Missing lessons for SA’s digital public infrastructure push from a decade of smart city efforts

Infrastructure has been the metaphorical backbone of society since the beginning of state formation. The Roman Empire was famed for its road building that enabled the far-ranging transport of people and materials. Ever since, the question of how to circulate not just people and materials but more recently electricity and now digital data has been the challenge of infrastructure. 

In Africa, the Economic Commission for Africa reports that infrastructural deficits on the continent reduce economic growth by 2% and productivity by 40% annually. In 2021, the African Union’s Programme for Infrastructure Development in Africa adopted a $161-billion project to improve infrastructure across the continent. Reliable and effective infrastructure is foundational to societal wellbeing – but exactly how to achieve that cost effectively is vigorously debated.

At the same time, the question of how to design cities to best serve their residents and their economy is just as old. Indeed, the old saying goes “all roads lead to Rome”. Cities are, from one perspective, immensely dense infrastructure; and anyone who has ever dealt with potholes, water leaks, blackouts or the sometimes wonderful, sometimes infuriating municipal bureaucracy knows exactly how central infrastructure is to urban wellbeing. As long as there have been ideas for how to design national or regional infrastructure there have been ideas about the design of cities too.

This year, as the rotating head of the G20, South Africa under Cyril Ramaphosa has continued to emphasise a new approach to infrastructure called digital public infrastructure (DPI). Championed under the Indian presidency of the G20 in 2023, DPI envisions a new layer of infrastructure, a digital one, intended to bring infrastructure into the age of data, online services and digital tools from smartphones to apps to artificial intelligence. 

Under the South African presidency, the goal for the G20 on DPI is to “harness the transformative power of digital technologies to bring about inclusive development for all”. In his State of the Nation Address, Ramaphosa reiterated this goal, promising to “invest in digital public infrastructure to give South Africans access to government services anytime, anywhere” and implement a “digital identity system”. Digital identity, digital payments, accessible digital services and secure digital data transfers are all envisioned parts of DPI.  

However, this current vision of DPI is based directly on the Indian Aadhaar model. While this model has been successful in India in bringing secure access to government services and digital payments to millions of previously marginalised people (although it has received its share of criticism too), this model cannot be so easily imported to either South Africa or Africa as a whole. 

As researchers like Diana Sang, Jane Munga and Nanjira Sambuli, as well as organisations like Research ICT Africa, have argued, the context in Africa is extremely heterogenous. Not only is it not one state but 54, there are also numerous digital projects around ID and payments already in place. In addition, the continent is home to thousands of different languages and cultures, as well as being marked by pressing digital inequality. 

If DPI is to support progress towards the Sustainable Development Goals (SDGs), as is often claimed, then locally adapted solutions are going to be necessary.

Yet DPI is not the first grand idea of the digital technology-driven developmental strategy. In the late 2000s, kickstarted by an IBM PR campaign, the concept of smart cities swept the world. It too drew on a vision of digital services, data-based technologies, algorithmic decision-making and the construction of digital infrastructure in partnership with the private sector – albeit on the scale of cities, not states. 

What began in the wealthiest cities quickly moved and became seen as a way to improve developmental standards throughout the global majority. More than a decade later, numerous smart city projects have been developed in Africa, including in Lagos, Nairobi, Kigali, Johannesburg, Casablanca and Accra and initiatives like the Smart Africa Alliance advocate strongly for smart city plans as well as, recently, DPI projects.

These years of smart city implementation have also produced extensive research on what works and what doesn’t. These lessons are frequently directly applicable to DPI, and it is important that they are learnt to avoid the same pitfalls and help ensure it can achieve its SDGs. 

Repeatedly, researchers have found that imported theories and ideas for the design and function of smart cities do not work well in Africa. Attempts to build smart cities from the ground up with these ideas are disconnected from the local realities and fail to solve urban challenges, particularly around inequality, inclusivity and environmental sustainability. Such projects frequently push the vision and objectives of donors, foreign technology corporations and developers whose business and political incentives are not aligned with those of the people on the ground. 

As a result, they ignore the numerous existing digital projects that already function within the communities, and they do not recognise the importance of local, traditional and informal authorities. Research shows that while smart city projects overlook local, bottom-up digital or civic tech projects, it is these projects that are the most socially sustainable and beneficial. Further, research shows that when local needs, challenges, contexts, cultures and projects are included and engaged in digital projects, the benefit to the people and communities in terms of skills, access and quality of life is considerable.

DPI has the potential to fall into the same traps of extractive corporate neocolonialism, digital exclusion and the exacerbation of inequality, and the failure to meaningfully achieve SDGs. However, much has changed since the beginning of the smart city trend. Across the continent many more countries have robust data protection laws, there are significant regional and continental digital strategies like the Malabo Convention, the Digital Transformation Strategy for Africa, the African Union Data Policy Framework, the AFCFTA Digital Trade Protocol, and more. All of which contribute to a considerably stronger governance environment. 

However, it will still be crucial that the lessons from smart city projects are learnt, and local communities, authorities and existing digital projects are actively included in the design and implementation of DPI projects. As much as DPI is intended to be a standardised and interconnecting infrastructure across people and places, it cannot do so if it does not take the uniqueness of those peoples and places into account. DM

Dr Thomas Linder works with several African NGOs on data justice and digital equity issues across the continent. He lives in Cape Town.