Africa’s Emerging Middle Class

What are the Implications for Development and Democracy?

11 December 2013

Danielle Resnick

Economic transformation and social mobility are currently popular themes in the development community. Both themes are especially relevant to analyses of the small but growing middle class in sub-Saharan Africa. Due to impressive rates of economic growth over the last decade—coupled with improvements in high-value services such as finance, communications, and technology—the region has witnessed a notable, albeit still modest, expansion of this constituency. In fact, according to the African Development Bank (AfDB), Africa’s middle class has grown from 200 to 300 million during 2000-10. [1]

Yet, aside from journalistic accounts and reports from the private sector, there has been little scholarly analysis of Africa’s middle class and its implications for the region’s development trajectory and democratic consolidation. UNU-WIDER’s project ‘Africa’s Middle Class’ attempts to fill this gap through a network of experts on African politics and economic development. As a side event of the 2013 African Studies Association annual meeting, project participants met in Baltimore on 23 November to discuss preliminary findings emerging from their collective research.

The research underscores a number of key issues. First, the conceptualization and measurement of the middle class remains a central challenge. The difficulty is finding a definition that resonates with common views of the middle class without conflating expected behaviors of this constituency within the definition. Absolute approaches, which involve defining income or expenditure thresholds within which the middle class should be situated, aim to get around this challenge but they can often be too simplistic. For instance, the AfDB relies on an absolute approach that results in classifying the middle class as those with a daily per capita consumption between US$4-20 per day in 2005 PPP. However, based on using living conditions monitoring surveys (LCMS), the UNU-WIDER project has undertaken a number of profiles of the middle class and uncovered that these AfDB thresholds obscure a wide degree of heterogeneity across the continent. Specifically, in Uganda, only 10 percent of those with an income of US$10 per day live in a house or apartment with electricity, piped water, and a flush toilet, compared with more than 70 per cent in Zambia. In other words, Zambia’s middle class is substantially better off in terms of housing than their Ugandan counterparts.

Project researchers have also examined alternative measurements using different data sources. Based on public opinion data from Afrobarometer, the middle class was also considered to be defined as those who have never gone without income, food, clean water, cooking fuel, and medical care, who own a television, and who have piped water in their dwelling. Using this definition in Africa’s largest economy, South Africa, reveals that only six percent of the black middle class would fall into this category.

Secondly, the middle class is not necessarily a new phenomenon in Africa but the project uncovers major differences with its prior incarnation. Notably, the African middle class historically was generated by close relationships with the state, which was the primary source of employment and contracts for the indigenous business sector. By contrast, today’s middle class has been driven much more by private sector activities, both domestic and foreign.

Thirdly, the role of the middle class for development is not unambiguously positive. While the middle class implies a growing resource base for tax mobilization and investment, many middle class constituents tend to simultaneously invest in private goods, such as their own security guards, electricity generators, and private schools for their children. This often breaks the expected link between taxation and accountability for service delivery. Moreover, the middle class have often been at the forefront of protests for distortionary economic policies. A case in point is Nigeria’s middle class, which was highly active in protests in 2012 against the removal of the country’s expensive and inefficient fuel subsidies.

With regards to democracy and good governance, the existing political economy literature suggests mixed possibilities. On the one hand, modernization theories suggest that the middle class is likely to promote democracy and mitigate against extremism due to possessing distinct values shaped by their higher levels of education and residence in urban areas. Another theory suggests that the presence of a middle class means that the median voter is wealthier, thereby encouraging autocrats to engage in political liberalization because they no longer fear that universal suffrage will result in demands for excessive redistribution. On the other hand, the middle class has also been found in other regions of the world to be motivated by material interests and to be more likely to support democracy and engage in political processes only when it is in their best interest to do so but to passively support autocracy when they benefit from the status quo.

The UNU-WIDER project examines this issue from multiple perspectives and finds that country context plays a significant role. In Kenya, the middle class is more likely to have a partisan affiliation than their poorer counterparts, and partisanship has long been shown to be a key impetus for high levels of political participation. In Nigeria, the middle class was also highly involved in protests against Olusegun Obasanjo’s attempts to change the constitution in 2007 to run for a third term. In Zambia, the middle class demonstrates a lack of trust of existing political institutions, which suggests that this constituency can be a beacon of vigilance over institutions such as the presidency, parliament, judiciary, electoral commission and police. However, in both Zambia and South Africa, the middle class is associated with a lower likelihood to vote in elections. In contrast to Kenya, South Africa’s black middle class also demonstrates low levels of partisanship, including to the ruling African National Congress.

Importantly, these political and economic outcomes very much depend on how the middle class is conceptualized and measured in the first place. In other words, income-based approaches equivalent to the AfDB’s show more significant correlations with some of these outcomes, such as political trust, than measures that incorporate employment status and educational background, which tend to influence political participation. 

Fuller versions of the project papers will be available on the UNU-WIDER website by mid-2014.

[1] This is based on the AfDB’s (2011) ‘lower-middle class’ and ‘upper-middle class’ designation. They also include a ‘floating’ middle class that consumes US$2-4 per day, but this category has been quite controversial since it suggests that those who live even slightly above the international poverty line of US$1 per day could be considered middle class. Therefore, the discussion here excludes that category.

WIDERAngle newsletter
November—December 2013
ISSN 1238-9544