Going the Distance people in economics
Source details
- Canonical URL
- Going the Distance people in economics
Other formats
Bibliographic details
- Authors: CHRIS WELLISZ
- Published: June 2, 2022
Profile summary
- Chris Wellisz profiles Harvard’s Melissa Dell, who pioneers micro-level methods to identify long-run effects of historical institutions.
- Published in F&D Magazine: "Melissa Dell: Going the Distance", CHRIS WELLISZ, June 2022.
- Notable recognition: her article “The Persistent Effects of Peru’s Mining Mita” was published in Econometrica in 2010 and featured in the American Economic Association’s decision in 2020 to award her the John Bates Clark Medal.
Key research contributions
- Emphasis on micro-level analysis of neighboring towns and villages to uncover channels of persistence in development.
- Identification of "channels of persistence" where historical institutions affect present infrastructure and livelihoods (example: the mita’s impact on roads and subsistence farming).
- Refinement and popularization of geographic regression discontinuity designs, extending discontinuity regression methods from individual eligibility cutoffs (e.g., age 65 for Medicare) to spatial boundaries.
Major studies and findings
- Peru mita study:
- Finding: Areas subject to the Spanish-era mita had fewer paved roads, higher likelihood of subsistence farming, and greater poverty relative to adjacent non-mita areas.
- Mechanism: Fewer haciendas inside the mita led to less political pressure for road construction; outside the mita, hacienda owners had roads built to access markets.
- Vietnam study ("The Historical State, Local Collective Action, and Economic Development in Vietnam", coauthored with Nathan Lane and Pablo Querubin):
- Comparison: former Dai Viet (strong, centralized, village-based administration) versus former Khmer (patron-client landlords).
- Findings: Household consumption in former Dai Viet lands was about one-third higher than in the former Khmer region.
- Civic outcomes: Citizens of former Dai Viet villages were twice as likely to participate in local civic institutions, were better educated, and had better access to health care.
- Vietnam wartime security-score analysis:
- Source data: surveys of 18,000 hamlets conducted from 1969 to 1973 with 169 questions per survey.
- Security scores: originally ranged continuously from 1 to 5 but were rounded to the nearest integer, producing discontinuities (example threshold effects around 2.49 vs 2.51).
- Result: After reconstructing unrounded scores using conditional probability matrices, Dell and Querubin concluded that overwhelming bombing made villagers more likely to support the communist insurgency and weakened noncommunist civic activities.
Methods and tools
- Geographic regression discontinuity design:
- Extended classic regression discontinuity (e.g., Medicare at age 65) to geographic boundaries, requiring GIS and more complex calculations.
- Archival and fieldwork approach:
- Combines detailed archival work (e.g., US National Archives, US Army Center of Military History) with field interviews (example: Cusco, Peru bus visits to Quechua-descendant villages).
- Technical innovations:
- Self-taught GIS mapping to precisely locate villages relative to institutional boundaries.
- Developed a package called Layout Parser based on open-source software to digitize early 20th century Japanese company records with unusual layouts and obsolete characters.
- Working on an alternative optical character recognition approach tailored to historical documents.
- Research practices highlighted:
- Persistent, hands-on data recovery (e.g., locating conditional probability matrices in basement archives).
- Teaching and disseminating methods (taught classes on GIS and deep learning methods for data curation).
Intellectual stance and policy implications
- Emphasis on grounding policy and theory in data rather than flashy untested models (critiques of advisors such as McGeorge Bundy and Walt Rostow for insufficient empirical testing).
- Demonstrates that interventions based on overwhelming force or simplistic theories can have counterproductive effects (Vietnam bombing example).
- Methodological diffusion: Dell’s geographic regression discontinuity work and data-curation tools lower barriers for other scholars to analyze historical institutional effects.
Personal background and life
- Grew up in Enid, Oklahoma (city of about 50,000).
- Education and activities:
- Undergraduate at Harvard (majored in economics, ran long distance competitively).
- Senior thesis on the mita; two-year fellowship at the University of Oxford; PhD at MIT.
- Family:
- Married to Austin Huang; three children ranging in age from four months to four years.
- Extracurriculars:
- Former ultramarathon runner; notes parallels between distance running and research persistence.
F&D Magazine — Melissa Dell: Going the Distance (CHRIS WELLISZ, June 2022).
Content in this bundle
- Going the Distance: People in Economics profile of Melissa Dell