The Covid-19 crisis has strained social relationships to unprecedented levels. Social distancing measures have prevented many forms of engagement in public affairs for two years now, often resulting in undermined cohesion and trust in institutions (Daniele et al. 2020). However, the erosion of British social capital is anything but a new phenomenon, with indicators of trust, social interaction, and civic engagement reportedly declining since the second half of the 1990s.
What is social capital?
Economists refer to social capital as all the features of social life, such as networks of relationships, civic engagement, and trust, that enable individuals to act together more effectively to pursue shared objectives. Civic networks can nurture members’ concern for public welfare and willingness to live by the norms of the community and punish deviant behaviours. From an economic perspective, the cooperative attitudes that stem from social capital can reduce transaction and monitoring costs, encourage investments, and improve the allocation of resources (e.g. Algan and Cahuc 2007, Bazzi et al. 2018).
Why has social capital been declining for three decades?
In his bestseller Bowling Alone, Robert Putnam (2000) suggested that television may have displaced relational activities in individuals’ leisure time in the last decades of the 20th century. In line with this argument, it seems plausible that the Internet, which provides on-demand content and allows for interactive communication, might have induced an even more powerful substitution effect in more recent years.
Does the time we spend online displace our civic engagement and political participation? Is the Internet weakening our willingness to comply with social norms of cooperation and mutual respect, making us more self-centred and isolated than before?
New study on the UK
In a recent paper (Geraci et al. 2022), we answer these questions by constructing a unique dataset on Internet access in the UK and matching this to survey data on social capital. We first collect the geolocation of the network nodes and the blocks served by each of them. This allows us to calculate the distance between each survey respondent’s telephone line and the relevant node of the voice network, and therefore to study how the topology of the old telephone network made of copper wires affected broadband penetration. To assess the causal impact of fast Internet on social capital, we match these data on Internet access with geocoded, longitudinal survey data collected annually between 1998 and 2018 by the British Household Panel Survey and the UK Longitudinal Household Study (Understanding Society).
The distance between each survey respondent’s telephone line and the relevant node of the voice network is a critical factor in the use of fast Internet. Broadband access long relied on the digital subscriber line (DSL) technology, allowing data transmission over the old telephone infrastructure made of copper wires. However, the speed of a DSL connection rapidly decays with the distance of a final user’s telephone line from the network node serving the area, also called the ‘local exchange’.
When the network was designed in the 1930s, the length of the copper wire connecting houses to local exchanges did not affect the quality of voice communications. However, the introduction of DSL technology in the 1990s unexpectedly turned distance from the local exchange into a critical determinant of broadband access. This technical change caused exogenous variation in the quality of Internet access, as access to a faster connection was increasing in the proximity to a node of the network (Figure 1, left panel), and stimulated greater use of the Internet (Figure 1, right panel).
Note: The x-axis measures the residuals of a regression of the distance between the house and the respective local exchange on the socio-demographic control variables. The y-axis of the left panel reports the residual of a regression for the presence of broadband at home on the socio-demographic control variables. The y-axis of the right panel reports the residual of a regression for the time spent online on the socio-demographic control variables.
The impact of broadband Internet on bridging and bonding social capital
Our empirical analysis shows that fast Internet substantially displaced several dimensions of social capital in the UK. After broadband take-up, civic and political engagement started to systematically decline with inhabitants’ proximity to the network node serving the area, i.e. with the speed of the Internet connection.
Time-consuming activities oriented to the pursuit of collective welfare, such as engagement in associations, suffered the most from broadband penetration. Putnam (2000) labelled associational life as a form of ‘bridging social capital’ that bridges people with different backgrounds, fosters cohesion, and encourages cooperative attitudes. Figure 2 illustrates the event study analysis supporting these conclusions.
The effect is statistically significant and sizable. A 1.8 km reduction in respondents’ distance from the local exchange, resulting in a faster connection, caused a 4.7% decline in the likelihood of participation in associational activities between 2005 and 2017. For political parties, broadband availability caused a statistically significant 19% reduction in the probability of involvement. For volunteering associations, the likelihood of people participating in these organisations reduced by 10.3%.
Note: Each panel plots the coefficients and 90% confidence intervals associated with the interactions between wave dummies and the Distance variable, using the first available wave as baseline. Standard errors are clustered at the level of the LSOA.
Nonetheless, the displacement spared relationships with family and friends, which is generally labelled as ‘bonding social capital’ in the literature. Social scientists often blame bonding social capital for pushing individuals to focus on particularistic goals, potentially harming trust in others, cohesion, cooperation, and development (Muringani et al. 2021).
In this context, while bonding social capital seems resilient to technological change, bridging social capital proves fragile and vulnerable to the pressure of new media on users’ time allocation choices. This result suggests an adverse effect of technological progress by undermining an essential factor of social capital and the well-functioning of democratic institutions, possibly paving the way for the rise of populism.
However, the pattern that we document for the British case may be case-specific, and will not necessarily hold in other contexts. Results for the UK must be understood in connection with previous, conflicting evidence on the outcomes of broadband penetration, suggesting that fast Internet did not displace social capital in Germany (Bauernschuster et al. 2014). The behavioural and societal impact of fast Internet may vary depending on the initial stock of social capital, institutional background, and activities users perform online.
Algan, Y, and P Cahuc (2007), “Social attitudes and economic development: an epidemiological approach”, VoxEU.org, 2 October.
Bazzi, S, A Gaduh, A Rothemberg and M Wong (2018), “How intergroup contact can foster nation-building”, VoxEU.org, 7 January.
Bauernschuster, S, O Falck and L Wößmann (2014), “Surfing Alone? The Internet and Social Capital: Quasi-Experimental Evidence from an Unforeseeable Technological Mistake”, Journal of Public Economics 117: 73-89.
Daniele, G, A F M Martinangeli, F Passarelli, W Sas and L Windsteiger (2020), “Covid-19 and socio-political attitudes in Europe: In competence we trust”, VoxEU.org, 1 October.
Geraci, A, M Nardotto, T Reggiani and F Sabatini (2022), “Broadband Internet and Social Capital”, Journal of Public Economics 206(3): 104578.
Muringani, J, R Fitjar and A Rodríguez-Pose (2021), “Bridging, not bonding, for regional growth”, VoxEU.org, 20 April.
Putnam, R D (2000), Bowling Alone: The Collapse and Revival of American Community, New York: Simon & Schuster.