Logo: to the web site of Uppsala University

Essay I (with Johannes M. Bos, Saravana Ravindran, and Abu S. Shonchoy): Can governments leverage existing service-delivery platforms to scale early childhood development (ECD) interventions? We experimentally study a large-scale, low-cost home-visiting intervention --- providing materials and counseling --- integrated into Bangladesh's national nutrition program without extra financial incentives for service providers (SPs). We find SPs partially substitute away from nutritional to ECD counseling. Intent-to-treat estimates show positive impacts on child's cognition (0.17 SD), language (0.23 SD), and socio-emotional scores (0.12-0.14 SD). Wasting and underweight rates decline. Older siblings' primary school attendance increases as well. Improved maternal agency, complementary parental investments, and higher take-up of the pre-existing nutrition program are important mechanisms. We estimate a sizeable internal rate-of-return of 18.9%.

Essay II: This paper evaluates a 2011 Swedish reform that introduced tuition fees for international students applying to universities. For identification, I use two variations in a difference-in-differences design: the timing of the reform and the differences in the pre-reform share of international Master's students across the faculties as a proxy for reform exposure. I find that, for the exposed faculties, the cohort size decreases sharply, driven by a decline in international students and partially mitigated by higher enrolment of natives. There are no effects on the composition of these native students in terms of age, gender, and academic ability. Native students who enroll in Master's programs under exposed faculties see higher labor earnings in the short run. These faculties also respond to the reform by offering a wider set of programs. Taken together, these results show how such reforms affect multiple inputs in the education production function and provide one of the first pieces of evidence on higher education policies related to international students.

Essay III (with Avenia Ghazarian): How does human capital investment respond to local economic opportunities? Income gains can increase the demand for schooling while new jobs raise the opportunity costs. We investigate this question in the context of rapid growth in artisanal gold mining in sub-Saharan Africa. We compile 45 waves of Demographic and Health Surveys covering 1.3 million individuals from 14 countries in this region. Identification comes from two sources of variation: one in the global gold price and the other in the exposure of households to places that are geologically suitable for artisanal gold mining. We find that a near-tripling of the global gold price -- reflecting changes between 2005 and 2010 -- leads to a decline in school attendance: by 3.1 pp for 11 to 15-year-olds and by 2.3 pp for 16 to 20-year-olds who live near gold-suitable areas. These reductions are higher for boys. Taken together, these results highlight the potential costs of economic development driven by natural resources.

Essay IV (with Erika Forsberg and Olof Rosenqvist): Family background shapes individual outcomes across the life cycle. While the literature documents how the importance of family background, measured often by the degree of sibling correlation in socioeconomic outcomes, varies across countries, less is known about heterogeneities across social groups within a country. Using Swedish register data, we compare sibling correlations in skills, schooling, and earnings across fine-grained groups defined by parental socioeconomic status (SES). We find that sibling correlations generally decline in parental SES. This pattern holds for skills, schooling, and earnings, and is robust to alternative definitions of parental SES. This result is consistent with theories on reinforcing parental investments but other mechanisms such as complementarities between parental investments and child ability could also be at play. While the exact mechanisms behind the observed socioeconomic gradient in sibling similarity are hard to identify, the results suggest that life is more formed by individual endowments and considerations for children from high SES backgrounds as compared to their low SES counterparts.

Essay V (with Erik Lundberg): Aging populations and labor shortages in skill-intensive sectors have led many countries to pursue targeted policies to attract international talent. We study a migration reform in Sweden that offered international doctoral students from outside the EU an easier path to permanent residency. Implemented in 2014, the reform shortened the required period of residence from eight to four years, allowed these students to obtain permanent residency immediately after graduation, and granted their spouses a work permit during their doctoral studies. Using the European students as a comparison group in a difference-in-differences design, we find that the treated international students are 13.5 pp (23%) more likely to stay in Sweden three years after graduation. Higher settlement prospects also increase their language investments and marriage rates during the PhD. These effects are larger for cohorts that have longer exposure to the reform and for those who carry out their doctoral research in STEM. In addition, the reform raises both employment and language investments among the partners of the treated international students. Taken together, the policy increases permanent residency among international graduates as well as leads them and their families to make long-term choices conducive to integration.