Discussion Paper

E-government and democracy in Botswana: observational and experimental evidence on the effect of e-government usage on political attitudes

Bante, Jana / Felix Helmig / Lara Prasad / Lea Deborah Scheu / Jean Christoph Seipel / Helge Senkpiel / Markus Geray / Armin von Schiller / David Sebudubudu / Sebastian Ziaja
Discussion Paper (16/2021)

Bonn: German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

ISBN: 978-3-96021-153-2
DOI: https://doi.org/10.23661/dp16.2021
Price: 6 €

Appendix to Discussion Paper 16/2021

This study assesses whether the use of electronic government (e-government) services affects political attitudes. The results, based on evidence generated in Botswana, indicate that e-government services can, in fact, have an impact on political attitudes. E-government services are rapidly being rolled out around the globe. Governments primarily expect efficiency gains from these reforms. Whether e-government in particular, and information and communication technology (ICT) in general, affect societies is hotly debated. There are fears that democracy may be compromised by surveillance, censorship, fake news, interference in elections and other strategies facilitated by digital tools. This discussion paper adds to the nascent literature by investigating if the expanding e-government usage in Botswana affects individual support for democracy, regime satisfaction and interpersonal trust. Methodologically, the study relies on observational and experimental evidence. The observational approach assesses the impact of the usage of different e-services such as e-payments and electronic tax return filings on political attitudes. The experimental approach incentivises taxpayers to file their tax returns electronically. Both approaches build on an original in-person survey gauging the political attitudes of 2,109 citizens in Greater Gaborone. The survey was conducted in February and March 2020. In terms of results, we do not identify a general substantive effect for the impact of all e-services on political attitudes. For some of the e-services and attitudes tested, however, we find significant evidence. Furthermore, our study yields significant results for several of the linkages between the causal steps within our causal mechanisms. For instance, we find that e-government can empower citizens to engage in political activities and that, although e-government users on average report that the government is not addressing their needs, a simple incentivising message can significantly improve people’s feelings in this regard.