Global financial inclusion surveys show that there are billions of low-income individuals who want a safe place to save, and evidence suggests that they are willing to pay for it. Many low-income individuals are currently able to store small values with licensed financial service providers. Yet these providers are usually reluctant to offer such services, because it is difficult to do so without losing money. Providing low-income individuals with a safe place to save is a challenging commercial proposition, because they tend to keep small balances, prefer to make many transactions and live and work far from traditional banking outlets. These challenges are further complicated by the upfront costs of opening accounts, the inability to charge clients enough to cover the costs of transactions, the difficulty of earning a return on funds mobilized and uncertainty about how to allocate revenues and expenses to determine profitability.
This paper focuses on improving the economics of servicing small-balance transactional accounts by low-income individuals in developing countries. It is estimated that less than 20 percent of adults have access to a bank account in most African countries.  While aggregate data on account balances is not available, it would not surprise us to find that only 20 percent of those that do have an account have balances greater than $50. Banks need to see a business case for extending service to those currently unbanked and continuing to serve those with small balances.
Understanding the Savings Needs of Low-Income Individuals
The average balances of low-income individuals’ savings accounts link to the poverty level of clients. However, averages must be understood in the context of how savings accounts are used by low-income individuals.
The book Portfolios of the Poor documents the variety of ways in which low-income households ”push” and “pull” money on a daily, weekly or seasonal basis. It documents how from time-to-time they act as microfinanciers, loaning funds out to friends and family when they have disposable funds and how they are in turn subjects of credit from local shops, employers and moneylenders.  Without finance, people would be relegated to a hand-to-mouth existence. For low-income individuals with little income security, finance – understood as the ability to transfer value over time, space and people – is a necessity. Indeed, low-income households typically employ a surprising number of informal financial tools.
We can segment the world’s low-income population (the 1.6 billion working age adults who live on less than $2 per day), based on their primary source of income. The main categories are as follows: 
Their income is concentrated in a few periods during the year. They need to set aside part of their harvest and save it to buy seeds and fertilizers for the next planting season. They need to spread their income over several months to tie them over until the next harvest season. And they need to build buffers against potential crop failures.
Many are paid daily but cannot be sure there will be a job for them to do tomorrow or next week. They need to insulate their daily expenditures from this income volatility by transferring income from good days/weeks to cover for bad days/weeks. They also need to build the discipline to set aside some excess cash on good days/weeks to build up to meet larger expenses.
They have a more steady income stream, but their income comes in small amounts. They need to accumulate small amounts from their regular wages to save up to a larger goal (e.g. school fees, house repairs or dowry).
They need to be able to self-fund their working capital needs to sustain and grow their business and to safely handle excess cash from one day to the next.
(610 million people)
(370 million people)
(300 million people)
(180 million people)
Illustrating the Challenging Economics of Small-Balance Accounts
Low-income individuals use their accounts in a wide variety of ways. Consider three sample clients with different types of usage patterns that all result in an average balance of $83 over the period of a year, as shown in Figures 1 (account balances) and 2 (number of transactions):
– Mercy is a farmer who receives one payment of $230 for her crops at harvest time. She then withdraws $80 at harvest, $100 at planting and $45 before the holidays.
– Peter is a market trader whose account fluctuates between $0 and $230 at the end of the year. Each month, he makes two deposits to store sales revenue and two withdrawals to buy inventory.
– Esther is a domestic worker who lives in the city and supports her family in a rural area. She deposits $15 per month (a quarter of her monthly wages) and withdraws all but $10 at the end of the year when she visits her family over the holidays.
Making the business case for small-balance savings can be tough, as the margin on float may not amount to much. To get a handle on the basic economics of small-balance savings, imagine that the three savings patterns depicted above were channeled through an account that offers free account opening, free deposits and one free withdrawal per month, with a $0.50 charge for each subsequent withdrawal. The account costs $6 to open (spread over three years) and has a 6 percent allocated margin. There is a $1 cost to the provider for each teller transaction and $0.20 per month in back-office costs.
Figure 3 shows that the costs to serve these clients far outstrip the revenues that can be earned. Peter offers the worst economics, given the large number of transactions he incurs. Esther’s profitability is undermined, because she does many more (free) deposits than withdrawals.
There are some levers that financial institutions can use to move these accounts toward viability. For example, if an institution can encourage higher balances or increase their interest margin, they would have more resources to cover the costs of the account. Transactional can be shifted to lower-cost channels, or higher transactional charging can be introduced to mitigate the high cost of transactions. Alternatively, higher-profitability products can be cross-sold to attribute selling commissions to the savings account.
Scenario B in Figure 4 shows the breakeven point needed to bring each client to viability using just one lever in turn and leaving everything else constant. None of the levers could be expected to solve the economics on their own across the three clients. For example, it would be difficult for Esther to maintain a balance of $290, or to bring down the cost of transactions to $0.01 for Peter, or for Mercy to pay a $1.47 withdrawal fee.
However, it would be possible to combine these levers to achieve viability for all three clients. Scenario C in Figure 4 shows a “combination scenario,” wherein all levers are moved slightly in tandem. By making smaller, more reasonable changes to each lever, such as requiring a $115 balance or lowering the costs of transactions by half, all three clients can be profitably served. Figure 5 depicts the profitability of the combination scenario for each client.
Levers for Increasing Revenues and Margin
One way to increase the revenue from accounts is to increase their average balance. Most low-income individuals spread their savings over many options, including in accounts, under the mattress, with friends, with informal groups and even in assets they can sell, such as jewelry or livestock. If financial institutions captured a larger share of low-income individuals’ savings, they could raise more of their funding from retail sources at a lower net cost. Banks would need to design products that cater to the diverse needs and desires of the target market, instill a greater sense of trust through marketing and branding efforts and increase the convenience of transacting.
Banks can also charge reasonable transaction fees to help defray costs and align the incentives of the client with the costs of the service. Low-income individuals prefer to pay on a per-transaction basis rather than having monthly fees that erode their balance over time, even if they are not transacting. They have also demonstrated a willingness to pay if they are offered more transactional convenience through proximity.
Providers can offer a fuller value proposition for clients, taking into account the indirect revenue from cross-selling credit, money transfers or insurance. For many microfinance institutions, the economics of savings is strengthened by microcredit – the notion that the account anchors the customer relationship, and the loan gives it profitability. But financial inclusion premised on credit is always going to leave some people behind: those who do not feel like credit is the right financial tool for them or who simply do not have the ability to commit to future payment streams.
Providers can also try to increase the return on assets within prudential guidelines, so as to widen the interest margin. Figure 6 shows how all these revenue enhancement and cost reduction options might play out for the three sample savers under the combination Scenario C.
Levers for Reducing Costs
Figure 3 shows that serving transactions (deposits and withdrawals) are the main cost component of small-balance accounts. These are typically conducted through a teller at a branch, which requires expensive labor and significant fixed costs. It is possible to increase process efficiency and reduce the time per teller transaction by adding bill counters, reducing required paperwork, streamlining queuing systems, automating transactions though point-of-sale (POS) devices and training staff. However, this will only bring down teller costs if the tellers are able to use the time saved to operate at full capacity or the number of tellers is reduced. Right-sizing outlets in low-income and especially rural communities might also lead to cost savings, for instance by reducing the size of outlets, renting existing space, using pre-fabricated buildings or even refurbishing shipping containers. Unit transaction costs for a mini-branch might be 20 percent lower than for a full branch (we assume $0.80 versus $1 respectively).
Shifting transactions to cheaper delivery channels provides a further step-change reduction in costs. A POS-enabled retail store might be able to process transactions at an average cost of $0.24, whereas a transaction through a mobile-enabled retail store might cost $0.16. (See Box 1 for a description of how we estimated this cost.) These channels rely on cash merchants based in low-income and rural communities to provide cash-in and cash-out services on
behalf of the provider with a real-time connection to client accounts. Such channels also bring services closer to the low-income individuals and make them more convenient to use, which can increase willingness to pay.
Figure 7 shows the significant reduction in total transaction costs in serving our three clients that is made possible by using cheaper delivery channels, based on the unit cost assumptions mentioned above.
Box 1: Unit Transaction Costs under a Mobile Phone-Enabled System
We estimated the unit cost (in U.S. cents) of conducting basic savings transactions in a mobile money platform that uses retail outlets as cash in/out points, based on a bottom-up analysis of the various incremental cost components. The following are the main items of costs, in descending order of importance:
Another chunk of costs relate to the opening of new accounts. These costs include form filling, identity verification, record-keeping, issuance of payment instruments and marketing costs. Some providers charge upfront fees that cover some of these costs, but such fees seldom cover the full cost and are unpopular with clients. Account opening costs can be reduced by delegating the task to retail outlets, hiring temporary staff paid on a commission basis, simplifying the account opening procedures through tiered Know Your Customer (KYC) requirements and reducing the amount of paperwork. The amortized costs of account opening can also be reduced by increasing client loyalty so that costs are spread out over more years.
Finally, providers can consider reducing the interest paid to depositors. Given the paucity of good savings alternatives, interest earned on deposits is relatively low on the priority list for low-income individuals compared to the safety and convenience of their savings.
Achieving Transformational Change in the Economics of Small-Balance Acccounts
Determining which levers of viability to pull will depend on the characteristics of the financial service provider and market dynamics. While all of the levers can help, we believe that transformational change in the economics of small-balance accounts will most likely come from storing a higher portion of low-income individuals’ savings, moving to a transactions-driven revenue model, and perhaps most significantly, moving transactions beyond bank branches. Small-balance accounts may offer challenging economics, but these challenges can be solved with the right combination of products, pricing and delivery to provide all people a safe place to save.
 Napier, Mark (2010), Real Money, New Frontiers (Claremont: Juta and Company Ltd), p. 1.
 Collins, Daryl, Jonathan Morduch, Stuart Rutherford and Orlanda Ruthven (2009), Portfolios of the Poor: How the World’s Poor Live on $2 a Day (Princeton: Princeton University Press).
 Segmentation data is from Oliver Wyman (2007), “Sizing and Segmenting Financial Needs of the World’s Poor,” unpublished paper commissioned by the Bill & Melinda Gates Foundation.