Essays in development economics
- Autores: Niklas Heusch
- Directores de la Tesis: Robin Hogarth (dir. tes.), Alessandro Tarozzi (codir. tes.)
- Lectura: En la Universitat Pompeu Fabra ( España ) en 2018
- Idioma: español
- Tribunal Calificador de la Tesis: Mohanon Manoj (presid.), Gianmarco. León (secret.), Judit Vall Castello (voc.)
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX
- Resumen
This doctoral thesis combines three self-contained essays on development economics, cutting across the fields of health, migration, and measurement. While independent in nature, the articles are united by all having an empirical approach, a research question relevant to policy, and roots in theoretical literature.
Informal health care providers are common across developing countries and drug sellers provide a large share of treatment for common (and often deadly) childhood illnesses in sub-Saharan Africa. The first chapter combines covert observation with formal surveys to study the quality of medical treatment for childhood illness provided by informal drug sellers in such a setting in Northern Ghana. I find the quality of treatment for four common childhood illnesses (malaria, diarrhoea, respiratory infections, and anaemia) to be poor: drug sellers provide appropriate treatment in only one third of interactions. I subsequently examine knowledge, effort, and financial incentives as determinants of provider behaviour and find that inadequate knowledge, rather than low effort or adverse financial incentives, is a main constraint to better treatment. Profit incentives, on the other hand, do not appear to be a bottleneck: a simulation exercise suggests that providing better treatment would not diminish drug sellers' profits or increase clients' expenditure.
The second chapter studies rural-urban migration. In developing countries, urban residents are more productive and consume more than their rural peers; recent research has suggested that these differences might stem from unobserved heterogeneity in skill between urban and rural residents, brought about by sorting. I study domestic migration flows in Tanzania and document three results consistent with this explanation. Firstly, above-median educated individuals are three times as likely to leave their agricultural households and move to urban areas as their below-median educated peers; urban movers are substantially better educated. Secondly, movers to urban areas are more likely to be part of formal labour markets at their prior rural locations, but appear to struggle to find adequate employment there and report unemployment more frequently. Thirdly, the out-migration of more educated individuals from agricultural households to urban areas is highly sensitive to households' economic conditions: being able to finance migration to urban areas appears to be an important pre-condition and bottleneck to the sorting of more educated individuals to urban areas.
The third chapter studies proxy means testing (PMT), which promises to identify poor households using information on household assets and demographic characteristics that can easily be collected and verified through surveys. Recently, Brown et al (2016) have criticised PMT for commonly failing to identify poor households. I hence revisit the authors' results and find that poor calibration is a major driver of the poor performance of PMT they find. When I calibrate the poverty rate predicted by PMT to match the known actual poverty rate of the population, I find that PMT performs substantially better: across the 5 countries I examine, PMT correctly classifies 60-70% of poor households, while chance would only correctly classify 40%. The poorest households are the least likely to be missed by PMT, while the households wrongly predicted to be poor do not tend to be among the richest households. While far from perfect, these results suggest that PMT might still provide useful information to users aware of its limitations.
