Urban Demography and Political Mobilization in African Cities

• In Event: New Perspectives on the Comparative Politics of Cities in the Global South

Abstract

How does the changing demographic context of unique African cities affect the political behaviour and mobilization of young people? Africa is undergoing a rapid process of demographic change characterized by urbanization which has profound implications for the politics of the continent. Demographically, urbanization has largely been driven by the ‘natural growth’ of urban populations from within (Menashe-Oren and Bocquier, 2021; Bocquier, Menashe-Oren, and Nie, 2023), implying national population structures increasingly defined by large youth cohorts. Yet there is considerable subnational variation in the rate and composition of this demographic change, leading to diverse population structures between cities within and across countries. As such, how the local demographic context of these cities conditions the political behaviour of individuals remains an important unanswered question.
Research on the political implications of demographic structure and change has failed to yield consistent results. The longstanding literature on formal political participation (i.e., voting in elections) in North American and European democracies almost universally shows that young people are less likely to vote, or that the age-voting relationship follows an inverted-U shaped curve (Henn and Foard 2012; Franklin 2004; Norris 2002; Wattenberg 2003; Nie, Verba and Kim 1974). Recent studies in the African context support this, showing lower rates of youth participation in formal politics (Resnick and Casale 2011; Lekeke and Gyimah-Boadi 2016; Tambe and Kopacheva 2023).
As such, research has often focused on the relationship between age and informal modes political participation, with so called ‘youth bulges’ thought to increase the likelihood of collective protest and/or violence (Goldstone, 2002; 2010; Urdal, 2006; 2008; Urdal and Hoelscher, 2012; Buhaug and Urdal, 2013; Menashe-Oren, 2020). Yet while young people may be more likely to participate in protests (Dim and Schafer, 2021), there is no clear relationship between large youth cohorts and contentious collective action at the meso or macro scale (Urdal and Hoelscher, 2012).
Empirically, the mechanisms linking youth cohorts to formal and informal participation and mobilization tend to be local or individual. However, theory is typically operationalised at the country level or in a comparatively small sample of large cities, which masks the heterogeneity of demographic structure between different urban areas within a country (Dorward and Fox, 2022). This presents a gap in our understanding of how the demographic context of individual cities affects formal and informal expressions of individuals’ political behaviour; if changing urban demography matters for political outcomes, this should be observed at the city-level. As individuals primarily compete in local labour markets rather than national ones, local urban contexts characterized by large youth cohorts are more likely to see young people unemployed or engaged in informal and insecure work. We anticipate this reduces belief in the efficacy of formal political participation (Resnick and Casale 2014) and increases dissatisfaction with formal politics. Moreover, we expect that individuals in urban contexts with large youth cohorts would provide larger recruitment pools to mobilize into protest actions, although economic grievances may not always translate into protest mobilization (Resnick 2015). All else equal, we expect to find that young people in cities characterized by large youth cohorts are less likely to participate in formal politics (i.e. voting behaviour) due to disaffection and more likely to participate in politics through informal routes (i.e. protest participation).
Building upon recent advances in geospatial data analysis (Dorward and Fox, 2022), we test these expectations by combining fine-grained data on the spatial extent of unique urban settlements with estimates of urban age and sex structures from World Pop, the location of contentious events from the Armed Conflict Location and Event Dataset (ACLED), and geocoded individual-level survey data from Afrobarometer. This cross-sectional dataset allows us to measure the political behaviour and mobilization of 11430 individuals in 347 cities in 32 African countries. We fit a series of multilevel regression models with cross-level effects nesting individuals within cities within countries to estimate how the demographic contexts of unique urban settlements are associated with indicators of individual political attitudes and behaviour (voting and protest participation) and the probability of urban riots and protests at the city level. In doing so, this paper presents new city-level evidence on how urban demographic composition is associated with the formal political participation and contentious mobilization of young people in cities.

Authors

• Nick Dorward, University of Bristol

• Kristian Hoelscher, PRIO