Unilateral Facilitation Does Not Raise International Labor Migration from the Philippines

How to cite

Beam, Emily; McKenzie, David; Yang, Dean. 2017. "Unilateral Facilitation Does Not Raise International Labor Migration from the Philippines." AEA RCT Registry. AEARCTR-0001888. https://doi.org/10.1257/rct.1888-1.0

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

Emily Beam* , David McKenzie , Dean Yang

* Primary investigator

Registry ID
AEARCTR-0001888
DOI
https://doi.org/10.1257/rct.1888-1.0
First registered
Last updated
Status
Completed
Trial dates
2010-03-01 – 2013-05-31
Countries
Philippines (Sorsogon province)
JEL codes
O15, F22, C93
Registry record
http://www.socialscienceregistry.org/trials/1888

Abstract

Significant income gains from migrating from poorer to richer countries have motivated unilateral (source-country) policies facilitating labor emigration. However, their effectiveness is unknown. We conducted a large-scale randomized experiment in the Philippines testing the impact of unilaterally facilitating international labor migration. Our most intensive treatment doubled the rate of job offers but had no identifiable effect on international labor migration. Even the highest overseas job-search rate we induced (22%) falls far short of the share initially expressing interest in migrating (34%). We conclude that unilateral migration facilitation will at most induce a trickle, not a flood, of additional emigration.

Intervention

Gains from international migration can be large for both the migrant, in the form of higher wages, and for the home country, in the form of remittances. However, there is less international migration than might be expected given the large gains possible. The PIs conducted a randomized controlled trial to test whether a set of unilateral facilitation policies increased international migration for work from Sorsogon Province in the Philippines. The Philippines is one of the largest sources of overseas migrant labor; however, the Sorsogon Province has low levels of migration relative to the country as a whole, despite having a comparably qualified population.

The intervention consisted of five treatments applied individually or in combination. The first treatment, "Application Information", consisted of information on typical overseas costs, steps to apply for work abroad, an advertisement for a jobs website, and warning signs of illegal recruitment. The second, "Financial Information", listed typical placement fees and companies which provided loans to cover them. The third, "Website Assistance", consisted of a paper application for an overseas jobs website, built specifically for this project. The fourth, "Passport Information", consisted of a flier on the importance of having, and the steps required to obtain, a passport. The fifth, "Passport Assistance", consisted of a letter inviting recipients to apply for a program which fully subsidized the costs of acquiring a passport and provided assistance during the application process.

In total, the treatments were administered together or separately to create fourteen different treatment groups and a control group. The PIs focused their analysis in particular on five treatment groups: "All Information" (Application, Financial, and Passport Information treatments), "All Information + Website", "Only Passport" (Passport Information and Assistance treatments), "All Information + Passport" (Application, Financial, Passport Information, and Passport Assistance treatments), and "Full Assistance" (all treatments). Overall, the PIs found that informational constraints are not a significant barrier to international labor migration, and that job matching assistance through a jobs website increases job-search effort and the likelihood of obtaining a job interview. However, not even the full treatment statistically increased migration itself.

Experimental design

To obtain the sample population, the PIs randomly selected 42 barangays from 6 municipalities in Sorsogan Province. A list of households from each barangay was collected and randomly sorted. Twenty-six percent of the households from each barangay were selected for the project by moving down the randomly sorted list. If a household could not be found, project interviewers moved on to the next household on the list. At each household, interviewers screened the first member they met who was aged 20-40 and had never worked overseas.

After completion of a baseline survey, respondents were randomized into four treatment groups and a control group, stratified by barangay and microfinance client status. Only the Application Information, Financial Information, and Website Assistance treatments were administered at the beginning of the intervention in 2010. In 2011, a subsample was randomly selected to receive the two passport treatments, stratified by baseline treatment, age, barangay, and whether they had enrolled in the jobs website. An endline survey was collected in 2012 by visiting the sample households again. If the members of the household could not be located, a "log survey" was collected by asking neighbors if the participants had migrated overseas for work at any point.

The PIs used an OLS specification. It included an indicator variable for treatment group, control variables for baseline characteristics, and stratification fixed-effects variables.

Design details

Randomization method
Randomization performed by computer; 42 barangays were randomly selected, then a list of households from each barangay was randomly sorted
Randomization unit
Baseline: Household, stratified by barangay and microfinance client status Passport treatments: Household, stratified by barangay, age, baseline treatment, and jobs website signup
Clusters
42 barangays
Observations
4,153 respondents in baseline survey
Arms
Control group: 645 respondents; All Information treatment: 107 respondents; All information + website treatment: 280 respondents; Passport information and assistance treatment: 103 respondents; All information + Passport assistance treatment: 104 respondents; Full assistance treatment: 275 respondents
Primary outcomes
Searched for an overseas job, Received an invitation to interview for a job, Attended a job interview, Received a job offer abroad, Migrated abroad

Papers from this trial

  1. Emily A. Beam, David McKenzie, and Dean Yang, "Unilateral Facilitation Does Not Raise International Labor Migration from the Philippines," Economic Development and Cultural Change 64, no. 2 (January 2016): 323-368. http://www.journals.uchicago.edu/doi/10.1086/683999

Cited by

No citation data available for this registration yet.

Keywords