Over the December holidays, local and foreign tourists to the Western Cape coastline could be forgiven for thinking South Africa’s economy is unmistakably on the up-and-up. In the New Year, as everyone returns to work (on average more rotund), the grim economic realities in the interior start to emerge.
How well did South Africa’s economy actually do, on New Year’s Eve? Unfortunately, economic activity estimates such as change in GDP arrive several months after the fact, which is at odds with the ever-greater push for in-time information and services.
How good could a forecast of the next six months be if the actual input arrives when the period is halfway done? This has seen the development of a range of approaches to “nowcasting”: forecasting the present.
While most of these in the public mind are linked to high-frequency financial data (think forex rates or stock prices), there are other sources in the real economy even more compelling.
In South Africa, as in much of the world, light emission is an excellent way to illuminate economic activity. Our research shows that economic activity in this country is highly concentrated in affluent urban areas, while large populations in informal or rural settlements generate limited night light emissions.
Though the suspension of load-shedding in 2025 caused a recovery in night light in many areas, night lights have increased less than 5% over the past decade. In fast-growing economies such as China, night lights increased more than 6% per year over much of this period.
Despite an improvement in efficient energy use in South Africa, our research raises questions about the depth of the structural slowdown and the ability of traditional indicators to fully capture shifts in spatial economic activity.
Provincial GDP data is published with a nine-month lag and is only available annually. No regular sub-provincial GDP indicators are available from official sources. Alternative data sources are therefore crucial to understanding real-time economic developments.
Delayed indicators
Though South Africa compares favourably to wealthy countries and other emerging markets in terms of the volume of economic and social indicators available, World Bank data shows that South Africa’s indicators are available on a less timely basis than for many major economies.
This makes it difficult for firms to identify economic opportunities and threats. The reality of light emission is unmistakable, available timely and highly granular. Our estimates show the economic activity of specific suburbs as it happens.
Another notable implication of our South African estimates is that there has been little growth in the intensity of night lights in South Africa over more than a decade. Over the same period, night light intensity has risen greatly in fast-growing emerging markets due to rising incomes and infrastructure development.
Despite population growth, ongoing urbanisation and growth of informal settlements in South Africa, night light intensity has shown a far lower increase than in other economies. In fact, the low increase in night light intensity in South Africa reflects declines in intensity in many urban and industrial areas.
Night light intensity has fallen greatly in most major special economic zones, which are designated areas offering tax and industrial incentives to promote industrialisation. Even at the high end, mean night lights fell in Sandton Central or Saxonwold in Johannesburg and Menlyn, Pretoria, between 2012 and late 2024.
South Africa stands out for the decoupling of night lights from economic and population growth.
This probably reflects the effects of municipal infrastructural degradation, including the breakdown of a large proportion of streetlights, as well as a shift to solar and off‐grid electricity supply. Another important reason is the decline in industrial and manufacturing production.
There has possibly been some shift to more efficient lighting technologies, though this does not explain why night light has increased so much faster in growing developing countries.
South Africa stands out for the decoupling of night lights from economic and population growth. In 2012-24, night light intensity rose at an annual average rate of only 0.4%. Real GDP rose at an average of just more than double that rate, while population growth grew at more than three times the rate.
Light intensity does show migration patterns clearly: middle- and upper-class migration from Gauteng to coastal regions and the Western Cape is visible from space.
The Harvard Growth Lab report from 2023 shows how South Africa’s urban sprawl has been incentivised by government urban planning regulations and zoning policies. The report documents how this raises transportation costs and survival wages, discriminating against job creation and efficient service provision.
Satellite data therefore sheds light on the implications of the collapse of state service delivery and South Africa’s decline in GDP per capita over the past decade. A streetlight is the quintessential non-rivalrous non-excludable public good: everyone benefits from it and you cannot use it up.
Satellite imagery is striking in its granularity and immediacy, providing a view of how well (or badly) municipalities are doing and where the money is really going.
The data highlights the widespread collapse in public service delivery, the decline in industrial and manufacturing production and persistent low levels of urban densification compared with other fast-growing emerging markets.
Once you take off the rose-tinted glasses of December, the economic reality in the interior is far darker than the fairy lights along the coast would make you believe.
• Dr Steenkamp is CEO of Codera Analytics and a research fellow with the economics department at Stellenbosch University. Roos is an associate with Codera. Codera’s report ‘Illuminating Economic Activity: Evidence from Night Lights Data in South Africa’ is available here.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.