SolepowerGet a quote

    The Infrastructure of Abundance

    Why Nigeria should, can and probably will run on the sun

    Solar power vs oil and gas, from first principles

    Introduction

    The argument about Nigeria's energy future usually runs along familiar lines. On one side are oil and gas: the industries that built the modern Nigerian state, that still earn most of its foreign exchange, and that Africa, their defenders say, has every right to use the way every rich country did. Gas, in this view, is the bridge, and the Federal Government declared 2021–2030 the "Decade of Gas". On the other side are renewables, argued for mostly in terms of what fossil fuels cost the world: emissions, spills in the Delta, and a changing climate that Nigeria did little to cause and will suffer from anyway.

    Both sides have largely been arguing from the numbers they had a few years ago. I think those numbers have moved so far, so fast, that much of the debate now describes a country that no longer exists.

    Consider what an analyst sizing solar against generators would have used in early 2023. Petrol was subsidised at around ₦185 a litre. Diesel, after the 2022 spike, sat at around ₦800. The naira traded at around ₦460 to the dollar. On those numbers, a petrol generator made electricity for roughly ₦100 a kWh and a diesel generator for around ₦250, and solar with batteries, priced in dollars, struggled to compete with either.

    Today petrol costs ₦1,400 or more, diesel ₦1,800–2,000, and the naira trades at around ₦1,330 to the dollar. A generator kWh costs ₦630–900. Over the same three years the hardware went the other way. Panel prices roughly halved in dollars and battery prices fell by more than half, while panels became more efficient and lithium batteries replaced the lead-acid ones that used to die within two years. A model built in 2023 is now wrong by a factor of several, and every bit of the error is in solar's favour.

    This essay is an attempt to redo the sizing from first principles, with today's numbers, and to show all of the working. Every figure is worked through in the text and sourced at the end. Every naira amount shows its dollar and pound equivalent. The household cases are run through the hourly simulator we use to size systems for customers. If a number is wrong, it should be easy to see where, and I'd like to be told. The aim is to make the case for solar in Nigeria on what it costs and what it does, measured honestly against what Nigerians actually use today.

    Why Nigeria's dysfunction makes it the ideal place

    It might seem strange to argue that a country with one of the least reliable grids on Earth is the best place to build a solar economy. I think the dysfunction is exactly what makes it so.

    In most countries, solar has to compete with a grid: power stations built decades ago, largely paid for, selling electricity at a fraction of what a Nigerian generator costs to run. Nigeria's grid reaches only some people, only some of the time, so the competitor solar actually faces here is the generator, running on diesel and petrol that Nigerians buy at world prices. At ₦630–900 a kWh, that is one of the most expensive incumbents on Earth, and the gap between it and solar is wider here than almost anywhere (chapter 2).

    Three more advantages come from the same dysfunction.

    The first is that the incumbent keeps getting dearer. The end of the petrol subsidy, a floating naira and an oil price set in London and Riyadh have together multiplied the fuel cost of a generator kWh roughly eightfold since 2019 (chapter 2). Solar sits on a manufacturing learning curve, and every year it gets a little cheaper and a little better (chapter 3).

    The second is that Nigerians already own their power stations. Tens of millions of households and businesses have bought a generator, a stabiliser or an inverter. They know what a kVA is and what a litre of diesel costs them each month. Switching to solar replaces a machine they already buy with a better one. It needs no new grid, no regulator's permission and no change in how people think about power.

    The third is that there is very little to protect. Countries with large, paid-off coal and gas fleets have utilities, unions and regulators with every reason to slow solar down. Nigeria's real power system, the generator fleet, has no lobby, and its owners would happily burn less fuel.

    Put together, these mean Nigeria can go straight to the cheapest electricity available, the way it went straight to mobile phones without ever building a landline network. The rest of this essay works out what that means: how big the market already is, how far the numbers have moved, how much it would take to build, and what a country could do with power that is cheap, and then nearly free.

    Chapter 1

    It's all sunlight anyway

    Walk down almost any street in Lagos after dark and you'll hear the country's real power system before you see it. It's a low, uneven hum coming from behind gates, under staircases and outside shops: millions of small engines, each burning fuel to make a few kilowatts for the people next to it.

    I want to start with a simple claim, because the rest of this essay depends on it. The diesel in those generators is sunlight.

    Hundreds of millions of years ago, plankton and algae in shallow seas captured sunlight through photosynthesis and turned it into carbon-rich tissue. When they died they sank, were buried, and over geological time heat and pressure cooked them into crude oil. When we pump that oil out of the Niger Delta, refine it into diesel, truck it to a filling station and pour it into a generator, we are releasing sunlight that fell on the Earth before the dinosaurs.

    Look at it that way and almost the whole energy system turns out to be sunlight, harvested in different ways.

    • Gas is the same buried organic matter as oil, taken deeper and hotter. Past a certain temperature, the long hydrocarbon molecules in oil crack apart into short ones, ending as methane. Much of Nigeria's gas sits alongside its oil for exactly this reason.
    • Coal is the land-based version: swamp forests that captured sunlight, died, and were buried and compressed before they could rot.
    • Wind is sunlight: the sun heats the Earth unevenly, and air moves to even it out.
    • Hydropower is sunlight: the sun evaporates water, it falls as rain on high ground, and we catch it on its way back down at Kainji, Jebba and Shiroro.
    • Firewood and charcoal are sunlight stored for a few years rather than a few hundred million.
    • Solar photovoltaics (PV) are the most direct harvest of all, with no plants, no burial and no burning in between. A solar panel is a sheet of silicon cells. When sunlight strikes a cell, its energy knocks electrons loose, and the way the cell is built pushes them out along a wire as electric current. There are no moving parts, and a panel keeps working for 25 years or more. Today's mass-market panels turn nearly a quarter of the sunlight that hits them into electricity; a plant capturing sunlight by photosynthesis, the first step on the long road to oil, manages about 1%.

    There are only three real exceptions. Geothermal energy is heat left over from the Earth's formation, topped up by radioactive decay deep underground. Tidal energy comes from the moon's gravity pulling on the oceans. Nuclear energy is released by splitting atoms like uranium.

    A useful test is to ask what would happen if the sun switched off. Within weeks, the wind would die as temperature differences evened out, the rain would stop, and plants would stop growing. We would still have whatever oil and gas was already underground, but that's a savings account with no more deposits. Geothermal, tides and nuclear would carry on. And even those exceptions trace back to stars: uranium was forged in exploding stars long before our sun formed. Nearly all the energy we use is starlight. Most of it is our own star's.

    So the real question, behind what most people would call "solar and renewables versus oil and gas," is which way of harvesting sunlight, at what cost, with how much complication, and with what side effects.

    On those three measures, the old answer was clear. Stored sunlight (oil and gas) won. Oil is extraordinarily dense: a litre of diesel holds about 10 kWh (kilowatt-hours) of energy, and you can carry it in a jerrycan, keep it for months and burn it whenever you like. Fresh sunlight is the opposite. It's dilute, it only arrives during the day, and for most of the twentieth century turning it into electricity was absurdly expensive. Solar panels powered satellites, calculators and little else.

    But stored sunlight has always come with a long and fragile chain. Somebody has to find the oil, drill it, ship it, refine it, import it (for most of Nigeria's history, we exported crude and imported the refined product back), store it, truck it across bad roads, sell it at a pump, and carry it home. At the end of all that, a small generator turns perhaps a fifth to a third of the fuel's energy into electricity and throws the rest away as heat, noise and fumes. The side effects are real: carbon monoxide in poorly ventilated rooms, tanker explosions, fuel queues, smog, and an economy whose cost of production rises and falls with a price set in London and Riyadh.

    Here's how the main options compare for electricity in Nigeria today. Each cost figure is worked through later in the essay.

    SourceCost per kWh (Oct 2026)ComplicationExternalities
    Petrol generator₦700–900+, fuel onlyLong fuel supply chain, frequent servicingFumes, noise, emissions, FX (foreign exchange) drain
    Diesel generator~₦630 fuel only; ~₦720 with maintenance and wearLong fuel supply chain, servicing every few hundred hoursFumes, noise, emissions, FX drain
    Gas-fired gridCheap to generate at the plant; Band A tariff ₦209.50–225Pipelines, vandalism, transmission, distribution, collectionEmissions, unreliability
    Large hydroVery low once builtSiting, rainfall, long build timesRiver and land disruption
    Solar + battery~₦70 over its life unfinanced; ~₦160–240 financed at 25%Installation, then light but real maintenanceManufacturing footprint, end-of-life waste

    (Firewood and charcoal belong in the family tree but not the table: they cook food rather than make electricity.)

    What changed is that the last row stopped being a niche.

    Two things happened. First, solar panels became manufactured goods produced at enormous scale, and manufactured goods follow learning curves: every time cumulative production doubles, the cost falls by a roughly fixed percentage. This is known as Wright's law. For solar modules that has been about 20% per doubling for decades, a pattern known as Swanson's law. Oil gets no such curve: it's extracted, and the cheapest barrels get pumped first. Over time, the thing you make gets cheaper and the thing you dig up doesn't.

    Second, batteries arrived. The great weakness of fresh sunlight was that it couldn't be stored, so it couldn't compete with a fuel you could burn at 9pm. LFP (lithium iron phosphate) batteries, pushed down their own learning curve by electric vehicles, fixed that. A panel plus a battery is, functionally, a machine that stores sunlight for hours instead of millions of years.

    This is the pattern Clayton Christensen described in The Innovator's Dilemma. The new technology starts out worse than the incumbent on the measures that matter to mainstream customers, so it serves the fringes: satellites, rural lanterns, telecom masts in the bush. The incumbents ignore it. It keeps improving at a pace the incumbent can't match, and one day it's simply better for most people. In Nigeria, I think that day has already come for anyone paying for diesel or petrol to make electricity.

    Nigeria sits in one of the sunniest belts on Earth. For fifty years we have sold our stored sunlight abroad and bought it back refined, while the fresh stuff fell on our roofs for free.

    Chart 1.1Interactive chart in progress
    the sunlight family tree, from sun to plankton, oil, gas, diesel, generator and kWh, versus sun to panel, battery and kWh, with the efficiency at each step; the three exceptions shown off to the side

    Chapter 2

    The generator nation

    Before getting to generators, it helps to see the whole system at once. Nigeria moves and powers itself on three things: petrol, diesel and the grid. Here is roughly how much of each it uses in a year, at 2026 rates, and what that comes to.

    Nigeria, a year at 2026 ratesHow muchEnergyWhat it mostly doesSpend
    Petrol~17–18bn litres~155 TWhMoves people (cars, okadas, kekes, buses); about a quarter runs small generators~₦23–25tn
    Diesel~5.5bn litres~55 TWhRuns generators; the rest moves heavy goods~₦11tn
    Grid electricity~35–40 TWh sent out~35–40 TWhPowers homes and businesses, when it's onroughly ₦3–5tn billed
    Total~245–250 TWh~₦37–41tn

    The fuel figures are the energy contained in the fuel, at about 8.9 kWh per litre of petrol and 10 kWh per litre of diesel. Engines turn only part of it into motion or electricity: a generator manages a fifth to a third.

    So Nigeria runs on around 250 TWh of energy a year, more than four-fifths of it liquid fuel, and spends close to ₦40 trillion on it, about ₦35 trillion of that on two fuels. Roughly a third of those litres, about 7.8 billion a year, is burned in generators to make electricity: around 18–19 TWh, against the 35–40 TWh the grid sends out, a large share of which is lost before it's billed. Once those losses are counted, generators supply somewhere between a third and two-fifths of the electricity Nigerians use, and their fuel alone costs about ₦13 trillion a year.

    In capacity terms, the gap is even more striking: Wood Mackenzie research from 2022 estimated Nigeria had 28 GW of diesel generator capacity, one of 17 countries where generator capacity exceeded on-grid power plant capacity. What Nigeria actually has is a vast, private, fuel-burning power system with a grid attached.

    That 19 TWh is also less than the electricity Nigerians actually want. It's what they can afford to generate. Households and businesses ration generator hours: on in the evening, or during working hours, and the rest of the time they go without. Rationing is already baked into the litres; it means the true demand is larger, a point I come back to in chapter 8.

    The rest of this chapter builds those numbers from the bottom up. Each step has its working folded underneath it, to open or skip.

    Where the litres go

    Why split the litres at all? Electrifying transport is a big story, and chapter 6 comes back to it, but this essay is about electricity: what Nigerians pay to power homes, shops and factories, and what solar can replace today. In most countries that question involves almost no petrol, because petrol moves cars. Nigeria is unusual. A large share of its petrol, and most of its diesel, is burned to make electricity in millions of generators. That share is the market a rooftop panel competes for today, so it's the number this chapter needs to get right.

    My estimate is that about 60% of diesel and a quarter of petrol go into generators: 3.3 billion and 4.5 billion litres a year, the 7.8 billion litres behind Nigeria's self-generated electricity. The diesel share matches the NNPC's own estimate. For petrol, the evidence puts the share somewhere between a fifth and a third, and a quarter sits towards the cautious end.

    The volumes themselves are my estimate from the data available: the regulator's monthly figures, the Dangote refinery's, and the import record from before the Dangote refinery opened. Together they point to about 45–50 million litres of petrol and 15 million litres of diesel a day, a fuel bill of roughly ₦35 trillion a year, of which generators burn about ₦13 trillion. The Dangote refinery's own sales figures suggest more, up to about ₦52 trillion; the working below shows each source and why I settled where I did. The one place I've deliberately leaned low is the share of petrol going into generators, so if anything the generator figures here are an underestimate.

    The workingHow much petrol and diesel Nigeria uses

    Start with total litres. The NMDPRA (Nigerian Midstream and Downstream Petroleum Regulatory Authority) publishes monthly figures. Its 2026 daily benchmarks are 50 million litres per day for petrol and 14 million for diesel, and actual truck-outs move around those numbers: in December 2025, petrol consumption was 63.7 million litres a day and diesel 16.4 million; in March 2026, diesel consumption fell from 20.3 to 14.5 million litres a day; and in August 2026, petrol truck-outs dropped to 41.5 million litres a day. Call it roughly 45–50 million litres of petrol and 15 million litres of diesel a day. Over a year, that's about 17–18 billion litres of petrol and 5.5 billion litres of diesel.

    At late-2026 pump prices of around ₦1,400 a litre for petrol and ₦2,000 for diesel, Nigerians spend about ₦23–25 trillion a year on petrol and ₦11 trillion on diesel: roughly ₦35 trillion, or about $26 billion, on two liquid fuels. That's somewhere around 8–10% of GDP.

    Checking the numbers from the other end. Truck-out data measures what leaves depots. We can cross-check it against what comes in: Dangote's domestic deliveries plus imports. August 2026 is the cleanest recent month (NMDPRA factsheet, imports, Dangote split):

    August 2026Petrol (million litres/day)Diesel (million litres/day)
    Dangote production41.918.0
    Dangote to domestic market35.912.4
    Dangote exports9.78.8
    Imports14.61.3
    Total receipts (supply side)50.514.5
    Truck-outs (demand side)41.5~13

    Supply runs ahead of truck-outs in August because stocks were being rebuilt. Over the year, the two methods agree within 10–20%: petrol has ranged from about 41 to 60 million litres a day in 2026, and diesel from about 12 to 20 million. My working figures sit in the middle of both.

    One discrepancy is worth flagging. Dangote's own H1 2026 report puts its domestic supply at around 50 million litres of petrol and some 25 million litres of diesel a day. The regulator's monthly figures are lower: in July, for example, NMDPRA recorded 25.8 million litres a day of Dangote petrol going to the domestic market. I use the regulator's numbers throughout, because they're published monthly and line up with truck-out data.

    What each source implies, all at ₦1,400 a litre for petrol and ₦2,000 for diesel, with a quarter of petrol and 60% of diesel going to generators:

    SourcePetrol (million litres/day)Diesel (million litres/day)Fuel spend a yearOf which generatorsGenerator electricity
    NMDPRA truck-outs, August 2026 (lowest recent month)41.5~13~₦31tn~₦11tn~16 TWh
    NMDPRA 2026 benchmarks5014~₦36tn~₦12.5tn~18 TWh
    Supply side, August 202650.514.5~₦36tn~₦13tn~18.5 TWh
    Working figures in this essay45–5015~₦34–37tn~₦12–13tn~18–19 TWh
    NMDPRA truck-outs, December 2025 (seasonal high)63.716.4~₦45tn~₦15tn~22 TWh
    Dangote's own H1 2026 figures, plus imports~65~26~₦52tn~₦20tn~29 TWh

    At ₦1,800 a litre for diesel, the low end of 2026, every row falls by about ₦1 trillion.

    Why the gap? Truck-outs count what leaves depots by road, month by month, so they swing with stock-building and may miss fuel delivered straight to large users. Dangote's figures count what it sells, and some of that may sit in stock before anyone burns it. The regulator's numbers are the ones I can check against each other every month, which is why I use them. If Dangote's turn out to be closer to the truth, Nigerians burn about half as much fuel again in generators as this essay assumes, and every argument that follows gets stronger.

    Chart 2.2Interactive chart in progress
    interactive, monthly petrol and diesel supply (Dangote domestic, imports, modular refineries) stacked against truck-outs, Jan 2025–Aug 2026, with toggles for naira, dollar and pound spend
    The workingWhat imports said before Dangote

    Before the Dangote refinery opened, Nigeria imported almost all of its petrol and most of its diesel, so the import record is close to a record of what the country burned.

    PeriodPetrol (million litres/day)Diesel imports (million litres/day)
    2019~57
    2021~61
    January–August 2022~68~11 (full year)
    First half of 2023, subsidy in place~66~15.6
    Second half of 2023, after the subsidy ended~45~11.5
    2026, this essay's working figures45–5015

    The record tells a consistent story. While petrol was subsidised, the country used 57–68 million litres a day. Within months of the subsidy ending in mid-2023, that fell to about 45 million, which is roughly where the regulator's figures have stayed since. Diesel ran at 11–16 million litres a day. The working figures in this essay match both: what Nigeria was importing and burning before the Dangote refinery, and what the regulator records now. The Dangote refinery's own figures would put diesel use at close to double its level before it opened, which is why I treat them as the top of the range.

    The workingWhat the fuel is used for

    Split by end use. Diesel does two jobs in Nigeria: it moves heavy things (trucks, buses, construction equipment, boats) and it makes electricity (generators at factories, banks, hospitals, hotels, estates, malls, telecom towers and wealthier homes). My view, from sitting with customers and watching where the tankers go, is that most of Nigeria's diesel ends up in generators rather than trucks. There's no official split, so I'll assume 60% goes to power generation and invite anyone with better data to correct me. That's about 3.3 billion litres a year. A 2015 NNPC study put the share at 50–60%, so 60% is at the top of that range.

    Petrol is mostly transport: cars, okadas, kekes and buses. But Nigeria is unusual in that a meaningful share of petrol is burned in small generators, the 1–3 kVA (kilovolt-ampere) machines that power a fridge, a fan and a TV in millions of households and small shops. I'll assume a quarter of petrol goes to generators: about 4.5 billion litres a year.

    That quarter is an estimate, and it matters, so here is the evidence behind it:

    EstimatePetrol in generatorsAt today's volumesBasis
    NNPC study, 2015~20%~3.5bn litres8 of 40 million litres a day went to power generation
    This essay~25%~4.5bn litresThe cautious middle of the evidence
    NBS Power Sector Data Preview, around 2021–22~25–35%~4.5–6bn litresPetrol generators supplied 22.6% of all electricity used, against 51.2% from the grid
    Counting household generatorsprobably higherAround a third of households own a generator, most of them petrol

    The NBS figure works backwards from electricity: if the grid delivers something like 25–28 TWh a year to customers, petrol generators were making about 11–12 TWh, which at 1.5–2 kWh a litre is 5.5–8 billion litres: a quarter to a third of the roughly 22 billion litres of petrol Nigeria used a year at the time. The household count works forwards. Surveys find about a third of Nigerian households own a generator, which is around 15 million machines. At half a litre an hour or so, 4.5 billion litres a year is only enough to run each of them for about an hour and a half a day, before counting the shops and workshops that run petrol sets too.

    The transport side pulls the other way. The remaining 13 billion litres, about 36 million litres a day, has to move every car, bus, okada and keke in the country: about 3 litres a day for each of the roughly 12 million vehicles on the road, before counting the millions of motorcycles and tricycles. Neither generators nor vehicles are counted properly in Nigeria, so the two checks don't settle it. They do bracket it: somewhere between a fifth and a third of petrol is burned for electricity. At a fifth, petrol generators make about 6 TWh a year and burn about ₦5 trillion of petrol; at a third, about 10.5 TWh and ₦8 trillion.

    Two more things the NBS survey shows. Petrol generators made more electricity than diesel ones, where this chapter has it the other way round; the survey predates the end of the petrol subsidy, when petrol cost about an eighth of what it does now, so the balance has probably shifted since. And larger users also self-generate on gas, about a tenth of all electricity used. This chapter counts only petrol and diesel, which leaves gas-fired self-generation out of the ₦13 trillion.

    What a generator kWh costs

    Fuel has become far more expensive in a short time. Diesel cost about ₦225–230 a litre in 2019 and crossed ₦800 during the 2022 price spike. Since the war with Iran shut the Strait of Hormuz in March 2026, it has swung between about ₦1,650 and ₦2,200 a litre; in October it's around ₦1,820. Petrol was held at around ₦145–185 by subsidy until May 2023; it's now ₦1,400 or more. Since 2019, the fuel cost of a generator kWh has risen roughly eightfold in naira.

    Chart 2.1

    What Nigerians pay for fuel, per litre

    DieselPetrolDiesel, Solepower estimate (Apr–Oct 2026)
    ₦0₦500₦1,000₦1,500₦2,000₦2,500’20’21’22’23’24’25’2612345678DieselPetrol
    1. 1Mar 2022 Russia invades Ukraine and diesel prices jump worldwide
    2. 2Jun 2023 The petrol subsidy ends and the naira is floated
    3. 3Feb 2024 The naira slides to around ₦1,500 to the dollar
    4. 4Apr 2024 The Band A grid tariff more than triples, from ₦66 to ₦225 a kWh
    5. 5Sep 2024 NNPC raises the petrol pump price
    6. 6Aug 2025 Dangote cuts diesel below ₦1,000 a litre at the depot
    7. 7Mar 2026 War with Iran shuts the Strait of Hormuz; oil passes $120 and diesel peaks near ₦2,200
    8. 8Sep 2026 Renewed Middle East crisis lifts diesel's import cost above ₦1,900 a litre

    The dashed diesel line, April–October 2026, is Solepower's estimate: each month's Dangote gantry price plus the ₦50–150 pumps have charged above it, following Brent. NBS's own April and May averages (₦2,475 and ₦3,277) are far above what depots charged, and it hasn't published since. Monthly averages hide the peaks: pumps briefly reached about ₦2,200. Petrol is around ₦1,400 in October.

    Sources: NBS Diesel and Petrol Price Watch, average price paid each month (2019–early 2021 worked back from NBS's year-on-year changes); Solepower estimates from Dangote gantry prices and Brent, April–October 2026; Band A tariff from Ikeja Electric and NERC orders (2021–22 steps approximate); exchange rates from the official investors' window (I&E, now NFEM).

    Show the numbers
    YearDiesel ₦/litrePetrol ₦/litreDiesel $/litreBand A ₦/kWh
    2019₦228₦145$0.63–
    2020₦224₦148$0.59₦52
    2021₦249₦166$0.62₦55
    2022₦665₦185$1.57₦62
    2023₦895₦469$1.48₦66
    2024₦1,381₦874$0.95₦174
    2025₦1,560₦1,101$1.03₦210
    2026 (NBS to May)₦1,735₦1,301$1.27₦210

    And people respond to price. In the first half of 2026, according to marketers' data, average petrol consumption fell from about 60 million litres a day in January to roughly 46 million in May, and diesel consumption dropped 17.5% over the same period. Some of that is people simply doing without. Some of it is people buying panels.

    Convert litres to kilowatt-hours. This is where the number most people quote, ₦630/kWh, turns out to be the optimistic case.

    ₦630/kWh is roughly what a well-matched generator achieves. The average Nigerian generator, oversized and lightly loaded, probably costs more. Then add what fuel cost leaves out: oil changes and servicing every few hundred hours, filters, overhauls, and the generator itself wearing out. On a typical commercial set, that adds something like ₦80–100 per kWh, taking the all-in cost to around ₦720/kWh. Small petrol generators are worse again: running at partial load, they produce perhaps 1.5–2 kWh per litre, which puts fuel alone at ₦700–900 per kWh.

    If the fleet runs less efficiently than assumed, the cost per kWh is higher and the electricity produced is lower. Either way, the ₦13 trillion is real money already being spent.

    The workingGenerator size, load and kWh per litre

    A diesel generator's efficiency depends heavily on how hard it's working. Run near its rated output, it might produce 3.2–3.4 kWh per litre. Run at a quarter of its rating, it produces far less, because a big engine burns a lot of fuel just turning over. And in Nigeria, generators are routinely oversized: a 60 kVA set bought for the occasional peak spends most nights powering a few air conditioners and lights at a fraction of its capacity.

    Generator loadkWh per litre (approx.)Fuel cost per kWh at ₦2,000/litre
    25%~2.5~₦800
    50%~3.0~₦670
    75%~3.2~₦625
    100%~3.4~₦590
    The workingFrom litres to 19 TWh

    Putting the national picture together, using the conservative per-litre figures:

    Litres/yearkWh per litreElectricityFuel spend
    Diesel generators~3.3bn~3.2~10.5 TWh (terawatt-hours)~₦6.6tn
    Petrol generators~4.5bn~1.8~8 TWh~₦6.3tn
    Total self-generation~7.8bn~18–19 TWh~₦13tn (~$10bn)

    The real incumbent

    When people say solar in Nigeria "competes with the grid," they usually compare it to a Band A tariff of around ₦210–225/kWh. That's the wrong comparison. Band A is the best grid supply in the country, it's only available to some customers, and even they run generators when it goes off. The real incumbent, the thing people are actually paying for at the margin, is self-generation at ₦630–900/kWh.

    How unusual is that? It helps to compare what solar is up against in other countries. Two things decide how valuable a solar panel is: the price of the power it replaces, and how much sun it gets.

    CountryWhat solar mostly replacesApprox. cost (US$/kWh)Annual yield per kWp (kWh)Value of 1 kWp of panels per year
    NigeriaDiesel/petrol generator$0.47–0.68~1,350~$650–900
    GermanyReliable, expensive grid~$0.40~950~$380
    South AfricaGrid, some diesel~$0.20~1,700~$340
    UKReliable, expensive grid~$0.33~900~$300
    PakistanExpensive grid~$0.20~1,500~$300
    IndiaCheap grid~$0.08–0.10~1,600~$140

    Germany and the UK have expensive grids but weak sun. India has strong sun but cheap grid power. Pakistan's boom happened at roughly a third to a half of Nigeria's value per panel. Nigeria is unusual in having both: some of the most expensive incumbent electricity on Earth and some of the strongest sunshine. A kilowatt of panels here is worth roughly two to three times what it's worth in Britain or Pakistan.

    That changes the economics completely. Nobody has to be persuaded that electricity is worth paying for. The millions of generators running every night are the proof.

    Pakistan is the parallel

    Pakistan in the early 2020s looked a lot like Nigeria today: a population of similar size, a strained grid, and a sudden supply of cheap Chinese panels. The trigger there was the grid price: grid electricity prices rose by roughly 155% between 2021 and 2024. The response was extraordinary. Pakistan imported 17 GW of panels in 2024 alone, double the previous year, and cumulative imports reached about 50 GW by August 2025, roughly equal to the country's entire grid capacity. Grid demand fell: a 10.4% reduction over a single year.

    Nigeria's version has started. In the 12 months to June 2025, Nigeria overtook Egypt to become Africa's second-largest panel importer, at 1,721 MW, and it imported 2.9 million panels worth more than ₦435 billion over 2025, with imports holding firm across all four quarters.

    Solepower research puts the growth at about 50% a year. China's customs records, compiled by Ember, show every panel China ships to Nigeria, and China supplies most of them. From those records we estimate that Nigeria imported about 0.2 GW of panels in 2019 and about 2 GW in 2025, doubling roughly every 21 months. 2026 is on course for about 3.5 GW, helped by a rush in March and April as the oil price spiked and China cut its export tax rebate. On these numbers Nigeria had imported around 6 GW of panels by the end of 2025, and will pass 9 GW during 2026. Not all of that is working on a roof: some sits in warehouses, some moves on to neighbouring countries, and some is small or poor-quality panels.

    Solar cells are arriving even faster. China shipped about 3.9 GW of cells to Nigeria in the first eight months of 2026, for panels assembled locally: far more than Nigerian factories can assemble, estimated at 300 MW to 600 MW a year. Some of those panels leave again. Nigeria exported ₦86 billion of panels in the first quarter of 2026, mostly to the United States. I leave cells out of the totals until it's clearer where they end up.

    Chart 2.3

    Solar panels Nigeria imports each year

    Panels China recorded shippingSolepower estimate: other sources, rest of 2026, scenarios
    02.5 GW5 GW7.5 GW10 GW12.5 GWScenarios0.2’19’20’21’22’23’242.0’253.5’26’27’28’297.0’30

    The estimate assumes China supplies 85% of Nigeria's panels by watts; NBS puts China's share of the value at about 75–86%, and panels from elsewhere usually cost more per watt. 2026 has 9 months of records; the rest of the year is estimated at the rate of the latest four months. From 2027 the bars are Solepower's central scenario, with the line showing the low and high cases. Imports are not installations: some panels sit in stock, some go on to neighbouring countries.

    Sources: Ember, China Solar PV Exports (Chinese customs data, to September 2026); NBS foreign trade statistics; Solepower research.

    Show the numbers
    YearChina recorded, panelsSolepower estimate, all panelsCells from China
    2019155 MW182 MW0 MW
    2020194 MW228 MW0 MW
    2021220 MW259 MW0 MW
    2022417 MW491 MW83 MW
    2023822 MW967 MW107 MW
    20241.3 GW1.5 GW187 MW
    20251.7 GW2.0 GW838 MW
    2026 (Jan–Sep recorded)2.2 GW3.5 GW6.7 GW
    2027 (scenario)–4.4 GW (3.2 GW–5.0 GW)–
    2028 (scenario)–5.3 GW (3.4 GW–6.7 GW)–
    2029 (scenario)–6.1 GW (3.5 GW–8.7 GW)–
    2030 (scenario)–7.0 GW (3.7 GW–10.9 GW)–

    Projected forward, our central case has imports growing more slowly, from 25% a year to 15%, as the customers who can pay cash are served and finance becomes the limit: about 4.4 GW in 2027 and 7 GW in 2030. If 2026's rush simply borrowed from 2027, imports could sit at 3–4 GW a year. At Pakistan's pace, they would pass 10 GW a year by 2030.

    So Pakistan built 50 GW at a fraction of the value per panel that Nigeria offers. The economic pull here is larger. What Nigeria lacks is the finance and installation capacity to respond at Pakistan's speed. That's the gap this essay is about.

    The workingHow we estimate solar imports

    The anchor. China's customs agency records every shipment of panels, cells and wafers by destination, and Ember converts the dollar values into megawatts using monthly prices. These are the panels China recorded shipping to Nigeria, before anything is estimated:

    YearPanels from ChinaCells from ChinaAverage panel price (free on board)
    2019155 MW–$0.27/W
    2020194 MW–$0.23/W
    2021220 MW–$0.27/W
    2022417 MW83 MW$0.28/W
    2023822 MW107 MW$0.18/W
    20241,300 MW187 MW$0.11/W
    20251,673 MW838 MW$0.10/W
    2026, January–August1,953 MW3,869 MW$0.11/W

    Checking it against Nigeria's own record. NBS valued Nigeria's 2025 panel imports at ₦435 billion, about $287 million at the year's average official rate of ₦1,518 to the dollar. Ember's China figures for 2025 come to $194 million, about 68% of that. NBS's own breakdown puts China's share at 71% for the first half of 2025. The two records describe the same trade.

    From China to every source. China's share of the value runs at about 70–80%. Panels from elsewhere usually cost more per watt, so China's share of the watts is higher. I assume 85%, so total panel imports are China's figure divided by 0.85: about 2 GW in 2025. For 2026, the last four months are estimated at the May–August rate, which leaves out the March–April rush.

    The scenarios. These are imports, not installations, in GW:

    Scenario2027202820292030Imported 2017–2030
    Low: 2026's rush borrowed from 2027, then 5% a year3.23.33.53.7~23 GW
    Central: growth slows from 25% to 15% a year4.45.36.17.0~32 GW
    High: Pakistan's pace, 40% slowing to 25%4.96.68.610.8~40 GW

    Chapter 3

    The crossover

    The case for solar in Nigeria rests on two price curves heading in opposite directions. One is fuel. The other is hardware.

    What hardware costs today

    Here are numbers from Lagos in late 2026, at four levels of scale. First, how each component's cost builds up from the factory gate in China to a Lagos warehouse:

    Step620W panel15 kWh LFP battery10 kVA hybrid inverter
    FOB (free on board) China~₦88,000 (~$66)~₦1.64m (~$1,235)~₦465,000 (~$350)
    CIF (cost, insurance and freight) Lagos~₦95,000~₦1.72m (shipped as dangerous goods)~₦480,000
    Plus duty, levies and VAT~₦105,000~₦1.80m~₦540,000
    Plus clearing, port charges, haulage~₦113,000~₦1.85m~₦560,000
    Landed in Lagos~₦113,000~₦1.85m~₦560,000
    Typical Lagos price₦125,000–150,000~₦1.9m₦0.8–1.0m mid-market; ~₦2.4m for a premium 12 kVA
    Markup over landed cost10–35%~3%40–80% mid-market; nearly 4x premium

    The three components behave very differently. Batteries already sell close to landed cost, so there's little to gain from importing them at small scale. Panels carry a modest markup. Inverters carry the most: a mid-market unit sells for up to nearly double its landed cost, and a premium brand for close to four times. Freight, clearing and haulage are also partly fixed, so an importer bringing in many containers spreads them more thinly and negotiates lower factory prices.

    For a standard 10 kWp (kilowatt-peak), 10–12 kVA, 15 kWh system for a large home or small business, that gives four price levels:

    ComponentRetail buyerPanel importer (one container)Importer of all componentsScaled aggregator
    16 × 620W panels₦2.0–2.4m₦1.81m₦1.81m~₦1.65m
    Inverter₦1.0m (mid-market 10 kVA) to ₦2.4m (premium 12 kVA)₦1.0m~₦0.56m~₦0.5m
    15 kWh LFP battery₦1.9m₦1.9m~₦1.85m~₦1.7m
    Mounting, cabling, protection, labour~₦0.8m~₦0.8m~₦0.8m~₦0.6m
    Total~₦5.7–7.5m~₦5.5m~₦5.0m~₦4.45m

    Competitor quotes I've seen for this spec sit at ₦6–6.9 million. In the rest of this essay I use ₦7 million as the conservative retail case and ₦4.5 million as the scale case: scale takes more than a third off.

    These are global commodity prices that a decade ago would have been unthinkable. Module prices fell roughly 90% over the 2010s. Battery pack prices fell by a similar amount. In Nigeria, Ember estimates a 420W panel retails for about $60 and produces around 550 kWh a year.

    Chart 3.1Interactive chart in progress
    your series, panel ₦/W, inverter ₦/kVA and battery ₦/kWh 2015–2026, in naira, dollars and pounds, plotted against diesel ₦/litre

    The naira trap. A common objection is that devaluation kills solar, because panels and batteries are priced in dollars. But diesel and petrol are also priced in dollars. Nigeria imports refined product or refines crude that trades at world prices. Devaluation pushes up both sides of the comparison at roughly the same rate. What actually decides the race is what happens in dollar terms, and there the story is one-sided: since 2019, diesel at the Nigerian pump has roughly doubled in dollars, while panels have roughly halved. The ratio between them has moved by a factor of four or five. Subsidy removal in 2023 then pulled petrol into the same trend.

    What a kilowatt-hour of solar costs

    In Lagos, 10 kWp produces around 13,000–14,000 kWh a year; in the north, more. Not all of it gets used, because loads don't perfectly match the sun. A conservative planning figure is 8,000 kWh of useful electricity a year; well-matched systems at heavy-use sites deliver 9,000 or more.

    • Over a 15-year life, including maintenance and component replacement: roughly ₦8.5 million for 120,000 kWh, or about ₦70/kWh unfinanced
    • Financed at a 25% cost of capital: about ₦240/kWh at retail prices, or about ₦160/kWh at scale

    Even the expensive, financed case is a third of what a generator costs.

    Payback depends on the grid, and the grid is worse than its labels

    The honest answer to "how fast does solar pay back?" is: it depends on how long the generator would otherwise run, and that depends on the grid.

    On paper, Band A means at least 20 hours of supply a day and Band B at least 16. In practice, Band A availability swings between roughly 40% and 80% of hours from month to month, and Band B often sees about a third. Real generator hours are far higher than the labels suggest.

    "Off-grid" doesn't mean the generator runs all day either. Most households and businesses ration it: on in the evening, or during working hours, and the rest of the time they go without.

    So I ran the same ₦7 million system through an hourly simulation covering a year of Lagos sunshine, a typical load of about 33 kWh a day peaking in the evening, and grid outages that vary by band and by month. When the grid is off, the generator covers outages from 6am to midnight. The battery holds a reserve for outages, so solar targets the hours the generator would otherwise run. Generator power is valued at ₦720/kWh all-in, Band A grid at ₦210 and Band B at about ₦65.

    CustomerGrid on (share of hours)Generator kWh/yr, before → afterNew kWh served that used to go withoutAnnual savingPayback: retail ₦7m / scale ₦4.5m
    Band A, good month~80%2,450 → 120~300~₦3.0m2.3 / 1.5 years
    Band A, typical~60%4,200 → 240~400~₦3.9m1.8 / 1.2 years
    Band A, bad month~40%6,300 → 490~500~₦4.9m1.4 / 0.9 years
    Band B~33%7,100 → 560~450~₦4.9m1.4 / 0.9 years
    No grid, generator 6pm–midnight04,300 → 380~5,500~₦2.8m2.5 / 1.6 years
    No grid, generator 8am–6pm (business)04,800 → 20~4,600~₦3.5m2.0 / 1.3 years
    No grid, generator 6am–midnight010,000 → 890~300~₦6.5m1.1 / 0.7 years

    Three things stand out.

    • Even Band A pays back in under two and a half years. Its reliability is mostly on paper, and the battery covers the hours the grid fails, which is exactly when the generator was costing most.
    • For households that ration their generator, solar brings more power as well as savings. A home running its generator only in the evening saves around ₦2.8 million a year, but it also gets roughly 5,500 kWh a year of electricity it previously went without. That's more than it used to buy from the generator. The pitch to that household is "spend the same, get twice the electricity."
    • About a quarter to a third of the panels' output goes unused. It's produced at midday when the battery is already full. Today that's wasted. In a networked future, it's power a neighbour could buy (chapter 8).

    These paybacks are faster than Ember's estimate of 18–36 months for full systems for two reasons. The model values generator power at ₦720 including maintenance, and it assumes the system is well matched to the load. A badly sized system will do worse, which is why design matters. The input that matters most, and that nobody publishes properly, is real grid availability by band, feeder and month. That's data I want to collect and publish.

    How much battery?

    The battery is what turns solar into a generator replacement. Here is the same 10 kWp array with different battery sizes:

    BatteryPayback, Band A (~60% grid)Payback, no grid with generator 6pm–midnight
    None2.3 yearsNever: the panels produce at midday, when the generator was already off
    5 kWh2.0 years5.8 years
    10 kWh1.8 years3.1 years
    15 kWh1.8 years2.5 years
    20 kWh1.9 years2.6 years

    At a heavy-use site, panels without a battery deliver about 5,200 useful kWh a year; with 15 kWh of battery, about 9,400. The battery nearly doubles what the panels are worth. The sweet spot is about 1.5–2 kWh of battery per kWp of panels. Beyond that, each extra kWh of battery covers rarer and rarer outages, and payback starts to lengthen.

    Chart 3.3Interactive chart in progress
    interactive battery sweep, with sliders for grid hours, generator window and battery price

    When did the lines cross?

    For anyone running a generator, I'd put it around 2022–23: the diesel spike and the end of the petrol subsidy, landing just as module prices hit record lows. Before then, solar made sense for the patient and the well-capitalised. After, it was the cheapest power you could buy.

    So why isn't everyone on solar already? Because a system that's cheap over its lifetime still has to be paid for on day one. A system that pays back in two years still costs ₦5–7 million upfront, and most households and businesses don't have that sitting idle, especially when naira borrowing costs 25–30%. The barrier is financing.

    That's exactly the gap capital is built to fill. If you believe the numbers above, and I've spent the last year and a half watching them hold on real sites, then there's a trade available where everyone wins: an investor pays the upfront cost, the customer pays less than they do today, and the investor still earns a fast return, all else being equal.

    The abundance point

    Here is the part I think investors have underrated.

    A diesel user is a proven customer. They have already shown, with their own money, every day, that they'll pay ₦630 or more for a kWh. You don't have to create demand, educate a market or wait for a regulator. The demand is there and it's being served badly.

    So consider the offer an EaaS (Energy-as-a-Service) provider can make to a heavy generator user. "We'll install the system at our own cost. You pay us half what you currently spend on fuel, ₦315 per kWh, for the power you use." The customer saves 50% from day one with no upfront spending. On around 9,000 useful kWh a year, the provider earns about ₦2.8 million on a ₦7 million asset, a payback of about two and a half years. At scale prices, it's under two. That's infrastructure with a long life and an existing, motivated customer.

    That's a different category of investment from most energy infrastructure in Nigeria. A gas power plant takes years to build, depends on a gas supply that gets vandalised, sells into a grid that struggles to collect payment, and needs long-dated finance to make the numbers work. A rooftop system takes days to install, depends on nothing but sunlight, and pays back in two to three years selling to someone who already wants to buy.

    When something is that much cheaper than what it replaces, it will happen. The open question is how fast capital, installers and supply chains can move.

    Chart 3.2Interactive chart in progress
    interactive, cost per kWh of petrol generator, diesel generator (with load slider), Band A grid and solar + battery from 2015 to 2026, with sliders for diesel price, grid hours and cost of capital

    Chapter 4

    The power station is local

    Every other power system Nigeria has tried is centralised. Gas comes from the Delta through pipelines that get vandalised. It's burned in power stations owned by one set of companies, sent through transmission lines owned by TCN (Transmission Company of Nigeria), and delivered by DisCos (distribution companies) that struggle with losses, metering and collection. Every link is a failure point, and the system is only as reliable as its weakest one.

    The generator was the people's answer: decentralise the generation, but keep the fuel. That traded one long chain for another: crude to refinery (or import terminal), depot, tanker, filling station, jerrycan, generator, mechanic. Every link in that chain is also a failure point, as anyone who has queued for fuel during a scarcity knows.

    Solar plus battery cuts the chain to its minimum. The hardware arrives once: panels, inverter, battery. After that, what remains is maintenance: cleaning panels, checking connections, watching battery health, and replacing the inverter once or twice over the system's life. That's real work, and the industry neglects it at its peril, but it comes to a technician's visit every few months. A generator needs a tanker every few days. Solar's fuel arrives every morning without a queue or a price.

    That's decentralisation (generation where the consumption is) and localisation (very little ongoing dependency on anything outside the building). The knock-on effects are bigger than they first look:

    • Fewer tankers. Every litre of generator fuel moves by road. Take billions of litres out of the system and you take a large number of tanker journeys off Nigeria's roads, with fewer accidents, explosions and jams. I suspect solar could take a measurable bite out of Lagos traffic, though I haven't seen anyone quantify it.
    • No fuel scarcity. A business on solar doesn't stop when there's a depot strike or a supply disruption.
    • Cleaner air and quieter streets. Generator fumes kill people every year through carbon monoxide poisoning, and the background roar of millions of engines is a public health problem we've learned to ignore.
    • Foreign exchange. Every kWh made from sunlight is a kWh not made from imported or world-priced fuel. In nine of Africa's top ten solar importers, refined petroleum imports outweigh solar imports by a factor of 30 to 107. Solar is the rare import that reduces the import bill.

    What's still hard

    It would be dishonest to stop there. The economics are excellent, but several things stand between good economics and a country that runs on sunlight.

    Finance. The single biggest barrier is upfront cost. A system that pays back in two years still costs ₦5–7 million on day one, and most households and small businesses don't have it. Naira lending rates are 25–30% or more. Development finance often arrives in dollars, which creates the next problem.

    Currency. Hardware is bought in dollars; customers pay in naira. A provider who signs a five-year naira contract and then watches the naira halve has a problem. The fixes exist (tariffs indexed to diesel prices or exchange rates, local-currency finance, partial guarantees) but they add cost and complexity.

    Quality and maintenance. Nigeria's solar market is full of undersized inverters, dying lead-acid batteries, fake or mislabelled panels and installers who disappear after payment. Every failed system becomes a story about how "solar doesn't work." The industry needs enforced standards from SON (Standards Organisation of Nigeria), real warranties, trained technicians, and remote monitoring so that faults get caught before customers lose faith.

    Design. The flip side of the payback tables in chapter three is that a badly sized system can destroy the economics. Too little battery and the generator keeps running in the evening; too much and the customer pays for capacity they rarely use. Getting sizing right for each site's load, grid pattern and generator habits is a skill, and it's scarce.

    Space and scale. A home or a shop can often fit what it needs on its roof. A factory running a 500 kW load usually can't, and a dense multi-storey building certainly can't. Larger users will need ground-mounted arrays, shared systems, or eventually a grid that can carry solar power from where it's generated to where it's used.

    Seasons. Output drops in the rainy season and under harmattan dust. Systems have to be sized for the worst months.

    The grid's finances. Pakistan shows the danger: when better-off customers cut their grid use, the grid's fixed costs fall on a smaller and poorer customer base, unit prices rise for those left, and more people leave, a cycle known as a utility death spiral. Nigeria needs a grid built to work with distributed solar.

    None of these is a reason to doubt the direction, and all of them are work that needs doing.

    Chapter 5

    Powering production

    Cheap electricity matters less as a household bill than as an input to production. The gains from solar look very different depending on who's using it. Going segment by segment:

    Homes. By number of systems, households are the biggest market, from wealthy homes running 10–20 kVA generators to families running a petrol "I-better-pass-my-neighbour" set for a few hours each evening. The gains are real: lower bills, quieter nights, no fumes, and for rationing households, far more hours of power. Pay-as-you-go solar home systems have already brought light and phone charging to millions who never had a grid connection. But a home mostly consumes power, so the economic multiplier is limited, except where the home is also a business, which in Nigeria it often is.

    SMEs (small and medium enterprises). For a barber, tailor, printer, pharmacy, cold-room operator, welder or restaurant, energy can be one of the largest costs after rent. A hairdresser who cuts her fuel bill in half has, in effect, raised her margin. A cold-room operator who can run 24 hours without diesel can stock more and waste less. The gains here are directly economic, spread across millions of businesses, and they show up as lower prices and more jobs. The challenge is finance at small ticket sizes, which is where lease-to-own and pay-as-you-go structures matter.

    C&I (commercial and industrial). Factories, malls, hotels, hospitals, banks, office blocks and residential estates are the largest diesel burners per site and the most creditworthy customers. This is where EaaS fits best: big loads, daytime consumption, proven willingness to pay, and balance sheets that can sign five- to ten-year contracts. They also tend to run the most oversized generators, so their real cost per kWh is often well above ₦630. A manufacturer whose energy cost falls from ₦700 to ₦300 per kWh is more competitive against imports and, in time, in export markets.

    Agriculture. This is where I think the biggest untapped gains are. Most Nigerian farming is rain-fed, so the land sits idle for months each year. Solar irrigation pumps make dry-season farming possible across the north without diesel. Solar cold storage cuts the enormous post-harvest losses on tomatoes, peppers and fish. Solar-powered milling, drying and processing let farmers sell finished products instead of raw ones. Each of these turns electricity directly into food and income.

    Mini-grids. For communities far from the grid, solar mini-grids run by private developers are often the cheapest route to 24-hour power. Their economics depend on anchor loads: a large daytime consumer such as a mill, a clinic, a telecom tower or a market, whose steady demand makes the system bankable and lets households connect at affordable tariffs. The REA (Rural Electrification Agency) and development partners have built a solid programme here. The next step is scale.

    Telecoms. Nigeria has tens of thousands of telecom towers, and many burn diesel around the clock, especially outside the cities. Towers are ideal for solar: constant load, secure sites, operators with balance sheets. Tower companies have been hybridising for years. Finishing the job would remove a large and very visible slice of diesel demand.

    Where are the biggest gains? It depends on what you measure. By diesel displaced, C&I and telecoms come first. By economic multiplier, agriculture and SMEs come first, because cheaper power there translates directly into more food, more output and more jobs.

    The strategic projects

    If I were designing a national programme, these would be my first priorities:

    1. Industrial parks and export zones on solar and storage, so that the businesses most exposed to international competition get the cheapest power first.
    2. Solar irrigation across the northern states, turning one growing season into two.
    3. Cold-chain corridors linking farms to cities with solar-powered storage at each step.
    4. Market electrification, putting solar on Nigeria's great markets, where thousands of traders currently run thousands of generators.
    5. Hospitals and clinics, where reliable power is a matter of life and death.
    6. Tower solarisation, a contained, high-impact programme with willing private partners.

    Each of these doubles as an anchor load: once a system serves a market, a clinic or an industrial park, connecting the homes and shops around it becomes cheaper and easier.

    Chart 5.1Interactive chart in progress
    interactive, segments plotted by diesel displaced vs. economic multiplier, sized by number of sites

    Chapter 6

    Moving things

    So far this essay has been about stationary power: the generator half of Nigeria's fuel bill. The other half is transport, and most of the petrol Nigeria burns moves people and goods.

    Two- and three-wheelers first. The fastest change will come from okadas and kekes, for a simple reason: they're commercial vehicles running long hours, so fuel is a huge share of the rider's costs, and the vehicles are cheap enough to electrify now.

    A rough comparison: an okada doing 100 km a day burns about 2.5 litres of petrol, or ₦3,500 at ₦1,400 a litre. An electric motorcycle covering the same distance uses about 4 kWh. Even charged at ₦300/kWh, that's ₦1,200. A keke doing 150 km a day might burn 6 litres (₦8,400) against 15 kWh electric (₦4,500). Electric motors also have far fewer moving parts, so maintenance falls too. For a rider who earns a living on the road, that difference is the difference between scraping by and saving.

    Battery swapping makes it practical: a rider swaps a depleted battery for a charged one in a couple of minutes at a station, rather than waiting hours to charge. And if those swap stations are themselves powered by solar and storage, the rider's fuel supply has gone from a refinery and a tanker to a panel on a roof down the road. That's the same localisation logic as chapter four, applied to movement.

    Buses next. City buses run fixed routes and return to the same depot each night, which makes them easy to charge. Lagos has piloted electric buses on its BRT (bus rapid transit) routes. The federal push towards CNG (compressed natural gas) vehicles is a sensible bridge: it uses Nigerian gas rather than world-priced petrol, and it builds the habit of switching away from liquid fuel. But electric buses charged from solar-plus-storage depots will, I think, win on cost by the early 2030s.

    Trucks last. Long-haul freight is the hardest case. Trucks are heavy, travel long distances and need big batteries, and Nigeria lacks charging corridors. I expect trucks to electrify in the late 2030s and 2040s, starting with short, repetitive routes such as port-to-depot, quarry-to-site and urban delivery. For long distances, the better answer is probably rail, which Nigeria should be building anyway.

    How much power does transport need? A rough estimate: if about three-quarters of Nigeria's petrol goes to transport, that's around 13 billion litres a year, holding roughly 120 TWh of chemical energy. Combustion engines waste most of that as heat; electric drivetrains are roughly three to four times more efficient. So electrifying today's road transport would need something like 35 TWh of electricity a year, about the same as the entire grid supplies now, and achievable with roughly 25 GW of solar.

    That's a big number. It's also a small one next to the destination.

    Chapter 7

    Electrostate

    Everything so far has been about replacing what Nigerians already burn. That's the beachhead. This chapter is about what comes after it.

    A petrostate is a country whose economy and politics are organised around selling stored sunlight. Nigeria has been one for half a century, with the paradox of importing much of its own refined fuel for most of that time. An electrostate is something different: a country whose economy runs on cheap electricity it generates at home from sunlight, and which uses that cheap electricity to make things.

    The word that matters in that definition is cheap. Solar will spread across Nigeria whatever the government does, because it already beats the generator. But it won't become cheap on its own. A market prices power against the alternative. While the alternative is a ₦720 generator kWh, solar will be sold at half of that, or a little under it, and the gap between that price and the ₦80–150 it actually costs becomes margin. That's a good business, and it leaves power expensive.

    So what separates a country that has a lot of solar from an electrostate is intent: a government that treats low-cost energy as the industrial strategy itself. China is the clearest example. It built the world's largest solar manufacturing industry and the world's largest solar fleet, and it treats cheap power as an input to everything else it makes. Most industrial strategies pick winners: this sector gets a tax break, that one gets a tariff wall. Cheap energy lifts every producer at once, from the welder in Aba to the fertiliser plant to the cold room at the market. It may be the only industrial strategy that doesn't require the government to guess which industries will win.

    It helps to think of it as a ladder. Each rung is a level of abundance, each needs more deliberate choices than the one below it, and each implies a much bigger fleet of panels and batteries. The capacities line up with the stages in chapter 9.

    LevelPrice of a kWhSolar installedBatteriesRoughly whenWhat gets it thereWhat changes
    Cheaper than diesel~₦300–400~5 GW today, ~20 GW by 2030~30–40 GWhNow to 2030Commercial capital chasing a two-year paybackBills fall and the generator goes quiet
    Priced at cost~₦80–150~20–25 GW~30–40 GWhEarly 2030sPatient capital that accepts a seven-year paybackEvery generator can be replaced by power cheaper than the grid
    Abundance~₦80–150, day and night~100–200 GW~150–400 GWh2040sEnough storage, and tariffs that reward usePeople stop rationing; demand grows three- to fivefold
    Super-abundanceNear zero, for hours each dayBeyond ~200 GW, towards 1 TW~1.5–2 TWhBy 2070Solar deliberately overbuilt, and industries sited to use the surplusIndustries that exist only because power is nearly free

    Cheaper than diesel

    The first rung needs almost nothing from government, and it's where Nigeria is now. Commercial capital in Nigeria wants its money back fast. With naira interest rates at 25–30% and real risks around currency, contracts and collection, investors look for paybacks of around two years. That's why EaaS pricing sits at around half of diesel: it's what a two-to-three-year payback requires. The panels are arriving by the million without a national programme, because the arithmetic already works. At the rate they're arriving (chapter 2), this rung takes Nigeria from around 5 GW installed today to around 10 GW by 2027 and 20 GW by 2030: roughly the size of fleet it would take to replace every generator in the country. What it won't do is make that power cheap.

    Government's job on this rung is mostly to protect it: no import bans, no new levies on panels, and standards that are actually enforced, so that bad systems don't teach people that solar doesn't work (chapter 10).

    But notice what the price is anchored to. A ₦315 kWh is a very good deal for someone paying ₦630 for diesel. For a factory deciding whether to make something in Nigeria or import it, it's still more than twice what manufacturers pay in most of the countries Nigeria competes with. This rung displaces diesel. Making Nigeria competitive takes the next one.

    Priced at cost

    The second rung is the first one that needs a decision. A government, a pension fund, or a development finance institution acting with government backing can afford to be more patient than commercial capital. Suppose public capital accepted a seven-year payback on the same systems.

    Using the numbers from chapter three, a ₦7 million system producing 8,000 useful kWh a year with a seven-year payback means charging about ₦125 per kWh. At the scale price of around ₦4.5 million per system, that falls to about ₦80 per kWh. That's well below Band A grid power, about a ninth of the all-in cost of diesel at scale, and available 24 hours a day.

    Why should public capital accept a lower return? Because the return to the country is far higher than the return on the asset. A government that finances solar at a seven-year payback is buying:

    • Competitive production. Factories, farms and workshops paying under ₦150 instead of ₦700 per kWh can compete with imports and, eventually, export.
    • Foreign exchange. Every litre of generator fuel displaced is world-priced fuel Nigeria no longer has to buy.
    • Household income. The ₦13 trillion a year now burned in generators is money that could go to food, school fees and investment.
    • Health and quality of life. Less smoke, less noise, fewer deaths from fumes and fires.
    • Fiscal room. Cheap distributed power reduces pressure on a grid that has absorbed decades of subsidy.

    Here's the number that I think makes the case. This rung is the size of the generator fleet: replacing its output, around 19 TWh a year, would take something like 15–25 GW of solar and 30–40 GWh of batteries, depending on how well systems match their loads. At scale prices of about $340 per kW installed, battery included, that's roughly $5–9 billion. Nigeria spends about $10 billion a year on generator fuel alone.

    At scale prices, the whole self-generation fleet could be replaced for less than one year's fuel bill.

    Abundance

    Pricing at cost makes power cheaper than the grid. Abundance is what happens when it's cheap enough, and reliable enough, that people stop rationing it.

    That's a bigger change than it sounds. Almost every Nigerian household and business manages electricity as a scarce good: the generator goes on at 7pm and off at midnight, the AC runs in one room, the freezer is switched off overnight to save fuel. The model in chapter three shows what happens when that constraint lifts. A household that ran its generator only in the evening ends up using thousands of kWh a year more than before. Across the country, that's how demand grows three- to fivefold (chapter 8).

    It's also the rung where the fleet grows to match what people want to do with power. Meeting the demand stack in chapter 8, ~180–250 TWh a year, mostly from solar, takes something like 100–200 GW of panels and 150–400 GWh of batteries: the 100 GW stage in chapter 9 and beyond.

    Abundance needs two things the lower rungs don't.

    The first is storage, because abundance means power at 9pm as well as at noon. The model in chapter three points to 1.5–2 kWh of battery per kWp of panels for a system that replaces a generator, and roughly half of the solar that reaches the load passes through the battery first, because so much Nigerian demand comes after dark. For the generator fleet alone that means 30–40 GWh of batteries; at today's Lagos retail price of about $95/kWh, roughly $3–4 billion, already inside the total above. For the abundance fleet it means 150–400 GWh, which is why falling battery prices matter as much as falling panel prices.

    The second is tariffs that reward use. A utility trying to cover its fixed costs from a shrinking customer base raises its unit price. An electrostate wants the opposite: low, flat or falling prices for each extra kWh, especially for anyone using power to produce something. This is the clearest test of whether a government means it. Does it treat a kWh as a source of revenue, or as an input to the rest of the economy? Every naira added to the price of a kWh, whether through levies, duties on panels or tariffs that protect weak local manufacturers, is a tax on everything the country makes.

    Super-abundance

    Solar has a property no fuel has: the more of it you build, the more hours of the day it produces more than anyone can use.

    You can already see this on Lagos rooftops. On a typical home system, around half of what the panels produce goes unused, because it arrives at midday when the battery is full and the house is empty. Today that power is thrown away. In California, South Australia and parts of Europe, so much solar now comes on at midday that wholesale prices regularly fall to zero or below.

    An electrostate plans for that surplus and puts it to work. Panels are now the cheapest part of a solar system, so the cheapest way to have enough power on cloudy days and through the rainy season is often to overbuild them and accept a midday surplus for most of the year. The strategy is then to build things that can use power whenever it's nearly free:

    • Water: pumping for irrigation and treatment, storing water rather than electricity
    • Cold: ice-making and cold rooms that bank the midday surplus as cold for the evening
    • Fertiliser: ammonia made by electrolysis in the hours when power is cheapest
    • Processing: drying, milling and processing crops in daylight
    • Computing: data centres with flexible workloads that run hardest at noon
    • Transport: charging okada, keke and bus batteries at midday swap stations (chapter 6)

    None of these happens by accident. They need the surplus sold to them cheaply: rules that let rooftops sell into estates, mini-grids and the grid; time-of-day prices that pass the near-zero midday cost on to whoever can use it; and industrial parks sited where the surplus is. This is the rung where government intent matters most, because it's the one the market is least likely to reach on its own. A single investor has no reason to overbuild. A country does.

    The surplus arrives well before the terawatt. At around 100 GW, panels on a clear day at noon would produce roughly twice what the whole country uses at that hour. Everything built beyond the ~200 GW that abundance needs is deliberate overbuild, and the terawatt in chapter 9, with 1.5–2 TWh of batteries, is where that strategy leads: about 1,400 TWh a year, 3,500–4,000 kWh per person, much of it at midday and nearly free.

    Choosing cheap

    Put together, the ladder is an industrial strategy, and the strategy is a price. I'd suggest Nigeria set a target for what a kWh of industrial power should cost, the way a central bank sets a target for inflation, and judge every energy policy by whether it moves the country towards that number.

    The machinery could be simple:

    • Public capital, private delivery. Government and pension money provides long-term naira finance; private installers design, install and maintain the systems, paid on performance.
    • Local currency, long tenors. Nigerian pension assets run to tens of trillions of naira. Even a small allocation to infrastructure bonds backed by solar assets would transform the market.
    • Bulk procurement. Buying hardware at national scale at factory prices, with quality standards built in.
    • Anchor-first rollout. Start with industrial parks, export zones, markets, hospitals and cold chains, then extend to the communities around them.
    • No taxes on the kWh. Keep duties, levies and protective tariffs off panels, batteries and the power they make.
    • A grid that buys the surplus. Let rooftops, estates and mini-grids sell midday power to whoever can use it.
    • Independent monitoring. Every system metered and visible, so performance is verifiable and fraud is hard.

    The risks are obvious, and Nigerians know them better than anyone: procurement corruption, politically chosen projects, white elephants. Design is the defence. If private firms are only paid when systems produce verified kilowatt-hours, there's much less room for waste.

    A petrostate makes its money by selling energy dear. An electrostate makes its money by selling energy cheap, at home, and putting it into everything the country makes.

    Chapter 8

    The real prize

    Nigeria uses astonishingly little electricity. Comparable data puts consumption at about 144 kWh per person per year, against 351 in Ghana and 4,198 in South Africa. Adding self-generation lifts Nigeria to perhaps 200–250 kWh per person, still a fraction of its neighbours'. One analysis estimated that, counting latent and suppressed demand, Nigerian per capita consumption should be four to five times its current level.

    The model in chapter three shows that suppressed demand at the scale of a single home. A household that runs its generator only in the evening, once given solar and a battery, ends up using roughly 5,500 kWh a year more than before: power it always wanted and couldn't afford.

    That gap is the market. Here's a rough stack of where demand could come from by around 2040, for a population of roughly 300 million:

    LayerAdditional TWh/yrLogic
    Generator displacement~19Chapter 2
    Making the existing grid reliable~20–30Backfilling outages for people already connected
    Suppressed demand~30–50Cheap, reliable power unlocks fridges, freezers, fans and air conditioning in homes that ration it today
    Universal access~30–40Bringing ~85 million unconnected people to ~400–500 kWh each
    Transport~35Chapter 6
    Cooking~15–20Electric and induction cooking replacing firewood, charcoal and part of LPG
    Productive use and industry~30–60Irrigation, cold chains, processing, light manufacturing
    Total~180–250~3–5 times today's grid and generators combined

    Generator displacement, the whole of chapters 2 and 3, turns out to be about a tenth of this.

    Three further layers sit beyond the table because they're harder to size but potentially larger:

    • Selling the surplus. A quarter to a third of rooftop output goes unused at midday. Across millions of roofs, that's a distributed power station nobody is paying for yet. Neighbour-to-neighbour sales, estate-level networks and mini-grids that buy from rooftops could turn spilled power into income.
    • Energy-intensive exports. Fertiliser from green ammonia, aluminium and other metals, food processing and data centres all compete on the price of power. A country selling electricity at ₦80–150/kWh, while its neighbours run on diesel, has a real industrial advantage.
    • Regional trade. Nigeria sits at the centre of a West African market of over 400 million people. Cheap power can be exported directly over the regional grid, or embedded in exported goods.

    Priced at even ₦150/kWh, 200 TWh a year is a ₦30 trillion annual market, more than twice what Nigerians spend on generator fuel today. And it's served by an asset that, once built, burns nothing.

    This is why I think people are underestimating the space. They size the market as "the diesel bill." The diesel bill is the proof that demand exists. The real market is everything Nigerians would do with electricity if it were cheap and always on.

    Chart 8.1Interactive chart in progress
    interactive stacked bar, today's ~50–60 TWh versus the ~200 TWh 2040 potential by layer, with a hover showing naira, dollar and pound value at a chosen tariff

    Chapter 9

    Stages to a terawatt

    My central claim is that Nigeria should, can and probably will reach a terawatt of solar in my lifetime. That sounds absurd today. It sounds less absurd when you break it into orders of magnitude and ask what each stage looks like and what has to be true.

    Stage 0: a few gigawatts (where we are). By Solepower's estimate in chapter 2, Nigeria had imported around 6 GW of panels by the end of 2025 and is adding about 3.5 GW in 2026. Not every imported panel ends up working on a roof: some sit in warehouses, some go on to neighbouring countries, and some are small or poor-quality and fail early. If about three-quarters end up working, Nigeria had around 4–5 GW of solar at the end of 2025, and will have around 7 GW by the end of 2026. Early adopters are wealthier households, banks, telecoms, some factories and estates. Diesel and petrol generators still produce far more electricity than solar.

    Stage 1: 10 GW and ~15–20 GWh of batteries (by 2027). At this stage solar displaces a large share of C&I and middle-class generator use. EaaS and lease-to-own become normal ways to buy power. Okadas and kekes begin to electrify in the major cities.

    What has to be true: imports holding at 3–4 GW a year, roughly where they are now, finance available at scale for C&I and SMEs, enforced quality standards, and no policy shocks (a point I return to in the next chapter). Keep growing along the central case in chapter 2 and Nigeria has around 20 GW by 2030.

    Prediction: By 2030, Nigeria will have 20 GW of installed solar, and generator fuel consumption will be at least a third below its 2025 level.

    Stage 2: 100 GW and ~150–200 GWh of batteries (by around 2042). At this point solar is the backbone of Nigerian electricity. Generators survive mostly as emergency backup. The grid carries utility-scale solar and storage alongside gas. Most two- and three-wheelers are electric, and buses are following. Panels and batteries are assembled in Nigeria at scale. 100 GW produces around 140 TWh a year, roughly two and a half times today's grid and self-generation combined, and most of the way to the demand stack in chapter 8.

    What has to be true: a national finance facility of the kind described in chapter seven; a reformed grid that can buy, carry and balance distributed solar; local manufacturing (domestic module capacity grew from about 110 MW to around 600 MW between 2023 and 2025, a start); and average installations of about 7 GW a year through the 2030s, about where chapter 2's central case has imports by 2030. That sounds like a lot until you remember Pakistan, with a similar population, installed 17 GW in a single year.

    Predictions: By 2035, a majority of new okadas and kekes sold in Lagos will be electric. By 2042, Nigeria will have 100 GW of solar, more than any other African country.

    Stage 3: 1 TW and ~1.5–2 TWh of batteries (by around 2070). A terawatt of solar produces around 1,400 TWh a year. For a population that may reach 350–400 million, that's roughly 3,500–4,000 kWh per person, approaching what Britain uses today. At this stage trucks and most industry are electric, Nigeria makes its own fertiliser from green ammonia, processes its own crops and minerals, runs data centres for the region, and exports goods made with cheap power.

    What has to be true: installations averaging around 30 GW a year from the 2040s onwards; a manufacturing base for panels, batteries and electrical equipment; a high-voltage grid that moves power across the country; something firm for the rainy season, when solar output falls by a fifth to a quarter (gas, hydro, or deliberately oversized solar); and an economy that has learned to use cheap electricity productively.

    Is 30 GW a year realistic? China installed well over 250 GW of solar in 2024 alone. Thirty gigawatts a year is about a tenth of that.

    What about land? Utility-scale solar takes roughly 1.5–2 hectares per megawatt. A terawatt would need about 15,000–20,000 km², around 2% of Nigeria's land area, and much of it would sit on rooftops, over car parks, or above crops grown in partial shade.

    Prediction: Nigeria will reach 1 TW of installed solar before 2075.

    Chart 9.1Interactive chart in progress
    interactive, the log-scale path from 1 GW to 1 TW with battery storage alongside, stage markers, installation rate required per year, and Pakistan and China overlaid for comparison

    Chapter 10

    What could change

    Predictions this far out will be wrong in the details. The question is whether they're wrong about direction or only about speed. Here's what could shift the pace.

    Technology gets cheaper faster. Sodium-ion batteries, perovskite-silicon tandem panels and continued manufacturing scale could push costs down faster than the curve suggests. That would accelerate everything in this essay, especially stages two and three.

    Oil shocks. A price spike, like the one in early 2026, speeds things up: every rise in diesel is a better sales pitch for solar. An oil price collapse would slow things down, but not reverse them. Even at ₦1,000 a litre, half of today's price, diesel's fuel cost would be over ₦300/kWh, still above what financed solar costs and four times its unfinanced lifetime cost.

    Currency. Devaluation hits fuel and hardware roughly equally, so the comparison barely moves. The damage falls on finance: naira contracts and dollar debt don't mix. Local manufacturing and local-currency capital are the long-term answers.

    Policy. This is the biggest risk to pace. Bad policy can slow the transition badly. When an outright import ban was floated in early 2025, first-quarter panel imports fell 89%. Pakistan, which had exempted panel imports from duties and sales tax, imposed a 10% sales tax on new imports in 2025. Grid rules that penalise self-generation, or tariffs that protect weak local manufacturers at the expense of everyone else, would cost Nigeria years. Good policy (standards, local-currency finance, a grid that welcomes distributed power) could compress the timeline just as dramatically.

    A working grid. If the grid were fixed tomorrow and delivered reliable power at ₦210/kWh, rooftop solar's advantage would narrow, as the Band A rows of the payback table show. But a grid fixed in a world of cheap panels and batteries would itself be built largely on solar and storage. The terawatt arrives by a different route.

    Why the direction holds. Strip everything else away and the argument comes down to this: fuel is dug out of the ground, and panels and batteries are made in factories. Things that are dug up don't get systematically cheaper; things that are manufactured do, every year, for as long as production grows. Nigeria pays for electricity mostly in fuel, at some of the highest prices on Earth, and sits under some of the strongest sunshine on Earth. The gap between those two facts will keep widening.

    The pace will depend on finance, policy and how quickly we build the industry. The direction is already set.

    Every Nigerian generator is burning sunlight that fell hundreds of millions of years ago. The same sunlight is falling on the roof above it today, for free. We are going to learn to use it.

    Sources

    1. NMDPRA March 2026 factsheet (Gazette)
    2. NMDPRA December 2025 data (Shipping Position)
    3. NMDPRA August 2026 factsheet (Gazette)
    4. Nairametrics: August 2026 imports
    5. Nairametrics: Dangote H1 2026 report
    6. Legit.ng: Dangote July 2026 supply
    7. Leadership: MEMAN H1 2026 report
    8. Ember: Africa's solar imports surge 60%
    9. African Business on Ember data
    10. Arise TV: Wood Mackenzie generator capacity
    11. BusinessDay: 2.9m panels imported in 2025
    12. Payne Institute: Africa's Solar Inflection Point
    13. Payne Institute: Pakistan's Solar Surge
    14. Profit: Pakistan grid demand fall
    15. TechEconomy: naira exchange rate, 9 Oct 2026
    16. IndexMundi: electricity per capita, Africa
    17. Sweetcrude: NESG/HBS True Cost of Electricity
    18. Industrial Info: NNPC study on petrol and diesel used for power (2015)
    19. Sweetcrude: NBS Power Sector Data Preview, generators' share of electricity
    20. Africa Check: how many generators Nigerians own
    21. ThisDay: NNPC on petrol supply, 2021 and 2022
    22. BusinessDay: petrol imports in 2019
    23. Ripples Nigeria: petrol imports, first half of 2023 (NBS)
    24. TheCable: petrol imports in 2023 (NBS)
    25. Ember: China Solar PV Exports dataset (customs data, monthly)
    26. Ember: Chinese solar exports double in a month (March 2026)
    27. BusinessDay: China's share of Nigeria's panel imports, H1 2025 (NBS)
    28. Daily Trust: Nigeria's panel exports, Q1 2026 (NBS)
    29. NBS: Diesel and Petrol Price Watch reports (microdata catalogue)
    30. Middle East Insurance Review: FRSC and NIA vehicle numbers

    Let’s power your next chapter.

    Tell us what you need and continue your enquiry on WhatsApp.

    Review your message in WhatsApp, then tap send.