Financing Clean Energy Access in Emerging Economies

Of the 671 million people without electricity in the latest data for these 118 countries, 594 million live in sub-Saharan Africa. Nigeria (89 million) and DR Congo (87 million) together account for more than a quarter of the total, followed by Ethiopia, Tanzania, Uganda, Mozambique, Niger, Madagascar and Malawi. The 2025 Tracking SDG7 report gives a global figure of 666 million for 2023, 85% of them in sub-Saharan Africa; the slightly different numbers reflect country coverage and data years.

Progress is real but uneven. The share of sub-Saharan Africans with electricity rose from 24% in 2000 to 39% in 2015 and 55% in the latest data, but population growth has absorbed much of the gain. Four in five of those still without power live in rural areas, often far from the grid and in low-density settlements, which is why off-grid solar and mini-grids are now the least-cost route for a large share of new connections.

The clean-cooking gap is three times larger and differently distributed: about 2 billion people, half in sub-Saharan Africa and a fifth in South Asia, with large numbers in East Asia too. Many people without clean cooking already have electricity, so cooking access is partly a matter of fuels, stoves and their financing rather than of new infrastructure. The IEA estimates that universal clean cooking by 2030 would require about $8 billion a year, roughly half of it concessional.

International public flows for clean energy to developing countries peaked at $31 billion in 2016 (constant 2023 dollars), fell to $14 billion in 2020, and recovered to $24.6 billion in 2024. Over the decade 2015–2024 they totalled about $201 billion, of which $172 billion can be attributed to the 118 countries in this dashboard. The least-developed countries received about $41 billion, or about $3.2 billion a year since 2020.

Sub-Saharan Africa received $48.6 billion, 28% of the attributable total, while holding 89% of the people without electricity. South Asia received 15% with almost none of the electricity gap, Latin America 18% with 2%, and Europe and Central Asia 13% with none. The money is concentrated: the ten largest recipients, led by India ($19.1 billion), Pakistan, Türkiye, South Africa, Brazil and Egypt, took 45% of the total.

This is not, in itself, misallocation. These flows fund renewable generation of every kind, and large middle-income power systems decarbonising quickly is a global good. But it shows that the existing channel of international public finance for clean energy is not an access channel: it follows bankable projects in large, creditworthy systems. Access in the poorest countries depends on finance that is designed for it.

Dividing each country's decade of clean-energy finance by its number of people without electricity gives a crude but revealing ratio. In sub-Saharan Africa it is $82; in East Asia $606, the Middle East and North Africa $863, Latin America $2,405 and South Asia, which has almost closed its gap, $8,681. Among countries with at least a million people without power, the median is about $50.

The typology uses that threshold. The forgotten frontier is 25 countries with at least a million people without electricity and less than $50 per person received: Nigeria ($19), DR Congo ($12), Ethiopia, Tanzania, Mozambique, Malawi, Niger, Sudan, Chad and others, holding 486 million people without power but only 7% of the finance. The funded frontier is 25 countries with large gaps but more finance, including Kenya, Uganda, Ghana, Senegal, Rwanda and Zambia: 175 million without power and 39% of the money. The last mile is 10 smaller countries below 95% access, and the near-universal group 57 countries with 6 million without power and 53% of the finance.

The forgotten frontier is not random. It includes most of the world's fragile and conflict-affected states, which donors and investors find hardest to reach, and several of the largest economies of sub-Saharan Africa, where macroeconomic crises have raised the cost of capital. Eleven of the 25 have lending rates of 15% or more or currencies that lost at least half their value since 2015, and 15 have government revenue below 15% of GDP or debt service above 15% of exports. They are where access finance has to be most concessional and most patient.

The IEA's Africa Energy Outlook 2022 estimated that universal access to modern energy in Africa by 2030 would require about $25 billion a year. The IEA's 2023 clean-cooking report put universal clean cooking worldwide at about $8 billion a year. The 2024 Off-Grid Solar Market Trends report estimated that reaching universal access through off-grid solar alone would require about $21 billion of investment, of which $9.2 billion would be needed to close the affordability gap through subsidies.

Against these, international public clean-energy finance to sub-Saharan Africa averaged about $4.1 billion a year in 2020–2024, and to the least-developed countries about $3.2 billion. Even if every dollar went to access rather than to large power projects, it would cover about a sixth of the IEA's estimate for Africa. Domestic budgets, utilities and private investors add to these flows, but in the forgotten-frontier countries all three are constrained by the macroeconomic conditions examined in the next sections.

The ESMAP study Mini Grids for Half a Billion People estimated that 210,000 mini-grids, mostly solar hybrids, could serve 490 million people at least cost by 2030 for about $220 billion, if their cost fell to $0.20 per kWh. The World Bank and the African Development Bank launched Mission 300 in 2024 to connect 300 million Africans by 2030. These are the right scale of ambition; the question is whether the finance can be made cheap and targeted enough.

A diesel generator's cost is mostly fuel, paid over time; a solar home system's or a mini-grid's cost is mostly equipment, paid up front and recovered over years. That makes the cost of capital the largest single driver of the price of clean access. Recovering $100 over 20 years costs $6.72 a year at a 3% cost of capital, $10.19 at 8%, $15.98 at 15% and $20.54 at 20%. The same asset is 2.4 times as expensive to run each year when financed at 15% instead of 3%.

That gap is the normal condition of the countries with the largest access deficits. The IEA finds that the cost of capital for utility-scale solar in emerging and developing economies is more than twice that in advanced economies, driven by currency, regulatory, offtaker and country risks, and estimates that narrowing the gap by one percentage point would save about $150 billion a year in financing costs. The median commercial lending rate in the 27 sub-Saharan African countries with data is 14.3%, against 10.5% in other low- and middle-income countries; in Malawi it is 37% and in DR Congo 24%.

For access companies the problem is sharper still: pay-as-you-go solar and mini-grid developers borrow working capital to finance the customer's purchase, so their cost of debt passes directly into monthly payments. A concessional loan or a guarantee that lowers the cost of capital by a few points therefore buys more access than the same money spent as a grant on a few projects, as the simulator in section 8 shows.

An access company in Lagos or Addis Ababa earns naira or birr from its customers and often owes dollars to its lenders and suppliers. Between 2015 and 2024 the naira lost about 87% of its value against the dollar, the Angolan kwanza 86%, the Egyptian pound 83%, the Ethiopian birr 75% and the Ghanaian cedi 74%. A company with dollar debt in those markets faced a multiplied debt burden just as customers' incomes fell. Lenders price that risk in, or stay away.

Public budgets carry the other half of the risk. Universal access needs public money for grid extension, utility viability and subsidies for poor households, but government revenue (excluding grants) is a median 15.8% of GDP in sub-Saharan Africa against 21.3% in other developing countries, and many governments spend a large share of revenue and exports on debt service. Where utilities are insolvent, private generators and mini-grid operators demand guarantees, which governments with thin budgets struggle to provide.

Two instruments target this directly. Local-currency finance, provided by local banks, by development-bank local-currency lending or by currency-hedging facilities, removes the mismatch between revenue and debt. Guarantees and political-risk insurance from multilateral banks transfer the risks that private lenders price highest to institutions better able to bear them. Both reduce the cost of capital more than they cost the public purse.

Across 106 countries, the share of people living on less than $3.00 a day is correlated with access to electricity at −0.86. The median poverty rate is 39% in sub-Saharan Africa and 2.4% in other developing countries. Where most people are extremely poor, even a kit costing a few dollars a month competes with food, school fees and health care, and payment collection suffers: the 2024 Off-Grid Solar Market Trends report found that only 22% of households without electricity could afford a basic Tier 1 kit on pay-as-you-go terms, and 1% a Tier 2 system.

That is why the same report estimated that $9.2 billion of the $21 billion needed for universal off-grid access would have to close an affordability gap, rather than earn a commercial return. Results-based financing, which pays companies a fixed amount for each verified connection in poor or remote areas, and end-user subsidies targeted through social registries or digital payments, are the main tools. They keep competition and private delivery while lowering the price that the poorest households pay.

Affordability also depends on what electricity does. Connections that power productive uses, such as irrigation pumps, mills, cold storage and workshops, raise incomes and so the ability to pay, and they make mini-grids viable by adding daytime demand. Pairing access finance with finance for appliances and productive equipment is one of the most effective ways to make access self-sustaining.

The simulator divides the investment in a connection into three layers. Grants or results-based subsidies are not repaid. Concessional loans are repaid at a low rate. Commercial debt and equity are repaid at the commercial cost of capital, which a guarantee or local-currency finance can lower. The annual capital charge, which customers ultimately pay through tariffs or kit payments, is the sum of each repaid layer's capital recovery. The public money used is the grant plus the concessional loans.

With the default settings ($300 per person, a 10-year life, 20% grant, 30% concessional at 1.5%, and commercial finance at 16%), the blended cost of capital is about 11% and the annual capital charge about $41 per person, a third less than with all-commercial finance. Raising the grant share lowers the charge but spreads public money over fewer people. A 4-point guarantee on the commercial layer delivers much of what a bigger grant would, cutting the charge to about $36, without spending more public money up front.

The figures are illustrative. Real projects add operating costs, taxes, foreign-exchange hedging and company margins, and the right blend depends on the technology and the country. But the arithmetic is general: in high-cost-of-capital countries, money spent lowering the cost of capital for everyone goes further than money spent subsidising a few projects at full cost, while grants are irreplaceable where households cannot pay.

The calculator multiplies the number of people without electricity and without clean cooking by an illustrative investment per person, spreads it over the years to the target, and compares the annual investment with three yardsticks: the international public clean-energy finance the country received on average in 2020–2024, its GDP and its government revenue. It also shows the annual capital charge that the investment carries at the chosen cost of capital, which is what customers and budgets must eventually service.

For Nigeria, at $300 per person for electricity and $30 for clean cooking, universal access by 2030 needs about $32 billion, or roughly $6.4 billion a year, 30 times the average international public clean-energy flows of about $0.2 billion a year in 2020–2024. For DR Congo the burden is proportionally larger still: about $5.9 billion a year, more than 30 times its average flows and about 6.5% of GDP. For many funded-frontier and last-mile countries, by contrast, universal access is within reach of existing flows if they are directed at it.

The investment-per-person values are deliberately adjustable because estimates vary widely. Densifying an existing grid in a town costs less per person than extending lines to dispersed villages; a solar home system for lighting costs far less than a mini-grid that can run a mill. Population growth adds to the totals over time and is not included. The point of the calculator is the order of magnitude and the comparison, not a project budget.

Target the frontier. Access finance should follow the gap, not the pipeline of bankable projects. National electrification plans that map least-cost technologies settlement by settlement, as Mission 300's national energy compacts aim to do, give financiers a pipeline in the countries that most need it. Concessional resources, including IDA, the African Development Fund and climate funds, are scarce, so they should be concentrated where the gap is large and private finance is absent: the forgotten frontier.

Lower the cost of capital. Guarantees, first-loss capital and political-risk insurance let a small amount of public money unlock much larger private finance at lower rates. Local-currency lending and hedging facilities remove the exchange risk that makes dollar debt so costly in Lagos or Lusaka. Standardised contracts, transparent procurement and credible regulators lower risk premiums at no fiscal cost, and utility reform that makes offtakers creditworthy reduces risk for every generator.

Pay for outcomes and close the affordability gap. Results-based financing that pays a fixed amount per verified connection, adjusted for poverty and remoteness, has helped scale off-grid solar while keeping competition. Targeted end-user subsidies, delivered through digital payments, reach households that cannot pay full cost. For clean cooking, finance for stoves and fuels, liquefied petroleum gas or electric-cooking distribution, and carefully verified carbon finance can lower prices. Finance for productive-use appliances raises incomes and demand together.

Countries: 118 low- and middle-income economies (World Bank classification) with at least 0.5 million people. Access gaps = population × (100 − access rate) for electricity (EG.ELC.ACCS.ZS) and clean cooking (EG.CFT.ACCS.ZS), latest year since 2015; rural share of the electricity gap from rural population and rural access. Finance: SDG 7.a.1 international public flows for clean energy R&D and renewable energy production, including hybrid systems, by recipient (OECD/IRENA via Our World in Data, constant 2023 US$), summed 2015–2024; flows not attributed to a country are excluded from country totals.

Typology: forgotten frontier = at least 1 million without electricity and under $50 of 2015–2024 finance per person without electricity; funded frontier = at least 1 million and $50 or more; last mile = under 1 million and access below 95%; near universal = access of 95% or more. Cost of capital: capital recovery factor r / (1 − (1 + r)^−n). Stack: annual charge = cost × (concessional share × CRF(rc) + commercial share × CRF(rp − guarantee)); grants are not repaid; blended rate = rate that gives the same charge on the repaid layers. Currency loss = 1 − rate₂₀₁₅ ÷ rate₂₀₂₄ (local currency per dollar).

Caveats: SDG 7.a.1 covers international public flows for renewables in general, not access spending, and excludes private, domestic and many non-OECD flows. Investment per person is illustrative. Lending rates are commercial bank averages, not project costs of capital, and are missing for many countries. Access figures differ slightly from Tracking SDG7 because of coverage and data years. Need estimates are from the cited reports and are not additive.