Removing Barriers to Integration: Experimental Evidence on Cash and Information Support for Asylum Seekers

This study evaluates two interventions delivered to asylum seekers in Costa Rica through UNHCR:

1. Unconditional Cash Transfers (UCT): Participants receive USD $880 total, distributed over either 3 months (higher monthly amounts) or 6 months (lower monthly amounts). The cash transfers are designed to alleviate immediate financial constraints and enable recipients to meet basic needs.

2. Information Campaign: Participants receive 10 informational flyers via WhatsApp about Costa Rican government social programs and public services available to asylum seekers and refugees. The materials cover: financial assistance programs (IMAS), educational degree validation, professional training courses (INA), medical care and health insurance, identification document renewal, and domestic violence support services. The information includes eligibility requirements and application procedures.

Study Design: The study uses a 2x3 factorial design with 6 treatment arms: (1) control group receiving neither intervention, (2) information only, (3) 3-month UCT only, (4) 6-month UCT only, (5) 3-month UCT plus information, and (6) 6-month UCT plus information. The interventions test whether temporarily removing financial constraints and informational barriers can help asylum seekers transition from humanitarian assistance to state-sponsored social protection while improving overall socioeconomic integration.

2024-09-02

2026-04-30

This study has two types of outcomes: survey outcomes and outcomes from administrative data.
Survey Outcomes:
1. Treatment-Implementation Outcomes
- Cash transfer received, amount, frequency
- Flyers received, number of flyers

2. Psychological Well-being
- Self-reported depression (PHQ-2 index)
- Perceived stress (PSS-4 index)

3. Social Integration
- Social integration index
- Self-reported connection to Costa Rica
- Discriminatory treatment experiences
- Intent to relocate

4. Navigational Integration: Knowledge
- Acquired knowledge index (knowledge of social programs)

5. Navigational Integration: Attitudes
- Trust in Costa Rican government index
- Self-reported navigational integration index
- Trust in UNHCR

6. Institutional Engagement and Access
- Participation in social programs index
- Applications to social programs index
- Integration in public services index
- Direct constraints to accessing social programs index
- Indirect constraints to accessing social programs index
- Dealing with constraints index

7. Economic Integration
- Economic integration index
- Short-term and long-term basic needs indices
- Economic activity index
- Financial inclusion

8. Perceived State Discrimination
- Self-reported discrimination in state interactions index

Administrative Outcomes:

Institutional engagement and access (social program participation, health insurance, education enrollment)
Economic integration (employment outcomes, formal income)

Please refer to the attached analysis plan for further details on outcomes, measurement, and survey instrument.

Please refer to the study analysis plan for detailed measurement specifications and survey question mappings.

This is a randomized controlled trial using a 2x3 factorial design with block randomization. The study has 2,587 participants recruited monthly over seven months (September 2024 to February 2025) from asylum applicants registering with UNHCR offices in Costa Rica.

Randomization: Participants are block-randomized monthly using a stratified approach. Each month, eligible participants are first stratified into three groups based on baseline knowledge and prior use of Costa Rica's government social programs (prior knowledge and use, prior knowledge but no use, no knowledge and no use). Within each stratum, participants are blocked on UNHCR's vulnerability score and into blocks of approximately 16 individuals with the most similar scores. Within each block, participants are randomly assigned to one of six treatment arms.

Treatment Arms:
Control (25%): Neither cash nor information
Information only (25%): Social program information via WhatsApp
3-month UCT only (12.5%): Cash transfers for 3 months
6-month UCT only (12.5%): Cash transfers for 6 months
3-month UCT + Information (12.5%): Cash transfers for 3 months + Social program information via WhatsApp
6-month UCT + Information (12.5%): Cash transfers for 6 months + Social program information via WhatsApp

Data Collection: The study combines telephone surveys (midline 6 months after first delivery of treatment, endline at 12 months post-treatment) with administrative data from Costa Rica's unified social information system (SINIRUBE), Social Security Fund (CCSS), and Ministry of Education.

Sample: Excludes individuals flagged as extremely vulnerable by UNHCR (who receive UCT with certainty). The final sample is 2,587 participants (reduced from planned 3,600 due to budget constraints). See analysis plan and below for implications of sample size reduction.

Please refer to analysis plan for further details.

Not available

Randomization was done using a computer.

Individuals

No

2587

2587

Control: 649
Information only: 661
3-month UCT only: 319
6-month UCT only: 321
3-month UCT + Information: 322
6-month UCT + Information: 315

This study is powered to detect effects on economic outcomes that are in line with those reported in previous studies of cash transfers and refugees. In particular, the study is well-powered to detect effect sizes around 0.11 standard deviations for the main pooled comparisons (e.g., UCT vs no UCT). Please see the attached analysis plan for more details on minimum detectable effects (MDEs) for all comparisons under different scenarios of attrition (for survey outcomes) and changes in effect sizes following budget cuts that left us with a final sample of 2,587.

Removing Barriers to Integration: Experimental Evidence on Cash and Information Support for Asylum Seekers: Pre-analysis Plan