Spatial Spillovers of Conflict in Somalia - joint with Thomas Cornelissen - Revise and Resubmit American Economic Review
Due to economic interconnectedness across regions, locally confined violent conflict may have welfare effects far beyond directly affected areas. This paper focuses on Somalia’s al-Shabaab insurgency and investigates whether the food transportation network propagates the effects of violent conflict to distant locations. Combining granular geospatial information on agricultural areas, roads, and itineraries, we show that conflict along transportation routes significantly increases food prices at markets located hundreds of kilometers away. Standardized estimates amount to up to half the magnitude of the effect of rainfall. Negative effects of conflict on road traffic as measured by satellite images of light emissions point towards decreases in food transportation. Moreover, conflict decreases food security, nutrition, health, and education for households living in far-away market areas. This suggests that food prices act as a propagating mechanism that links – among others – human capital to far-away conflict. Back-of-the-envelope calculations suggest that spatial spillovers add an additional 30% to the welfare cost of local conflict.
Media: The Conversation, Yahoo! News
Strategic Drones - joint with Margaux Clarr, Jaime Marques-Pereira, and Jean-Francois Maystadt
US drone strikes are popular with the electorate and overseen by the President. This paper investigates whether the US President uses drone strikes strategically for political gain. We document that US drone strikes in Afghanistan, Pakistan, Somalia, and Yemen are significantly more likely before US elections, when popularity has high payoffs. We find no changes for unpopular, piloted airstrikes. Consistent with unusually high drone approvals, abnormally cloudy skies before US elections lead to a postponement or redirection of strikes to other target countries. To examine whether drone strikes are used strategically to divert attention from damaging media coverage, we gather closed captions from all cable TV coverage of the President and analyze their tone using natural language processing. Drone strikes are more likely in weeks when news anchors cover the President more negatively, a relation that holds both during and outside of election periods. We find no such relationship for piloted airstrikes or during weeks of high news pressure.
Coping with Weather Shocks - joint with Joseph-Simon Gorlach
Accounting for multiple responses to weather shocks drastically changes policy implications for adaptation to increasingly variable weather. Kenyan households react to temperature anomalies by sending migrants, by transiting to less climate-sensitive occupations, and by changing livestock species. Evidence suggests these are short-term adjustments, which respond significantly to common interventions. Randomised income transfers cushion consumption losses and decrease adaptation pressure, such as migration. Better infrastructure, instead, eases occupational transitions, reducing alternative adjustments, including migration and livestock composition. A model of joint migration, occupation, and livestock choices reveals long-term effects of these short-term shocks. Transitions to non-agriculture first act as a substitute for migration and subsequently as a stepping stone for later migration.
Temperature and Confidence in Government in Africa - joint with Gabriel Aboyadana
This paper documents a significant effect of daily temperature anomalies on confidence in government. We combine Africa-wide attitudinal survey and high-frequency climatic data to calculate temperature on the day and at location of interview. We find that positive temperature anomalies on interview day decrease ten self-reported measures for confidence in government. Anomalies two days before and after the interview have no effect. Results also show effects on voting intentions, government satisfaction, and aggression. Evidence suggests that positive temperature anomalies decrease confidence predominantly by amplifying pre-existing grievances of disadvantaged respondents. We also find that temperature anomalies affect voting behaviour and protests.
Widening the Gender Gap: The Unintended Consequences of Conditional Cash Transfers in India - joint with Wiji Arulampalam and Artemisa Flores
This paper investigates the unintended consequences of the introduction of a large conditional cash transfer programme (Janani Suraksha Yojana) aimed at improving maternal and neonatal mortality in India. We concentrate on two outcomes not targeted by the scheme: sex ratios at birth and gender gaps in neonatal survival. We exploit the incremental implementation of the programme using the District Level Household Survey data to estimate these effects in a difference-in-differences framework, which compares districts treated in the first year of the scheme with districts treated in subsequent years. The results suggest that the programme significantly suppressed sex ratios at birth, especially in areas characterised by strong preferences for sons. By contrast, we do not find an impact on gender specific mortality. We also provide evidence that the introduction of the programme increased pregnant women’s exposure to ultrasound scans.
"Female autonomy and the education of the subsequent generation: Evidence from three contrasting states in India" - joint with Wiji Arulampalam and Uma Kambhampati - IZA Discussion Paper 6019
This paper makes a significant contribution on both conceptual and methodological fronts, in the analysis of the effect of maternal autonomy on school enrolment age of children in India. The school entry age is modelled using a discrete time duration model where maternal autonomy is entered as a latent characteristic, and allowed to be associated with various parental and household characteristics which also conditionally affect school entry age. The model identification is achieved by using proxy measures collected in the third round of the National Family Health Survey of India, on information relating to the economic, decision- making, physical and emotional autonomy of a woman. We concentrate on three very different states in India – Andhra Pradesh, Kerala and Uttar Pradesh. Our results indicate that female autonomy is not associated with socio-economic characteristics of the woman or her family in Kerala (except maternal education), while it is strongly correlated to these characteristics in both Andhra Pradesh and Uttar Pradesh. Secondly, while female autonomy is significant in influencing the school starting age in UP, it is less important in AP and not significant at all in Kerala.
In Refereed Journals
"Terrorism and Education: Evidence from Instrumental-Variables Estimators" - joint with Joseph-Simon Gorlach
Journal of Applied Econometrics (forthcoming)
Media: VoxDev
"Terrorism, Media Coverage, and Education: Evidence from al-Shabaab Attacks in Kenya " - joint with Joseph-Simon Gorlach
Journal of the European Economic Association, 21, 2, 727 - 763 (2023)
"Cross-occupational effects of immigration on native wages in the UK" - joint with Ross McKenzie and Graeme Roy
Applied Economics Letters, 1-5 (2023)
Media: Faculti Interview
"Islamic Law and Investments in Children: Evidence from the Sharia Introduction in Nigeria"
Journal of Health Economics, 85, 102660 (2022)
"Daughters, Dowries, Deliveries: The Effect of Marital Payments of Fertility Choices in India"
Journal of Development Economics, 125, 89 - 104 (2017)
"Physician shortages in rural Vietnam: using a labor market approach to inform policy" - joint with Marko Vujicic, Bakhuti Shengelia, Ha Bui Thu
Social Science and Medicine, 73 (7), 970 - 977 (2011)
"Access to non-pecuniary benefits: does gender matter? Evidence from six low- and middle-income countries" - joint with Neeru Gupta
Human Resources for Health, 9 (25), (2011)
Book Chapters
"Immigration and the UK: Reflections after Brexit" (2016) in Refugees and Economics Migrants: Facts, Policies and Challenges, CEPR - joint with Christian Dustmann and Tommaso Frattini
Policy Reports
"What do we know about migration? Informing the debate" (2014) available as CReAM Discussion Paper - joint with Michele Battisti, Charles Clarke, Thomas Cornelissen, Christian Dustmann, Francesco Fasani, Tommaso Frattini, Simon Gorlach, Luigi Minale, Anna Okatenko, Ian Preston and Jonathan Wadsworth.
"Policy options to attract nurses to rural Liberia: evidence from a Discrete Choice Experiment" (2010) available as HNP Discussion Paper - joint with Marko Vujicic, Mandy Ryan, C. Sanford Wesseh, Julie Brown-Annan.