The effects of cash transfers, subsidized childcare, and life skills training on adolescent mothers' participation in school or vocational training in Malawi: evidence from the PROMOTE project pilot randomized controlled trial

Technical Summary: The PROMOTE Project Pilot RCT in Malawi

Problem Statement
Adolescent mothers in Malawi, as in many low-resource settings, face a convergence of structural barriers that prevent their re-entry into formal education or vocational training. These barriers include economic hardship, the lack of affordable childcare, and a deficit in life skills and psychosocial support. While single-intervention approaches (such as cash transfers or childcare subsidies alone) have shown promise in specific contexts, there is a significant evidence gap regarding integrated intervention packages that simultaneously address financial constraints, caregiving responsibilities, and personal development needs. The study posits that these interconnected challenges require a multifaceted solution to effectively disrupt cycles of poverty and educational exclusion among young mothers.

Methodology
The study employed a three-arm, parallel-group, individually randomized controlled trial (RCT) conducted between 2022 and 2024 in Blantyre, Malawi. The trial, registered as AEARCT-0009115, enrolled 288 adolescent mothers aged 15–19 (note: the abstract specifies 15–19, while the methods section mentions 10–19; the analysis focuses on the enrolled cohort) who had a biological child aged 1–3 years.

Participants were randomized in a 1:1:1 ratio into three intervention arms:

• Arm 1 (Control): Life skills training only. This included fortnightly group sessions covering sexual and reproductive health, mental health, gender-based violence, childcare, and financial literacy.
• Arm 2: Life skills training plus subsidized childcare. Participants received vouchers for community-based childcare centers staffed by trained caregivers for the 12-month duration.
• Arm 3: Life skills training plus subsidized childcare plus conditional cash transfers (CCT). This group received the full package of Arm 2, plus quarterly cash transfers of approximately $30 USD via mobile money, contingent on verified enrollment in school or vocational training.

The primary outcome was self-reported participation in school or vocational training during the 12-month intervention period, assessed at endline. The analysis followed an Intention-to-Treat (ITT) framework using binary logistic regression, adjusted for baseline schooling status and stratification variables. Sensitivity analyses, including Inverse Probability Weighting (IPW) and worst-case bounds, were conducted to address the 38.2% attrition rate.

Key Results
At endline, data were available for 178 participants (62% follow-up rate). Attrition was balanced across all three arms. The results revealed a stark divergence in outcomes based on the intervention package:

• Participation Rates:

  • Arm 1 (Life skills only): 3.0% participation.
  • Arm 2 (Life skills + Childcare): 0.7% participation.
  • Arm 3 (Life skills + Childcare + CCT): 12.0% participation.

• Statistical Significance:

  • Compared to Arm 1, participants in Arm 2 had lower odds of enrollment (Adjusted OR = 0.2, 95% CI: 0.03–0.83), though the authors note this difference was not statistically significant in conventional ITT models but became significant in sensitivity analyses.
  • Participants in Arm 3 had significantly higher odds of participating in school or vocational training compared to Arm 1. The adjusted odds ratio was 6.5 (95% CI: 2.54–16.89, p < 0.001). This indicates that the comprehensive package made participants nearly six to seven times more likely to return to education or training than those receiving life skills training alone.

• Sensitivity Analysis:

  • IPW models, which corrected for attrition bias, yielded even larger effect estimates for Arm 3, suggesting the true intervention effect for the most disadvantaged mothers may be larger than the primary ITT analysis indicated.

Key Contributions
The paper contributes empirical evidence that layered interventions are superior to single-component approaches for adolescent mothers. Specifically, it demonstrates that:

1. Life skills training alone is insufficient to overcome structural and economic barriers to education.
2. Subsidized childcare alone (without financial incentives) did not improve re-entry rates and, in some analyses, was associated with lower odds of enrollment compared to life skills alone. The authors suggest this may be due to mothers reallocating freed-up time to immediate income-generating activities rather than education when financial barriers remain.
3. Conditional Cash Transfers (CCT) act as a critical catalyst. The combination of childcare support, life skills, and financial incentives specifically tied to enrollment created a synergistic effect, addressing both direct costs (fees) and opportunity costs (lost income).

Significance and Claims
The authors claim that the study provides robust evidence that a comprehensive intervention package, anchored by conditional cash transfers, can significantly increase the likelihood of adolescent mothers returning to school or vocational training. The significance of the findings lies in demonstrating that integrated social protection and empowerment strategies are necessary to address the interlinked barriers facing young mothers.

The paper concludes that while foundational support (life skills) and logistical support (childcare) are necessary, they are not sufficient without the removal of financial barriers. The study underscores the importance of "real-life help" that removes everyday obstacles, suggesting that policy interventions must move beyond single-focus programs to holistic models that simultaneously tackle economic, caregiving, and psychosocial constraints. The authors note that further research is required to understand how these interventions can be scaled in low-resource settings, particularly given the challenges of attrition and external shocks (such as Cyclone Freddy) encountered during the trial.