Pessimistic Labor Market Expectations and Educational Investment: Evidence from a Migration Shock

Last registered on August 27, 2026

View Trial History

Pre-Trial

Trial Information

General Information

Title

Pessimistic Labor Market Expectations and Educational Investment: Evidence from a Migration Shock

RCT ID

AEARCTR-0018743

Initial registration date

August 25, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

August 27, 2026, 12:33 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

Colombia

Region

Primary Investigator

Name

Alejandra Quintana-Barrera

Affiliation

Columbia University, Teachers College

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

Completed

Start date

2026-05-26

End date

2026-06-10

Keywords

Education, Labor

Additional Keywords

Beliefs formation, Migration, Returns to Education, Survey Experiment

JEL code(s)

J24; I26; F22; D83

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Large and unexpected migration shocks can shape educational investment decisions not only through labor market competition, but also by altering perceptions about the returns to higher education. This project combines two complementary components linking belief formation to realized educational choices.

First, quasi-experimental evidence from Colombia shows that exposure to the large inflow of Venezuelan migrants between 2016 and 2018 reduced native students’ enrollment in bachelor’s degree programs. These effects cannot be explained by changes in academic performance, native student mobility, or increased competition for university slots. Instead, the declines are concentrated in fields where migrants face fewer labor market barriers, suggesting that perceived and/or real labor market conditions may shape students’ educational decisions.

Building on this evidence, the second component directly examines the process of belief formation. I will implement a large-scale survey experiment in Colombia with recent high school graduates to test whether making migration salient affects perceived returns to higher education and and whether attending university is worth it. Respondents are randomly assigned to one of two conditions that vary the order of two survey blocks: (i) a migration-information block presenting factual statements about Venezuelan migration, and (ii) a block eliciting beliefs about the returns to education and perceptions of whether attending university is worth it. In the treatment condition, respondents receive the migration-information block before reporting their beliefs and perceptions; in the control condition, the order is reversed. All remaining survey content is identical across conditions.

This design is motivated by pilot evidence showing that individuals randomly exposed to migration-related prompts report expected returns to a bachelor’s degree that are approximately 15 percent lower relative to the control group. The experiment tests whether migration salience causally shapes belief formation regarding the economic value of higher education and, in turn, intended educational investment decisions.

External Link(s)

Registration Citation

Citation

Quintana-Barrera, Alejandra. 2026. "Pessimistic Labor Market Expectations and Educational Investment: Evidence from a Migration Shock." AEA RCT Registry. August 27. https://doi.org/10.1257/rct.18743-1.0

Sponsors & Partners

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

Request Information

Experimental Details

Interventions

Intervention(s)

The intervention consists of a light-touch salience and information treatment. Respondents are exposed to information about Venezuelan migration to Colombia before reporting their beliefs about the returns to higher education and perceptions of whether attending university is worth it. Specifically, respondents receive a brief module presenting factual statements about the scale of Venezuelan migration and migrants’ labor market participation, followed by a short comprehension check. They are then asked about their expectations regarding wages and labor market opportunities associated with obtaining a bachelor’s degree, as their assessment of whether attending university is worth it, elicited through a peer-advice framing to reduce social desirability bias.

Intervention (Hidden)

Intervention Start Date

2026-05-26

Intervention End Date

2026-06-10

Primary Outcomes

Primary Outcomes (end points)

Assessment of whether attending university is worth it, elicited through a peer-advice framing to reduce social desirability bias; and expected returns and labor market opportunities associated with obtaining a bachelor’s degree.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

The experiment consists of a randomized survey module embedded at the end of a larger follow-up survey designed to measure post-program trajectories and postsecondary educational expectations. Respondents are randomly assigned to one of two survey sequences that vary the timing of exposure to migration-related information relative to questions about educational expectations and intended investments.

All respondents complete the same two core blocks. The first is a migration-information block that presents factual statements about the scale of Venezuelan migration to Colombia and migrants’ labor market participation, followed by a brief comprehension check. The second is a returns-to-education block that elicits beliefs about wages, labor market opportunities, perceived social mobility associated with obtaining a bachelor’s degree, and postsecondary enrollment intentions.

In the treatment condition, respondents complete the migration-information block before answering questions about educational expectations and intended investments, making migration salient at the moment beliefs are formed and reported. In the control condition, respondents complete the returns-to-education block before encountering any migration-related content. Because respondents cannot revise earlier answers, differences across conditions identify the causal effect of migration salience on perceived returns to higher education and intended postsecondary enrollment decisions.

Experimental Design Details

Randomization Method

The randomization procedure is implemented in a computer program and is replicable.

Randomization Unit

3271 graduates students where cluster randomize in 4 municipalities, Barranquilla, Bogotá, Pensilvaina and Tolima.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

4 municipalities Barranquilla, Bogotá, Pensilvaina and Tolima

Sample size: planned number of observations

The survey is administered to the full sample of 3,271 graduating students across four municipalities, Barranquilla, Bogotá, Pensilvania, and Tolima, which constitute the study clusters. However, the effective sample used in the analysis will depend on survey response rates.

Sample size (or number of clusters) by treatment arms

Each treatment arm will have 50% of the sample 1635 in treatment and 1636 in the control

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

The pilot survey experiment provides a useful benchmark for determining the minimum sample size required under individual-level randomization with 80% power. Detecting an effect equal to half the size of the pilot estimate, a conservative and plausible target, would require approximately 1,662 respondents, or about 831 per treatment arm. Additional benchmarks come from the Alesina et al. (2022) survey experiment on question order. Detecting an effect comparable to their social tax outcome would require roughly 2,222 respondents, while detecting an effect similar to their donation outcome would require approximately 3,288 respondents. By contrast, the social budget outcome corresponds to a much smaller effect size and would require nearly 5000 respondents, making it substantially less feasible in practice. Taken together, these calculations suggest that a sample of approximately 3,000 to 4,000 respondents would provide sufficient statistical power to detect effects within the range reported in related studies.

Supporting Documents and Materials

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

Request Information

IRB