Building a Robust Case for Microsavings


This page in:

Editor's Note: The following post was submitted jointly by Jake Kendall and Ignacio Mas of the Bill and Melinda Gates Foundation.

At the Financial Services for the Poor team at the Bill & Melinda Gates Foundation we have made a deliberate choice to focus on promoting savings (you can read about our strategy here). We think that saving in a formal, prudentially regulated financial institution is a basic option that everyone should have. Having a safe place to save allows people to manage what little they have more effectively and to self-fund life-improving or productivity-enhancing investments without paying the high interest rates associated with small loans. Accessing other people’s money through credit may not be right for everyone, but making the most out of your own income surely is. From a donor perspective, we need to move beyond microcredit and support the development of broader markets. In fact, too much focus on microcredit risks tilting the incentives of local financial intermediaries to funding their credit portfolio from external soft funds rather than via mobilizing local deposits.

As I go around the world talking up these issues, I am struck by how often I need to justify the value of savings for poor people intellectually. Sure, we should do more to demonstrate these benefits with actual data, and we are funding a bunch of studies in this regard. But why is the notion so counter-intuitive for many people? I would trace that to two misconceptions and two fears.

Misconception #1: Poor people don’t save.

They do, just not often in formal financial institutions. Poor people’s income is not only small but often irregular (e.g. small-holder farmers with seasonal income, or day laborers without guaranteed employment). Occasionally they face large shocks which can easily overwhelm their means, arising from entirely predictable lifecycle events (marriage, death) or unpredictable but not altogether unforeseeable occurrences (accident, illness, drought). Savings help them maintain consumption and plan certain investments in the face of erratic income streams and occasional disruptive events. There is a broad base of empirical evidence that this does happen: we can observe the prevalence of informal savings groups and deposit collections in many countries; financial diaries of poor households show that the marginal propensity to consume out of current income is often less than one; and people report wanting a bank account or formal savings arrangement when asked in surveys. The book Portfolios of the Poor documents the many ways in which poor people manage what little they have in order to survive on less than $2 per day. The question is not so much whether poor people save, but rather why they are not saving in formal institutions.

Misconception #2: If the poor are saving, that means they are not investing in their microenterprises, and that limits their ability to escape poverty.

The fallacy here is to view savings as an end in itself rather than as a means to an end. In developed countries, saving is often thought of as a nest egg for a broad purpose such as retirement, while credit is thought of as a tool geared toward specific short- to medium-term needs like purchasing a flat screen TV. However, because they often lack easy access to credit, poor people often save for a specific short term purpose: to provide for food tomorrow, to buy a bicycle, to pay school fees, to prepare for a daughter’s wedding, to build assets to pay for an eventual funeral—or indeed to invest in a family business. Savings and credit are alternative paths to setting aside small daily or weekly sums to fund a larger expenditure or investment. The differences are in timing (with credit you may make the expenditure or investment faster) and cost (with savings it will be a lot cheaper). If you believe microentrepreneurship can lift large numbers of people out of poverty, imagine if they could do that while retaining the full value of their efforts because they don’t need to pay onerous terms to credit suppliers.

Fear #1: Raising local savings is a way of taking resources from the poor and their communities in order to invest elsewhere.

This certainly may happen, but it may not be altogether a bad thing. The poor are still better off if bank accounts give them the option of taking their savings out from unproductive assets (cash under the mattress) or low-return assets (a chicken in the backyard) and putting them into higher-yielding bank savings, even if the capital is being used outside of their community in the meantime. Since they are likely to be saving for a shorter- or medium-term purpose, they will get their money back from the bank when they are ready to invest their funds in their microenterprise or to build household assets. In the meantime, the saved bank balances might be used by enterprises to create employment, which serves to boost the income of local populations, in turn bolstering their savings opportunities. Access to finance thus benefits poor people in multiple direct and indirect ways, and a strict accounting of dollars saved and loaned to poor people does an inadequate job of capturing the net welfare effect. Of course, in countries where the banking system is controlled by a few powerful groups that use their banks primarily to fund themselves, this fear may be entirely founded. But this argues for banking sector reforms which prevent connected lending, rather than keeping the poor outside of the banking system.

Fear #2: Bank savings are not safe for poor people anyway.

Banks can and do go under and may perpetrate fraud. People may reveal their PIN codes to friends and family who then siphon off their accounts, or they may be assaulted at gunpoint and taken to an ATM to empty their accounts. These are real risks, which need to be addressed through proper prudential and consumer protection rules. We stand for safe savings, and our concern is to make sure that the banks serving the poor are no more risky than banks serving the affluent. But it strikes me as ironic when these concerns are voiced by richer people as an argument to keep poor people out of banking, when they themselves have not given up their bank accounts. Often these concerns express themselves more insidiously as an unwillingness to accept new or unfamiliar banking service delivery models geared to reach the poor, simply because the newness or unfamiliarity of these models brings to mind risks and consumer protection concerns.

We need to put together a compelling evidence base that shatters these myths, and we are working towards that with a range of research partners.

(Photo Caption: A line forms in front of a bank branch in Haiti)


Ignacio Mas

Consultant on Mobile Money

Charlotte Hill
January 07, 2011

You're absolutely right -- the idea that low-income people don't save is a completely misconception, one that's been proven wrong time and again by empirical evidence. Our research at EARN, the U.S.'s leading provider of microsavings accounts to low-income workers, proves that our clients save at the same rate as the average American, if not higher.

Thanks for writing such a comprehensive, thoughtful article!

Douglas McLain
January 07, 2011

I am a long time Kiva microfinance lender and agree that microsavings accounts would be preferable to microfinance loans. Saving one's own money is far better than being in debt. And the obvious way to do it globally is with cell phones. Cell phones are becoming ubiquitous around the world, including Africa.

The M-PESA system being developed by Safaricom in Kenya not only provides cell phone communications but also money transfer services and bank accounts to customers.

I think every cell phone buyer should automatically get a savings account and email address along with his cell phone.

M. Rikhana
January 09, 2011

it is an excellent observation, yet a challenge how this financial source of the poor to be productive without a fear. . . .

Nachiket Mor
January 30, 2011

This is an excellent post by Kendall and Mass of BMGF. I want to add to this debate by sharing a few more perspectives on this issue:

1. Very strictly speaking, if we carry currency notes, we are all financial included because the currency notes we carry are actually cheques issued by the Central Bank of our country on deposits placed with it. These deposits are unique in that they do not carry any interest (though it is indeed the case that during the Civil War the Southern Confederates did issue interest bearing currency notes: but in return are fully negotiable ("holder in due course" / “bearer” has good title). While I am not an expert on this issue, my understanding is that the principal reason why Central Banks do not offer to pay a premium upon redemption of a currency note is that that they implicitly assume full financial inclusion -- i.e., access to higher interest rate deposits. This is the same logic that the India Central Bank (the Reserve Bank of India) offers for imposing an implicit "tax" on individuals and companies that keep money in savings accounts and current accounts by repressing those interest rates.

2. If we accept the (above) position that as individuals who carry currency notes we do actually have the ultimate "no frills" bank / savings account (UNFSA), the relevant questions then shift from access to any kind of savings account to the features that are offered and how they compare to the UNFSA that we already have.

3. There is a concern that the full negotiability and liquidity of the UNFSA may itself be a problem on one very important count -- since it is fully negotiable it can be converted into a non-financial asset by giving into short-temptation (cup of tea) and not allow the holder to accumulate enough to build a house which arguably could be more "necessary" for her happiness. If indeed this is a valid concern then traditional savings and checking accounts may not be very useful because of the “easy debit” feature that they may offer. Professors Abhijit Banerjee and Sendhil Mullainathan presented a theoretical paper on this at IFMR ( a little while ago on this issue and the empirical work done at the Centre for Microfinance at IFMR by Professors Abhijit Banerjee and Esher Duflo both suggest that this indeed could be a problem and that traditional microloans could actually be a savings device with the saver purchasing a “commitment service” by paying the interest rate to the lender. To many of us that have borrowed money to purchase a home so that we are “forced” to save this will be a familiar problem. Viewed from this perspective the hard line that is often drawn from savings and loans may be actually be a real one and they may actually be substitutes for each other with loans having features that make them a form of savings superior to UNFSA and ordinary savings accounts. One more feature of the loan is that since the final result of the “saving” is already with the client (the repaired or the new home or the study table) and all that she has to do is ensure that she now “saves” regularly by paying her instalments, if the local lender goes belly-up for any reason, her “savings” are still safe with her.

4. In countries like India where the only accounts on offer from banks are low-interest rate savings accounts it is possible that when one subtracts the transactions costs associated with even local level banking the nominal rate of return obtainable on UNFSAs is higher than from traditional savings accounts offered by local financial institutions. There is the added fear that if the local financial institution (or Self-Help Group or a Village Level Savings and Loans Association) is less secure than the Government of India perhaps a relatively high risk premium needs to be subtracted from that savings account as well making the UNFSA a superior alternative. It is not surprising to me therefore that attempts by the Indian banks to offer regular bank accounts have not met with much success in terms of savings account balances in them.

For all these reasons I would argue that promoting access to minimalist savings accounts with regular banks without any other services or promoting Village Level Savings Associations (VLSA) or Self-Help Groups that take in savings or cooperative credit institutions that offer savings and loans may not necessarily offer a superior alternative to regular microloans. IFMR Trust ( has experimented with offering, in remote rural locations, Money Market Mutual Funds that invest in short-dated government securities with great success – the transactions costs are near zero, minimum investment amount is 2.5 cents, interest rates are several percentage points higher and since the money is transferred intra-day to the Mutual Fund and away from the Local Financial Institution there is no risk that the insolvency of the local financial institution will “eat into” the savings of the low-income household. For longer-term savings they have just launched the National Pension Scheme and are already signing up more than 100 clients a day with absolutely no marketing and for the rural middle class will soon launch an index based mutual fund. In my view these may be more interesting directions to explore further if one is looking for a product superior to the UNFSA.