Late-arriving votes and electoral fraud: A natural experiment and regression discontinuity evidence from Bolivia

Author-hosted working-paper version, dated 17 August 2023, of the article published in World Development volume 173 (January 2024) as article 106407. Treating the interruption of the count as a natural experiment and using the 2016 referendum and other parties' votes as controls in difference-in-differences and regression discontinuity designs, the authors estimate fraud amounting to 2.51% of valid votes, which they say was sufficient to change the outcome, and report a statistically significant discontinuous jump in the gap between the incumbent and the runner-up during the shutdown.