Research
Publications:
Publications:
When Private Beats Public: A Flexible Value-Added Model with Tanzanian School Switchers, Economic Development and Cultural Change, Forthcoming.
The popularity of private schools is increasing across Sub-Saharan Africa. Surprisingly, however, little is known about the return to private secondary education in this region. In this paper, I estimate a private school learning premium in Tanzania, using unique administrative data on primary and secondary school exams for 635,000 students. Specifically, I match secondary school students with their primary school schoolmates who achieved the same primary school exam scores, and control for peer effects and unobserved ability. The size of the dataset further allows me to investigate various dimensions of heterogeneity. On average, private schools improve exam scores by 0.54 of a standard deviation in two years. An instrumental variable model suggests the effect is causal, and subject-specific estimates are all positive but higher for mathematics relative to Kiswahili and English.
The popularity of private schools is increasing across Sub-Saharan Africa. Surprisingly, however, little is known about the return to private secondary education in this region. In this paper, I estimate a private school learning premium in Tanzania, using unique administrative data on primary and secondary school exams for 635,000 students. Specifically, I match secondary school students with their primary school schoolmates who achieved the same primary school exam scores, and control for peer effects and unobserved ability. The size of the dataset further allows me to investigate various dimensions of heterogeneity. On average, private schools improve exam scores by 0.54 of a standard deviation in two years. An instrumental variable model suggests the effect is causal, and subject-specific estimates are all positive but higher for mathematics relative to Kiswahili and English.
WORKING PAPERS:
WORKING PAPERS:
Illicit Financial Flows and the Global South: A Review of Methods and Evidence (link to working paper)
Illicit financial flows (IFFs) constitute a major challenge for development in the Global South, as domestic resource mobilization is imperative for providing crucial public services. While several methods offer to measure the extent of IFFs, each has its benefits and drawbacks. Critically, methods based on the balance of payments identity may capture licit as well as illicit flows, and a method based on macroeconomic trade discrepancies suffers from doubtful assumptions. The most convincing estimate to date demonstrates that individuals hold financial assets worth around ten percent of global GDP in tax havens. Evidence further indicate that countries in the Global South are more exposed to individuals and firms illicitly transferring money out of the country. Further research is warranted on profit shifting out of countries in the Global South and the effectiveness of anti-IFF policies in countries outside Europe and the US.
Illicit financial flows (IFFs) constitute a major challenge for development in the Global South, as domestic resource mobilization is imperative for providing crucial public services. While several methods offer to measure the extent of IFFs, each has its benefits and drawbacks. Critically, methods based on the balance of payments identity may capture licit as well as illicit flows, and a method based on macroeconomic trade discrepancies suffers from doubtful assumptions. The most convincing estimate to date demonstrates that individuals hold financial assets worth around ten percent of global GDP in tax havens. Evidence further indicate that countries in the Global South are more exposed to individuals and firms illicitly transferring money out of the country. Further research is warranted on profit shifting out of countries in the Global South and the effectiveness of anti-IFF policies in countries outside Europe and the US.
WORK IN PROGRESS:
WORK IN PROGRESS:
The Impact of Eliminating Secondary School Fees: Evidence from Tanzania(with Beatrice K. Mkenda)
Tanzania implemented a fee-free secondary school reform in January 2016. Using variation in district and cohort exposure to the reform, we employ a difference-in-differences strategy to estimate the short-term impacts of the reform. The reform substantially increased enrolment into secondary education. While these enrolment effects were predominantly driven by an increase in public school enrolment, there was also a delayed positive effect on private school enrolment. Districts mostly affected by the reform experienced a significant drop in exam scores relative to less affected districts, which cannot be explained by academic abilities of new students. These findings are in line with a theoretical model on school choice, where some individuals are credit-constrained and the quality of public education is harmed by increased enrolment.
Predicting Local State Capacity in Sub-Saharan Africa: A Machine Learning Approach(with Gustav Agneman, Christoffer Cappelen, and David Sjöberg)
Despite the need to measure state capacity at a sub-national level, most studies still use country-level indicators as rough approximations of the local counterpart. We estimate a measure of state capacity at the 2.5×2.5 arc-minutes grid cell level (≈ 5 kilometers) for Sub-Saharan Africa. The measure builds on geocoded survey-based data on local state capacity which we predict and extrapolate using an ensemble of regression trees. We demonstrate the usefulness of measuring state capacity at a disaggregated level by including our local state capacity index as a moderating factor in the relationship between oil wealth and armed conflict. The findings suggest that cells with higher local state capacity face lower risks of conflicts caused by oil price hikes.
Tanzania implemented a fee-free secondary school reform in January 2016. Using variation in district and cohort exposure to the reform, we employ a difference-in-differences strategy to estimate the short-term impacts of the reform. The reform substantially increased enrolment into secondary education. While these enrolment effects were predominantly driven by an increase in public school enrolment, there was also a delayed positive effect on private school enrolment. Districts mostly affected by the reform experienced a significant drop in exam scores relative to less affected districts, which cannot be explained by academic abilities of new students. These findings are in line with a theoretical model on school choice, where some individuals are credit-constrained and the quality of public education is harmed by increased enrolment.
Predicting Local State Capacity in Sub-Saharan Africa: A Machine Learning Approach(with Gustav Agneman, Christoffer Cappelen, and David Sjöberg)
Despite the need to measure state capacity at a sub-national level, most studies still use country-level indicators as rough approximations of the local counterpart. We estimate a measure of state capacity at the 2.5×2.5 arc-minutes grid cell level (≈ 5 kilometers) for Sub-Saharan Africa. The measure builds on geocoded survey-based data on local state capacity which we predict and extrapolate using an ensemble of regression trees. We demonstrate the usefulness of measuring state capacity at a disaggregated level by including our local state capacity index as a moderating factor in the relationship between oil wealth and armed conflict. The findings suggest that cells with higher local state capacity face lower risks of conflicts caused by oil price hikes.
BOOK CHAPTERS:
BOOK CHAPTERS:
Characteristics of the VARHS Data and other Data Sources(with Finn Tarp)
In Finn Tarp (ed.) Growth, Structural Transformation, and Rural Change in Viet Nam: A Rising Dragon on the Move, Oxford University Press, 2017.
In Finn Tarp (ed.) Growth, Structural Transformation, and Rural Change in Viet Nam: A Rising Dragon on the Move, Oxford University Press, 2017.
POLICY WORK:
POLICY WORK:
Characteristics of the Vietnamese Business Environment: Evidence from a SME Survey in 2015(with John Rand, Smriti Sharma, Finn Tarp, and Neda Trifkovic)
UNU-WIDER: Helsinki. October 2016 report.
UNU-WIDER: Helsinki. October 2016 report.