Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
The impact of vocational training on labor market outcomes in Mongolia
Last registered on October 26, 2017


Trial Information
General Information
The impact of vocational training on labor market outcomes in Mongolia
Initial registration date
October 25, 2017
Last updated
October 26, 2017 12:57 AM EDT
Primary Investigator
National University of Mongolia
Other Primary Investigator(s)
PI Affiliation
University of New Hampshire, Department of Economics
PI Affiliation
Carleton University, Department of Economics
PI Affiliation
National University of Mongolia, Department of Economics
PI Affiliation
Additional Trial Information
Start date
End date
Secondary IDs
Mongolia is a lower-middle income country characterized by high youth unemployment and a large informal sector. Training programs have been implemented to promote employment and tackle these issues. In this study, we evaluate the impact of a vocational training program (VTP) through a field experiment conducted between 2013 and 2015. This time frame allows us to explore short and medium term impacts. We focus on three outcomes of interest: employment, earnings and job quality. We also want to investigate how the high dropout rate registered in the VTP program is related to the fact that participants ignore the potential gains of the training in terms of wages and employments.
External Link(s)
Registration Citation
Alzua, Maria et al. 2017. "The impact of vocational training on labor market outcomes in Mongolia." AEA RCT Registry. October 26. https://doi.org/10.1257/rct.2461-1.0.
Former Citation
Alzua, Maria et al. 2017. "The impact of vocational training on labor market outcomes in Mongolia." AEA RCT Registry. October 26. http://www.socialscienceregistry.org/trials/2461/history/22715.
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
The primary intervention is a short-term vocational training program (VTP), which was introduced in 2003. The program's primary goal consists of promoting employment by providing vocational skills through training. In 2013, there were 80 vocational skills programs provided by the VTP, which included heavy machinery operating, hairdressing, cooking, and various types of construction work, among others. The target population is unemployed youth between 15 and 30 who want to get vocational training. Participating in the VTP gives them an opportunity to improve their skills and competitiveness in the labor market in the short-term, by increasing their odds of getting a job and thus their income.

The Employment Promotion Service Center (EPSC) of the Ministry of Labor is responsible for the overall design of the program and selection of training institutions. It is financed by the State Employment Promotion Fund and is targeted at the unemployed or those vulnerable to unemployment, youth not enrolled in formal education, and low-income citizens.

The EPSC selects training institutions through a competitive bidding process. Institutions must show their ability to provide adequate training at the time of submitting their bids. In 2013, 75 institutions were selected for VTP. Approximately, 15% of them were non-profit and 85% were for-profit institutions.

The length of the training varies from 20 days to 45 days, depending on the type of class chosen, with a minimum duration of 144 hours. Each training program consists of classroom training and a subsequent internship to provide on-the-job work experience. Classroom teaching should not exceed 30% of the 144 total hours. The second part of the training is split into practical training and internships.

In Ulaanbaatar, the Metropolitan Employment Department (MED) implements the VTP once a year, from April to November. The MED signs a contract with the selected institutions and monitors the courses.

VTP covers the tuition cost of the training with no other benefit included using vouchers, the trainees hand in to the selected training institutions. Tuition fees are approximately US$140 per trainee. Eligibility criteria include household income and employment status.

Applicants apply for the program in their respective Khoroo offices and are assigned the training. Historically, the VTP has shown a very high dropout/no-show rate. To that end, and additionally to the effect of the training itself, our intervention test the potential information pitfalls of the beneficiaries regarding the returns to vocational training. This is called the "information treatment". To that end, we delivered a series of letters to trainees. The letters contained detailed information about labor market outcomes of skilled vs.unskilled workers in Mongolia. The letters were delivered every Friday during the training.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Primary Outcomes (explanation)
We look at several outcomes related to employment (labor force participation, employment status, hours worked, wages, formal vs. informal work, industry)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
We look at several outcomes related to drop-out (no-show, attendance, drop-out dates)
Experimental Design
Experimental Design
We designed our experimental protocol together with MED. We followed the regular registration process of the MED, which consisted of the following stages: First, khoroo officials screened applicant eligibility and sent eligible applicants to the district labor divisions. In particular, the VTP targets people who are unemployed, vulnerable to unemployment, those having difficulty finding a job and school dropouts. After the screening process, eligible applicants are randomized into treatment and control group. Applicants assigned to the treatment group continued the application process and received training funded by the MED.

At the second stage, district officials checked place availability for the chosen vocation and provided the contract form. Depending on both vocations and training institutions, some applicants had to sign an agreement with an employer, while other applicants did not need to. Once the contract was signed by all the parties, applicants went to the MED office to receive the training.

Experimental Design Details
Randomization Method
Randomization was implemented after the first stage of the registration process. Khoroo officials were instructed to register applicants normally. At the moment of registration, they had to ask young applicants -between 15 and 30 years old- to an interview for the baseline survey and tell them they may not be selected for training.

At the end of every workday, we called khoroo officials for the lists of applicants’ names. Based on those lists, we conducted randomization at the individual level on a daily basis. Individuals were then informed if they were selected to be trained or not.
We first set the probability of being assigned to the control group at 1/3 and the probability of being assigned to either treatment at 2/3, replicating usual proportions for the VTP. We used random number generator in STATA for randomization.

We then assigned the second treatment among those who had been already selected and had started their training. This time, random assignment was carried out at a class level, rather than at an individual level, to prevent spill-over effects. The class size of short-term vocational training varies highly across training institutions depending on vocation demand and nature (from 3 to 30 students).

Randomization Unit
We have two treatments. Vocational training program (treatment 1) was randomized at the individual level and information treatment (treatment 2) was randomized at the class level.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
For information treatment we used class level randomization. We had 141 classes.
Sample size: planned number of observations
We have planned to include 2774 young unemployed applicants of the program.
Sample size (or number of clusters) by treatment arms
The original sample size of the baseline survey was set at 2,100 individuals of which we would assign 1,400 to the treatment group and 700 to the control group. The second treatment sample size will depend on the number of classes.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations to determine sample size considered employment as primary outcome. A sample size of 2100 would allow us to detect an increase of 3 percentage points in employment. We adjusted the sample for an estimated dropout of 30%.
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 Name
Metropolitan Employment Department
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Intervention Completion Date
April 15, 2014, 12:00 AM +00:00
Is data collection complete?
Data Collection Completion Date
September 15, 2015, 12:00 AM +00:00
Final Sample Size: Number of Clusters (Unit of Randomization)
We have 141 classes and 1188 individuals.
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
1188 individuals
Final Sample Size (or Number of Clusters) by Treatment Arms
Treatment 1: 414 individuals in control and 774 in treatment Treatment 2: 40 classes in control and 101 classes in treatment
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)