Ember's analysis of Chinese customs data, published in late August 2025, provides the clearest evidence yet that solar is taking off across Africa. Imports of solar panels from China rose 60 per cent in the 12 months to June 2025, to 15,032 MW, from 9,379 MW in the previous 12 months. More striking is where the growth happened. Outside South Africa, imports nearly tripled over two years, from 3,734 MW in the year to June 2023 to 11,248 MW in the year to June 2025. Twenty countries set new import records, and 25 countries imported at least 100 MW, up from 15 a year earlier. Nigeria overtook Egypt to become the second-largest importer after South Africa, and Algeria rose to third after a 33-fold increase.
The Ember report is careful about what the data can and cannot show. Exports from China are not the same as installations in Africa. Some panels are re-exported, some sit in warehouses, and there is a lag of months between shipment and installation. But the scale and breadth of the increase leave little doubt that something structural is under way. Our view is that most of this solar is being installed by households, shops, offices and small factories, not by utilities, and that it will force a change in how African power systems are planned and regulated much sooner than governments expect.
Why this looks like distributed solar
Ember's initial analysis suggests that the growth is driven more by distributed solar than by utility-scale projects. Its review of satellite data found that most announced utility projects had not started construction, and TransitionZero's satellite tracking identified only one new African utility project outside South Africa in 2025 up to August. If panels are arriving in large quantities but large solar farms are not being built, the panels are going somewhere else: rooftops, small ground-mounted systems and commercial premises.
The economics explain why. Many African countries share the conditions that drove Pakistan's solar boom: unreliable grid supply, high and rising tariffs, expensive diesel and access to very cheap Chinese modules. Ember estimates that in Nigeria a 420 W panel retailing at about USD 60 would produce around 550 kWh a year, while the same USD 60 spent on diesel at USD 0.66 a litre would generate only around 275 kWh, implying a payback of roughly six months on the panel alone. Inverters, batteries and installation add cost, but even allowing for them, solar paired with modest storage beats a generator for many users within a couple of years. Wood Mackenzie research cited by Ember estimated that 17 African countries had more diesel generator capacity than on-grid power plant capacity, including 28 GW in Nigeria.
Not the next Pakistan, yet
Ember notes that Pakistan imported more solar panels in the year to June 2025 than the whole of Africa, despite having about a sixth of the population. Africa is therefore not yet at Pakistan's scale. But Pakistan's experience is a direct warning. In Pakistan, imports tripled in a year, and the rapid growth of rooftop and behind-the-meter solar caught the power sector off guard. Grid sales fell in the daytime, capacity payments to idle plants were spread over fewer units, and tariffs for remaining grid customers rose, which pushed more customers towards solar.
Several African utilities face a similar structural risk. Their best-paying customers, commercial and industrial users and wealthier urban households, are exactly the ones most able to install solar. If those customers reduce their grid purchases sharply, utilities lose their most reliable revenue while retaining the obligation to supply at night, during the rainy season and to customers who cannot afford solar. Where utilities are locked into take-or-pay contracts with thermal generators, the problem is worse, because the fixed costs remain while sales fall.
What governments should do now
The first priority is data. Ember's central message is that better tracking is urgently needed, and we agree. Most African countries have no monthly data on installed solar, and registration schemes, where they exist, capture only a fraction of systems. Pakistan imported about 16 GW of panels in 2024, but only about 1.2 GW was registered under net metering. Utilities planning generation and network investments without knowing how much solar sits behind the meter will make expensive mistakes, building capacity for demand that has already been met by rooftops.
The second priority is tariff design. Utilities should move towards tariffs that recover fixed network costs more fairly, without punishing solar adopters so harshly that they leave the grid altogether. That means careful use of fixed charges, time-of-use tariffs that reward evening supply, and fair compensation for exported power. Abrupt changes, such as sudden cuts to net-metering rates, tend to push customers off the grid entirely, which is the worst outcome for the utility.
The third priority is storage. Solar is most valuable when it is paired with batteries that shift supply into the evening. Encouraging batteries at commercial sites and in mini-grids, and procuring utility-scale storage, would turn the solar boom into a resource for the system rather than a threat to it.
The upside is large
It would be wrong to frame this purely as a problem. Ember points out that the value of refined petroleum imports exceeds the value of solar panel imports by a factor of 30 to 107 in nine of the ten largest importing countries. Every panel that replaces diesel generation reduces fuel imports and foreign exchange pressure. In several smaller countries, the panels imported in a single year could add electricity equivalent to more than 10 per cent of reported generation, and in Sierra Leone around 61 per cent, although Ember cautions that official generation data often omits off-grid supply. For countries with chronic shortages, that is a substantial contribution to supply at almost no cost to the public budget.
Our assessment
The surge in solar imports is the most important energy development in Africa that most policymakers are not yet tracking. It is being driven by consumers and businesses making rational decisions in response to unreliable grids and expensive diesel. Governments cannot stop it and should not try. They should measure it, adapt tariffs and planning to it, and encourage storage so that distributed solar strengthens the grid. The lesson from Pakistan is that waiting for the data to become undeniable means reacting after the utility's finances have already been hit.