S2 Ep52: Has the randomista revolution gone too far?

Episode Summary

Executive Summary: This episode examines the 2019 Nobel Prize in Economics awarded to Esther Duflo, Abhijit Banerjee, and Michael Kremer and the debate it sparked over randomized controlled trials (RCTs) in development economics. Lant Pritchett argues the field has overemphasized methods and targeted programs at the expense of growth. Oriana Bandiera counters that RCTs and growth research are complementary, but insists stronger collaboration with policymakers is essential.

Main Topics: The Nobel Prize and the rise of RCTs: The episode opens with praise for Duflo, Banerjee, and Kremer and notes that their work on randomized controlled trials transformed development economics and influenced policy worldwide. Critique of method over subject: Lant Pritchett argues the laureates advanced a method, not a subject, and that commitment to RCTs has crowded out broader work on global poverty and growth. Poverty reduction versus growth: Pritchett insists that median income and economic growth are the dominant drivers of poverty reduction across countries, making growth the central policy priority. Policy relevance of RCTs: Bill Easterly’s view is presented that many RCT studies focus on small interventions that are not the kind of large-scale solutions policymakers want or use. Complementarity of research agendas: Oriana Bandiera argues the field should not choose between RCTs and growth theory; instead, both are needed because poverty traps and underdeveloped institutions limit the power of growth alone. Improving researcher–policymaker collaboration: Bandiera emphasizes that the biggest divide is not inside economics but between academics and practitioners, and says research should be co-generated with policymakers.

Key Arguments: Pritchett argues that development economics should focus less on targeted anti-poverty programs and more on raising median income, because cross-country poverty outcomes track income levels almost perfectly. He says the recent attention to RCTs reflects philanthropic incentives to fund visible, creditable anti-poverty action rather than the most important poverty-reducing strategy. He contends that the biggest gains lie in understanding and triggering growth, even if the mechanism is uncertain, because the payoff from growth episodes is enormous. Easterly’s criticism is that the literature has become too centered on small, testable interventions that are unlikely to interest policymakers or meaningfully shift development outcomes. Bandiera responds that RCTs brought development economics into the mainstream and improved the field’s practical impact by engaging policymakers and studying real-world constraints. She argues that poverty reduction and growth are complementary, noting that poverty traps mean growth alone may not be enough to lift everyone out of poverty. Bandiera’s strongest concern is not the RCT agenda itself but the frequent absence of policy involvement in study design, which reduces adoption and impact. She concludes that the real institutional task is building structures that train researchers to co-produce knowledge with policymakers.

Data Points: Nobel Prize year: 2019 - Awarded to Esther Duflo, Abhijit Banerjee, and Michael Kremer Correlation between median income and poverty rate: 0.99 - Pritchett cites cross-country data linking typical income levels to poverty Share of variation explained by other factors: less than 1% - Pritchett says measurement error, distribution, and poverty programs explain under 1% of cross-country poverty variation India growth acceleration gain to GDP: about $3 trillion - Pritchett’s estimate of the net present value of India’s growth acceleration China growth acceleration gain to GDP: $11 trillion - Pritchett’s estimate of the net present value of China’s growth accelerations Time reference for a paper: June 2019 - Pritchett says he completed a paper making these arguments before the Nobel announcement Institutional reference: over 20 years - Pritchett describes the debate with randomized-trial advocates as ongoing for decades

Pivotal Quotes: "They made a commitment to a method, not a subject, and a commitment to method prevented them from studying global poverty." — Lant Pritchett: His critique of the Nobel laureates and of the RCT-centered direction of development economics "The correlation between Median income, the income of the typical person, and the poverty rate in the country is 0.99." — Lant Pritchett: Evidence used to argue that growth, not targeted programs, explains most poverty reduction "I think the true division is, is between practitioners and economists." — Oriana Bandiera: Her view that the main challenge is bridging academia and policy implementation

Implications: The debate suggests development economics must balance rigorous causal evaluation with big-picture growth research. Future impact will depend less on choosing one method and more on integrating RCTs with policymaker-led, scalable interventions.