In Focus by The Hindu ‘Evidence-based’ approach and RCTs: Can they distort policy-making?
Oct 27, 2025
In a compelling discussion, Jean Drèze, a development economist known for his work on poverty and social policy in India, critiques the reliance on Randomized Control Trials (RCTs) in policymaking. He reveals their ethical limitations and risks of harm, highlighting the Bihar payment experiment as a cautionary tale. Drèze argues that overvaluing RCTs narrows the evidence pool and depoliticizes policy discussions. He emphasizes the need for values and deliberation beyond mere data, warning against aligning RCTs too closely with technocratic approaches.
AI Snips
Chapters
Transcript
Episode notes
What An RCT Actually Does
- An RCT randomly assigns subjects to treatment and control groups to estimate intervention effects by comparing outcomes.
- Jean Drèze cautions this method has limits and can miss why outcomes differ, so it isn't a universal superior evidence standard.
RCTs Can Distort Local Knowledge
- RCTs can disrupt natural learning processes like farmers' observational adoption of new seeds and create ethical and practical complications.
- They often fail to explain mechanisms, create losers, suffer spillovers, and lack external validity across places and years.
Don't Let RCTs Crowd Out Other Evidence
- Value multiple evidence sources instead of elevating RCTs above all other forms of knowledge.
- Treat RCTs as one tool among many and avoid starving policy debates of other valid evidence.