The Uneven Economics of Protests: Understanding Social Unrest and Income Dynamics in Pakistan
Description
This dataset supports the study examining the association between social unrest and district-level income in Pakistan during 2011–2021. It integrates household income data from the Pakistan Social and Living Standards Measurement Survey (PSLM/HIES) with geocoded records of political violence and protest events from the Armed Conflict Location and Event Data Project (ACLED). The dataset was constructed as a balanced district-year panel to facilitate longitudinal and spatial analysis of economic outcomes and social unrest. The research is guided by the hypothesis that different forms of social unrest have heterogeneous associations with district-level income. Specifically, protests and riots may exhibit different economic associations from violence against civilians, strategic developments, remote violence, and battles, and these relationships may vary across the income distribution and geographic space. Income observations are available for the 2011, 2013, 2015 and 2019 PSLM/HIES survey rounds. Forecast values were generated for 2017 and 2021, while intermediate years were obtained through interpolation to create a balanced annual panel. The repository contains both original and extended panel datasets, district-level summary statistics, and spatial outputs used in the empirical analysis. Social unrest variables are derived from ACLED event records and include annual district-level counts of riots and protests (RP), violence against civilians (VAC), strategic developments (SD), remote violence (RV), and battles (BTL). Additional variables include fatal and non-fatal event counts, fatality-weighted measures, spatially lagged unrest variables, district coordinates, literacy, household size, urban population share, working-age population share, province identifiers, and alternative income measures used for robustness analysis. The data are suitable for panel data, spatial econometric, and distributional analyses. The accompanying variables enable replication of fixed-effects, panel quantile regression, spatial lag, and robustness specifications reported in the associated study. Spatial variables were constructed using district centroids and spatial weight matrices to facilitate tests of spatial dependence. The data should be interpreted as observational rather than causal. Social unrest variables represent annual counts of recorded ACLED events aggregated to the district level, while income represents average district household income derived from PSLM/HIES. Forecasted and interpolated income values are included solely to construct a balanced panel and are clearly distinguished from directly observed survey rounds for robustness analyses. The dataset may also be reused for research on regional development, political economy, conflict economics, spatial inequality, and district-level socioeconomic analysis in Pakistan.
Files
Institutions
- University of the PunjabPunjab, Lahore