Improved Early Action Through Precise Targeting, Timely Cash and Early Warning to Mitigate the Impacts of Climate Shocks in Bangladesh

This study, conducted by the BRAC Institute of Governance and Development (BIGD) in collaboration with the University of Oxford with support from GiveDirectly, examines a targeted risk-informed early action pilot program to evaluate the effectiveness of timing of cash transfers, early warning messaging and information, and data-driven innovations in targeting approaches in response to flood risk and vulnerabilities in Bangladesh. This study finds key opportunities and critical obstacles to adopting early actions in high-risk contexts of vulnerability to extreme climate events, climate displacement, and river erosion in the Jamuna river basin.

Researchers: Dr. Munshi Sulaiman; Dr. Rohini Kamal; Rocco Zizzamia

Partner: University of Oxford, GiveDirectly

Timeline: 2023–2025

Status: Ongoing

Contact: Rohini Kamal PhD;


Considering the global risk of flooding, Bangladesh is ranked as the seventh most impacted nation on the climate change account (Global Climate Risk Index 2020). The Ganges and Brahmaputra river erosion by the Himalayan flow and glacial melting will further increase due to global warming, resulting in the displacement of thirty million people and one-seventh of the country’s landmass erosion. The constrained capacity of governments and humanitarian organizations to respond to idiosyncratic climate shocks creates further vulnerabilities to risk exposure. Timely actions to provide relief immediately require having more reliable systems to identify the neediest recipients and deliver aid effectively.


The study will leverage mobile technology and data science innovations to test novel data-driven solutions to identify when to act, whom to target, and deliver cash support and early warnings in a timely manner with limited physical accessibility. The study will vary its timing of providing reliefs as well as the big-data driven targeting approach to select beneficiaries in anticipatory climate-shock and post-climate shock response. It will also generate evidence in three potential knowledge-gaps including the need to design more cost-effective response systems, the need to improve “operational readiness” for early action, and the greater learning demand for risk-informed early actions. Furthermore, the study will address the constraints in identifying and verifying eligible beneficiaries, especially those from marginalized groups such as women.

This study is relevant to SDG 13 (climate action), particularly taking immediate action to combat climate change and associated impacts.


The research design comprises a three-armed randomized control trial (RCT). 8,000 households will be targeted and randomized across the three arms prior to the first treatment. The treatment groups are: 2000 T1 Households (early warning) as a control group, 4000 T2 Households (early warning and early cash transfer before flood) as a treatment group with variation in targeting technology, and 2000 T3 Households (early warning and late cash) as a treatment group providing cash after floods.

Findings and Recommendations