Political Economy of Development and  Conflict in Afghanistan

This thesis explores the political economy of development and conflict in Afghanistan through four empirical papers, focusing on subjective well-being (SWB), social norms, foreign aid, corruption, and conflict. The papers examine these factors at the regional, provincial, and district levels. The first paper (Chapter 2) examines the relationships among income, SWB, and violence, as well as the moderating roles of insecurity and violence in the income-SWB relationship. Using ordered logit models and nationally representative cross-sectional surveys from over 62,000 respondents from 2016 to 2021 covering 34 provinces and approximately 400 districts, the results show that income is positively and statistically significantly associated with SWB. More importantly, this relationship weakens substantially in the presence of insecurity and violence, particularly among higher-income individuals. These findings highlight the complex interplay between economic resources and security threats in shaping well-being in conflict-affected contexts. The second paper (Chapter 3) examines men’s attitudes toward women’s education in Afghanistan at the primary, secondary, and tertiary levels, as well as women pursuing studies in other provinces or abroad. Using nationally representative repeated cross-sectional survey data from over 69,000 respondents collected by the Asia Foundation (2014–2021) across all 34 provinces, we apply multivariate regression analysis to assess attitudinal patterns. We find that men’s attitudes are largely negative, particularly in rural areas. Women's financial contributions to their households are significantly associated with more supportive attitudes toward women's education, particularly at the secondary and tertiary levels, and when studying outside their home province. Perceived insecurity exacerbates negative attitudes, though this effect is less pronounced when women contribute financially. The third paper (Chapter 4) investigates the relationship between foreign aid and conflict in Afghanistan, with a particular focus on the sector-specific effects of aid in education, agriculture, and health. Using provincial-level panel data from the National Statistics and Information Authority (NSIA) and the Armed Conflict Location & Event Data Project (ACLED) covering the period 2017–2023, I employ fixed-effects, lagged, and System Generalized Method of Moments (System GMM) models to address endogeneity and unobserved heterogeneity. The findings show that education aid consistently reduces conflict, demonstrating strong and robust effects, whereas agriculture aid has weaker and less consistent effects. In contrast, health aid has no significant effect on conflict. Overall, these findings highlight substantial sectoral and regional variation in the effects of foreign aid. Finally, the fourth paper (Chapter 5) examines the dynamic relationship between corruption and conflict-related fatalities at the subnational level in Afghanistan. Using data from the Uppsala Conflict Data Program (UCDP) and the Global Data Lab (GDL) covering the period 1990–2022, we construct a panel dataset for eight regions. The empirical strategy integrates fixed-effects models as a baseline specification, count-data estimators appropriate for conflict-related fatalities, and system GMM to account for persistence and potential endogeneity. Across specifications, higher levels of corruption are consistently associated with increased conflict-related fatalities, suggesting that corruption is linked to environments conducive to sustained violence rather than short-lived instability. Lagged estimates indicate that increases in corruption are associated with higher conflict intensity in subsequent periods, underscoring the dynamic nature of the relationship. The findings contribute to the conflict literature by providing rare subnational evidence from a long-running conflict setting and highlighting the role of governance-related distortions in shaping both the intensity and persistence of armed conflict.