Using an original survey, I also find that phone use is associated with greater discussion of the economy, corruption, and politics with relatives. (10/n)

Using panel data and questions about contact with specific family members over time, I show that “non-physical” interaction with urban relatives rises significantly as rural households gain access to a phone between waves. This directly shows that phones increase urban-rural connectivity. (9/n)

I find that entering coverage reduces rural trust in government, across a range of institutions of the state. This effect is driven by those living geographically further from urban centres, and in countries with relatively greater pre-existing urban-rural attitudinal divides. (7/n)

I triangulate several pieces of data to support for the argument. First, I draw on Afrobarometer surveys and mobile coverage maps from across the African continent, locating survey respondents as in our outside coverage over time in a difference-in-differences setting. (6/n)

I suggest that rural households use mobile phones to keep regular communication with relatives who have migrated to towns and cities. This is politically consequential, as urban areas across Africa tend to be significantly less trusting of government and so provide new, negative, information. (4/n)

The paper is motivated by rising mobile phone access around the world. This represents a rapid shock in our ability to send and receive information, but, unlike other types of information or media technology, is strikingly homogeneous and without clear income gradient. (1/n)