Microfinance is the business of lending small amounts of money to people banks do not serve — market traders, smallholder farmers, street vendors and the self-employed. CGAP, the World Bank-hosted research centre on financial inclusion, defines it as the supply of loans, savings and other basic financial services to the poor. The scale is large: more than 142 million people borrowed from microfinance institutions worldwide in 2024, according to the Microfinance Barometer. In Sub-Saharan Africa, the sector served 6.3 million borrowers with a loan portfolio of $10.3 billion in 2018, the latest year for which a regional breakdown was published — and 64% of those borrowers were women.
It exists because most African adults still cannot borrow from a bank. Account ownership is rising — in Sub-Saharan Africa it grew to 58% of adults in 2024, up from 49% in 2021, according to the World Bank’s Global Findex 2025 — but an account is not credit. A trader with a bank account and no collateral, no payslip and no credit history is still, in the bank’s eyes, unlendable. Microfinance was built for exactly that person.
What microfinance actually is
Microfinance is not one product. It is a set of small financial services: microloans for working capital, savings accounts that accept tiny deposits, micro-insurance, and money transfers. The loan is the best-known part, but savings matter just as much — a safe place to keep money is often what a trader needs first, before credit.
The providers are just as varied. Some are non-governmental organisations. Some are cooperatives and savings groups. Some are licensed non-bank microfinance institutions, and some are commercial banks running a microfinance arm. What they share is a client base the formal banking system has mostly ignored, and a loan book made up of thousands of small loans instead of a few large ones.
Two things microfinance is not: it is not charity — the loans are real loans, repaid with interest, and the institution has to cover its costs or it closes. And it is not a cure for poverty on its own. The evidence from randomised trials of microcredit is that its effects on household income are modest rather than transformative. A loan can stock a stall or buy seed; it cannot fix a bad harvest, a sick child or a market with no customers.
How a loan reaches a borrower
The classic model, developed by the Grameen Bank in Bangladesh, works in steps — and most African group lenders still follow a version of it.
- The borrower joins a group. Typically five people from the same village, often women, who know each other’s circumstances. The group, not the lender, does the first credit check: members will not accept someone they believe will not repay.
- A field officer visits, not the other way round. There is no branch to walk into and no pile of documents a bank would demand. The officer records the business, the household’s cash flow and the group’s members.
- Loans go out in sequence, not all at once. In the classic design, two members borrow first. If they repay on schedule, the next two become eligible. The group chairperson borrows last.
- Repayment is small and frequent. Weekly instalments, matched to the daily or weekly cash flow of a trader or farmer — not a monthly salary cycle most borrowers do not have. The group meets weekly, repayments are collected in public, and savings are made alongside the loan.
- The group is the collateral. No asset is pledged and no court is involved. If one member struggles, the group covers the payment or puts pressure on them to pay — and every member knows that a default can cut off the group’s access to future loans.
- Repayment builds the next loan. Borrowers who repay on time qualify for larger loans. This is progressive lending: the loan grows as the borrower proves they can handle it, and savings accumulate alongside.
The model was pioneered by Muhammad Yunus and the Grameen Bank, formally established in Bangladesh in 1983. Yunus and the bank jointly won the Nobel Peace Prize in 2006. Not every African lender still uses the group model — many now lend to individuals — but the logic survives in all of them: when a borrower has no collateral, the lender substitutes knowledge of the borrower, frequent contact and the promise of a bigger loan next time.
How the phone changed the model
The biggest change to African microfinance did not come from a bank. It came from a phone. M-Pesa, launched in Kenya by Safaricom in 2007, put a payment account in the hands of people who had never had one. Then, in November 2012, Safaricom and the Commercial Bank of Africa launched M-Shwari, a savings account and mini-loan service inside the M-Pesa menu. There were no forms to complete and no branch to visit. Eligibility was decided from the customer’s own M-Pesa transactions and savings history, and the loan money was sent straight to their M-Pesa account. At launch, users could borrow as little as KSh 100.
The uptake was fast. Within two years of launch, M-Shwari had more than 9 million savings accounts and had disbursed over 20 million loans. The phone had replaced the weekly group meeting with an algorithm: instead of five neighbours vouching for a borrower, the lender read the borrower’s own transaction history.
Mobile money is now the channel through which much of African microfinance moves. 40% of adults in Sub-Saharan Africa had a mobile money account in 2024, up from 27% in 2021 — the highest mobile-money use in the world, according to the World Bank’s Global Findex 2025. Formal saving in the region rose 12 percentage points to 35% of adults over the same period. But the phone also removed the human checks the group model provided. An algorithm cannot see that a borrower’s stall burned down, or that they already owe three other lenders.
Why the interest rate is high — and what can go wrong
The first question most people ask about microfinance is why the interest rate is so high. The answer is cost. CGAP found that microcredit rates are higher, often much higher, than normal bank rates, mainly because it inevitably costs more to lend and collect a given amount through thousands of tiny loans than to lend and collect the same amount in a few large loans. A field officer visiting a village every week costs the same whether the loan is small or large — so that cost is a far bigger share of a small loan. The rate has to cover it, or the lender cannot survive.
That does not mean every rate is fair. When several lenders chase the same borrowers, or when loans are taken for consumption rather than for a business that earns, debts pile up. Kenya shows what the strain looks like from the lender’s side: the Central Bank of Kenya’s 2024 supervision report found the country’s 14 licensed microfinance banks shrinking — their loan book fell 16.8%, from Sh37.5 billion in 2023 to Sh31.2 billion in 2024, non-performing loans rose to Sh7.38 billion, and the sector posted a pre-tax loss of Sh3.5 billion, its ninth consecutive year of losses. Branches fell from 115 to 107 as banks and digital lenders took their customers.
For the borrower, the risks are different. A loan taken to pay school fees or a medical bill does not generate the income to repay itself. A missed payment on a phone loan can be reported to a credit reference bureau, which then blocks future borrowing. And a lender that lends too fast, to too many people, can turn a useful service into a debt trap — which is why the rules around microfinance matter as much as the loans themselves.
What to watch
The rails getting cheaper. The cost of a small loan is mostly the cost of moving small amounts of money and collecting small repayments. South Africa is opening its national payments system directly to non-bank operators, including micro-lenders, from mid-2027, after a pilot in January 2027 — so a loan can move from lender to borrower’s wallet without passing through a commercial bank sponsor. If that cuts the cost of each transaction, it should, in time, cut the price of the loan.
Regulators licensing digital lenders. Microfinance is a licensed activity in most African countries, and regulators are now licensing digital-first firms. In Côte d’Ivoire, Djamo Finances secured a microfinance licence in September 2025, letting it offer credit, interest-bearing savings and current accounts from a smartphone. Financial inclusion in Côte d’Ivoire has risen from 41% in 2017 to 58% in 2024, according to the World Bank — but access to credit and savings still lags behind account ownership, and the licence is meant to close that gap.
Whether protection keeps up with speed. A loan approved in seconds from a phone needs rules written for that speed: clear disclosure of the total cost, limits on how many loans one person can hold at once, and a way for borrowers to complain. The question that will decide whether microfinance helps or harms in the next five years is not whether the loans get faster — they will — but whether the cost of a small loan finally comes down, and whether the borrower is protected when it does not.
Microfinance works when three things hold: the loan is small, the repayment rhythm matches the borrower’s cash flow, and the lender is regulated. When any of the three breaks, the borrower pays for it.
