected employment and earnings.

The followup, which is still being designed and will likely be 18 months since recruitment, will ask about subjective well being, assets and debt, employment histories, family/ marriage outcomes. We plan on linking the subjects that actually do migrate to administrative data on earnings and time of migration. We expect many of the subjects to migrate even if they are randomized out, as there are a large number of other recruitment firms on the ground.
External Link(s)
Registration Citation
Citation
Naidu, Suresh, Yaw Nyarko and Shing-Yi Wang. 2016. "Estimating the Returns to Offers of International Migration and Skills in the United Arab Emirates." AEA RCT Registry. July 20. https://doi.org/10.1257/rct.1294-1.0.
Former Citation
Naidu, Suresh et al. 2016. "Estimating the Returns to Offers of International Migration and Skills in the United Arab Emirates." AEA RCT Registry. July 20. https://www.socialscienceregistry.org/trials/1294/history/9483.
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)
Intervention Start Date
2015-05-20
Intervention End Date
2015-09-20
Primary Outcomes
Primary Outcomes (end points)
For migrants that we have administrative data for: Earnings and start dates, as measured in administrative data as well as tenure and renewal rates upon contract expiration. If the firms measure productivity we will look at that also.

For all subjects we will look at current wages and employment, wage and employment histories, location (country), job search behavior including reservation wages, experiences with labor brokers/middlemen, subjective well-being measures, loneliness and locations of close friends, and assets and debts. In the longer-term followup, we also plan on estimating the impact of a job offer on marriage outcomes including dowry and marriage quality, political participation and attitudes towards inequality, and tolerance including for other caste groups and other religious groups,

Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
At each recruitment center, workers are screened for job relevant skills by interviewers employed by the company. All workers are surveyed, but 2/7 workers successfully screened are randomized so as to not receive a job offer from the recruitment center.
Experimental Design Details
The construction companies will over-sample workers by identifying larger numbers of qualified workers than they have permits (visas) to hire. We then have a list of randomize from among those who have applied to be migrant construction those who will meet the target number of workers of the company and thereby go to the UAE. The remainder will be randomly chosen qualified workers not going to the UAE because of visa quota restrictions, but otherwise similar to those who left for the UAE. We expect to have 2800 workers in our pool - 2000 will go to the UAE (which is the maximum the firms can send because of quotas) and 800 who are equally qualified but can not go this time because of the quotas. (This is a minimum guaranteed by current visa agreements with the UAE ministry of labor, but the number may increase.) The workers sent to the UAE will also be randomized into a construction skills training program. We will run followup surveys after 1 and 2 years, and will link the migrants to detailed administrative data on earnings, productivity, remittances, and retention.
Randomization Method
We generated a sequence of numbers, with a random 2/7 designated as treatment, electronically, and the survey enumerator worked with the interviewer to randomize workers in the order they are screened. The randomization was not public.
Randomization Unit
individual workers who have been screened by the recruitment firm.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
2800
Sample size: planned number of observations
2800
Sample size (or number of clusters) by treatment arms
We expect 2000 workers to be selected to get visas at the recruitment centers, with 800 in the control.
We expect 1000 of the selected workers to be trained, with the other 1000 untrained.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We do power calculations for two key outcomes that we expect to change as the result of receiving a job offer in the UAE, income and happiness. Happiness is measured from 1 to 10 with 10 being the happiest. So far in our baseline, happiness has a mean of 5.5 and standard deviation of 2. With a power of 0.8 and 95 percent confidence, a sample of 2800 observations will allow us to detect a change in happiness measure of 0.237, or one-tenth of a standard deviation. Income has a mean of 114,051 rupees (or USD 1721) and standard deviation of 79,876. We can detect a change in income of 9365 Rupees (or 8 percent). One potential threat to the power of our analysis is compliance in the control group in particular. While individuals in the control group do not receive an offer to go to the UAE in the context of our study, they may find other job opportunities with other firms in the UAE or in other developed countries. Unfortunately, we do not have good estimates of the arrival rate for international jobs for this population, but our field work suggests that it is unlikely that they can expect to find another international job placement immediately. Thus, we believe it is important to do a mid-line survey after 6 months; this relatively short-term follow-up maximizes the probability that members in the control group have not yet found another position in a developed country. At the same time, some of these outcomes, such as the negative psychological effects of separation from one’s family may take a longer time, so we want to do a follow-up again at 2 years.
Supporting Documents and Materials

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

Request Information
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Columbia
IRB Approval Date
2015-05-17
IRB Approval Number
IRB-AAAP5057
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers