Is it a good idea to subsidize microfinance?

Microfinance is trendy. But is it just another development fad, or is there evidence that microfinance really helps to stimulate economic growth? Even if there is, should donors do more to support microfinance?

My take on this is that microfinance, provided on a commercial basis, is self-evidently good. But that does not mean that we should welcome the stampede to subsidize microfinance. There are not strong arguments – either in principle or from evidence – that this is the best way to use scarce resources to help poor countries to grow.

Here I look at six reasons why microfinance is no panacea.

This is the International Year of Microcredit, no less, and throughout the world ten thousand microfinance institutions service some 40 million customers. Grameen Bank of Bangladesh, the world’s largest and most successful MFI, serves more than five million clients.

Donors and foundations are climbing on the bandwagon.

  • On Friday, the founder of EBay, Pierre Omidyar and his wife Pamela gave $100 million in eBay stock to Tufts University to create a fund that will invest in international microfinance, or lending to people who are too poor to qualify for traditional loans.
  • On the same day, Bill Clinton launched a new microfinance fund based on $75m from several multinational companies, charities and government development agencies.
  • The Shell Foundation has announced $100m in micro loans.

We celebrated the blogswarm that supported Kiva. And here among the self-appointed experts in the blogosphere there is rare consensus between the Globalisation Institute, WorldChanging, NextBillion, Global, the PSD blog, Adventures in Ethical Consumerism, Dan Drezner, Cafe Hayek, Unitus, Atlas Blogged, and others.

So presumably we have good theory and evidence to suggest this is an effective way to help the poor? Err … not exactly.

Microfinance enables people to lift themselves out of poverty. It is bottom-up, empowering people to lead their own development instead of top-down development. Support for microfinance removes market failures that put barriers in the way of the poor. What’s not to like?

Let’s be clear: access for the poor to financial services is definitely a good thing. Access to credit can help households to increase their consumption, reduce risk and increase food security, reduce malnutrition, and empower women. (ref: Khandker)

But saying that commercially provided microfinance is good is not the same thing as saying that it is a good idea for donors, philanthropists or governments to subsidize microfinance. Here’s why.

First off, it is not clear that there is a market failure to correct.
On Friday, The Financial Times (who should know better) said this:

In practice, however, there is a market failure: capital does not flow to start-ups in poor countries. The returns are not high enough to compensate for the risk.

That isn’t a market failure. If the returns are not high enough to compensate for the risk, then the market is doing exactly what it should by excluding the poor from access to credit.

(An aside: the fact that some people are too poor to obtain goods or services that we wish they could afford – such as food, water, credit, health care, transport etc – is not a market failure in the economic sense.)

If the market incorrectly perceived the risk, or if there were government regulations or taxes that made credit unaffordable, then there would be a market failure, and there would be a case for policy intervention to correct it. But in this case, the best approach would be to correct the market failure directly. If governments have better information than markets about default risk, then they should publish it. If there are taxes that make credit unaffordable, they should be abolished. It is very unlikely that, if there is a market failure, the best approach is to provide subsidized loans.

Second, subsidies for microfinance suffer the same shortcomings as other industrial subsidies.

Most microfinance institutions get some or all of their funding from grants or loan guarantees from individuals, philanthropists, foundations, and governments and international institutions such as the World Bank.

When developing country governments subsidize businesses through export subsidies, import tariffs, public procurement, grants or tax breaks, they are roundly denounced as interventionist and corporatist. They are not praised for supporting bottom-up, market-led growth. It is economically inefficient to use public money to distort the market. So why are subsidies to the cost of capital for small businesses any different?

Third, these subsidies may stifle competition that would improve financial services for the poor.

The poor desperately need access to affordable credit and insurance markets and other financial services. They are most likely to get that as the financial services industry, through a process of competition, drives out costs and produces effective and affordable services that meet their customers’ needs.

If we have donors and governments supporting uneconomic microfinance institutions, that evolution of the financial services industry isn’t going to happen. (The obvious parallel is domestic firms behind a protectionist import tariff which have no incentive to become competitive in world markets). The unintended consequence of well-meaning support for the microfinance industry may be that it delays the very evolution of the industry that the poor most need.

Fourth, increased access to credit may not significantly improve productivity in the absence of complementary inputs.

Almost all studies of access to credit find that most loans taken by the poor, especially in the rural and informal economies, are NOT taken for the purpose of investment in businesses but to finance consumption. This is not a bad thing – the poor benefit from the insurance that access to credit provides; but it suggests that hope that microfinance will enable people to grow their way out of poverty may be too optimistic.

There are few really good empirical studies of the effect of access to microfinance on investment; but such as there are show that returns to access to credit depend crucially on access to other complementary production inputs such as seeds, irrigation, or transport. Without the presence of these other factors of production, access to credit has little or no impact.

(Another aside: even in cases in which the borrower is formally a woman, there is a great deal of evidence that women only rarely control the use of the loan.)

This fourth point is not an argument against microfinance (which, as you will recall, is unquestionably A Good Thing): but it is an argument that microfinance is a complement to, not an alternative to, all those other policies to improve the opportunities for the rural poor (many of which are despised as "top down" interventions by the most articulate advocates of microfinance).

Fifth, there is an opportunity cost to these subsidies.

The total subsidy to Grameen Bank is about $20 per person per year, or $0.22 per dollar-year borrowed. (Schreiner, 2003) It seems likely that this is good value for money in the sense that the recipients of the subsidized loans use them in ways that produce a surplus of more than the cost. But there are many development investments for which funds are not available (because aid resources are scarce) that would yield a positive return. The question is whether the return to subsidizing credit for the poor is the best possible investment compared to, for example, the purchase of vaccines, teacher training, building roads, R&D into more productive crops, or the cost of peace-keeping in conflict-ridden countries.

Measuring the impact of microfinance is hard to do. An individual who takes the initiative to access credit is likely to have above average entrepreneurial skills; and will typically have a higher income than someone who does not access microfinance. It is difficult statistically to separate the effect of this entrepreneurial trait from the impact of access to credit. As far as I know, there are no studies using fully random assignment to measure the effectiveness of microfinance schemes which would permit a comparison with other development interventions.

We do know that investment in these public goods – such as vaccines – produces huge benefits, far exceeding their costs. The returns to microfinance would have to be very large indeed to justify their position as the intervention of choice for well-meaning donors. And yet the best evidence we have is indifferent on the impact of access to credit. (ref IFPRI, 2000)

Sixth, subsidies for microfinance is no more "bottom up" or less interventionist than many other government and donor policies

Here is The Economist:

What makes microfinance such an appealing idea is that it offers “hope to many poor people of improving their own situations through their own efforts,” says Stanley Fischer, former chief economist of the World Bank and now governor of the Bank of Israel. That marks it out from other anti-poverty policies, such as international aid and debt forgiveness, which are essentially top-down rather than bottom-up and have a decidedly mixed record.

Eminent though Stan Fischer is, this does not bear a moment of reflection.

All economic growth will come about because of the efforts, investments, courage, luck and skills of individual people, working to improve their lives and those of their families. This process requires access to a range of complementary inputs, such as skills, market infrastructure, inputs for their production (eg seeds, tools, labour), access to credit, good health, information, and security.

When countries receive debt relief or aid, they can use those resources either to provide those goods and services which are needed to enable the private sector to have successful businesses (eg investing in security, education or health), or to help to reduce the obstacles that the government has erected (eg taxes).

Providing public subsidies for credit is no more "bottom up" than providing subsidies for road-building or education. Indeed, it is arguably more interventionist than many other policies, since it takes governments and donors from the business of ensuring efficient supply of public goods such as transport infrastructure or security (which would be undersupplied by the free market) into the business of subsidizing services that should be efficiently provided by the market.


Lest I be misunderstood, let me repeat that the poor need access to credit, insurance and other financial services. That they cannot afford these services demands policy action; in just the same way as the fact that they cannot afford to access food, education, health care, transport, communications, security, housing.

But it does not follow that well-meaning donors, whether rich philanthropists, governments or international institutions, should subsidize credit. If there are market failures, we should identify and correct them. If there are not, we should allow the market to work, and concentrate instead on good public policy: correcting market failures, ensuring the affordable supply of public goods, and creating an environment by which the private sector can invest and grow. It is not at all clear that subsidizing microfinance is a high priority for helping the developing world to grow its way to prosperity.

19 thoughts on “Is it a good idea to subsidize microfinance?”

  1. The interesting thing about this issue is that the free market types who oppose top down socialism seem in favour of top down capitalism.

    This strikes me as further evidence that we should not be thinking about left vs right, capitalism vs socialism, or market vs government…. rather top down institution building vs the facilitation of emergent outcomes. A subjectivist, cultural approach to development economist will help.

    Examples exist of available micro-finance being ignored by developing entrepreneurs. Instead of using macro institutions to flood the market, the institutions that these people are trying to create should be allowed to flourish.

    The best way we can support people to release their hidden capital is to let them make their own choices. Even though micro-financing has the air of micro policy, when it’s being advocated and stimulated by a macro organization it becomes a contradiction.

  2. What I would like to see is not so much microfuinance being subsidised, but it being made more efficient. And I think the internet can help with that.

    Owen replies: Right. My worry is that by subsidizing it, we make it less not more likely that the industry will become more efficient. That, after all, is what we think about subsidy and protection for most industries – firms don’t become competitive until and unless they face the chill wind of competition.

  3. It seems to me that there are two kinds of subsidy to microfinance. First, there is the donation of start-up capital to a fund which then becomes self-sustaining. Second, there is the lender who requires sustained subsidisation because its loans are not performing and it will go under without support. Aren’t your criticisms, which are well made, more directed at the second than the first kind?

  4. I agree that the FT’s comment on “risk” was a little silly. I’d also point out that The Economist survey made an interesting point that I hadn’t thought about at all. That the poor value the ability to have secure savings just as much as they do access to credit. It is the full panoply of banking and insurance services that they wish to have.

    Which leads to the “risk” comment. It isn’t as the FT says, risk which is the problem. It is overheads, the cost of assessing potential borrowers. If you want to lend someone $300 and it costs you $150 to evaluate the loan then your interest charges are going to have to be very high.

    That’s where I’d like to see the subsidies go, into the design and development of low cost methods of doing that work. Which of course requires funding people to go and try it out.

    Very much the same as the oft identified “investment gap” for small companies in advanced nations. It’s horribly difficult to find equity finance in the $10k to $500k area, for exactly the same reason. The overheads of investigating the plan mean that the return must be very high. It isn’t the risk, it’s the assessment of risk if you wish.

    Now this was something that we (partly) solved ourselves a few generations ago. Via the Friendly Societies and so on. Raising the funds to be lent from the savings of the local community, paying those who administer them something around the average local income and lending the money within that community.

    It’s at this point that the Economist survey is worth careful attention again. For those who are indeed trying to do this on a profitable basis find their costs hugely raised by banking regulations (in many places ) better suited to large scale finance.

  5. Tim

    The cost of assessing potential borrowers isn’t exactly an “overhead” (as you put it). It is the core of the business. Doing it well and cost-effectively is what makes a lending business succeed. If the “overheads” are too high, then this isn’t a profitable business selling to poor people.

    I agree that it would be good if it could be done cheaper. I also wish we could make cheaper tractors and computers. But I don’t think that government and philanthropic subsidies to make this cost go away is going to help us find a solution. Which other businesses whose costs are too high to be profitable in poor countries should we subsidize? How will the existence of firms that continue to operate despite being unprofitable create the space for potentially competitive firms to develop cost-effective techniques which would enable them to enter and operate profitably in these low unit value but large volume markets? The existence of inefficient incumbents living off subsidies is not generally thought to be the ideal way to create a vibrant private sector market.

    The banking regulation point is a good one. I didn’t have much faith in the Economist Survey, which seemed to me to be sketchy on facts, but if it is true that (in some countries) the costs are increased by unnecessary banking regulations then we should work hard to get those regulations changed. That is more economically efficient, cheaper and more sustainable than subsidizing the businesses that bear the burden of those costs.

    What is striking is how quickly the notion of “bottom up” development has gone out of the window. There is a need for judicious intervention to help financial services work better, and allow the poor to benefit from them, as there is to support so many other sectors of developing countries, from skills to seeds, from credit to transport, from communications to energy. I am not at all convinced by the sales pitch made by many advocates of microfinance that subsidizing microfinance is “bottom up” whereas subsidizing transport or training is “top down”. All these sectors are essential complements to enable a healthy private sector to emerge and prosper. My instinct is that governments (and donors) should focus, at least at first, on the markets in which public goods need to be provided, or market failures need correcting, before moving on to markets which the private sector should be able to sort out itself, if given suitable space to do so.


  6. Not quite my point…the FT said “risk” which is different from the cost of assessing that risk. In banking terms risk is usually taken to mean the likelihood of non-repayment while the staff required to assess that likelihood are regarded as an overhead.

    Bankers, accountants and economists do often use the same words with slightly different meanings.

    We can in fact make cheaper tractors and computers. By reducing or ignoring labour and environmental protections, for example.

    Still, the point that I got from the Economist survey was that micro-finance was just at a possible tipping point, going from being something that only worked with public (or philanthropic) subsidy to something that might actually be profitable on its own. Examples like ProFund and ProCredit.

    I regard methods of organisation as a technology just as much as light bulbs or solar panels are. So I’m not surprised that in their early days they require subsidy. It is, after all, a consistent criticism of free markets (and one I partially sign up to)that they do not provide sufficient incentives to invest in basic research.

    Whether Governments are in practice any better is an argument for another day.

  7. In countries where accessing government or providing most other services provided solely by government or correcting market failures is not possible or cannot be influenced by donors.. microfinance is a direct way to help many receivers and hope the receivers will help themselves. Ofcourse it shouldn’t be subsidized since that wouldn’t put the pressure on receivers to make the output of the loan most efficient.

  8. Owen,

    I’m not sure it’s accurate to characterize the Omidyar-Tufts Microfinance Fund as a microfinance subsidy. It was founded through an individual endowment from Pierre and Pam Omidyar, who are university alumni (not through the Omidyar Network), and is intended to kill two birds with one stone: funding microlending programs and generating income for the university. From the President’s statement on their website :

    We hope the Omidyar-Tufts Microfinance Fund will demonstrate that microfinance can be commercially viable, and in the process draw additional capital into this market. This fund clearly demonstrates that it is possible to do well by doing good. We expect Pierre’s innovative idea to empower millions of people in developing nations – and to generate significant income to Tufts as the microfinance investments produce returns. Fifty percent of the investment income from the Omidyar-Tufts Microfinance Fund will be available for Tufts to use in support of both our faculty and students. The remaining income will be reinvested in microfinance. We expect to use the returns generated by the Fund to create a loan forgiveness program for graduates pursuing careers in public service, to create a funded internship program for students working in the non-profit sector, and to support additional investments in financial aid. Faculty will also benefit from this program with additional support for their teaching and research.

    In terms of institutional arrangements: “The Omidyar-Tufts Microfinance Fund will be invested solely in microfinance initiatives. An independent supporting organization controlled by a Board of Trustees will have fiduciary responsibility for investing the funds with the expectation of risk-appropriate financial returns.“

    So at it’s heart, this is should be seen first and foremost as a major endowment gift to Omidyar alma mater with a creatively earmarked investment strategy rather than a microfinance initiative implemented through a university (which is also accurate, but represents the means rather than the end). However, it is a good indication that Omidyar firmly believes in both the development effectiveness and commercial viability of microfinance initiatives, and might be interested in funding an analysis of “best practices” to guide the investment of his endowment funds. But this the only charitable recipient (as opposed to commercial beneficiary) is Tufts.

  9. I think a lot of it is going to be wait and see.

    The sums you mention are relatively small compared
    to the sum of foreign aid and charity.

    Which is pretty small compared to the wealth of industrial nations.

    We do know a lot of top down aid goes to corrupt officials and waste. So simply getting money into the bottom is something. If they stimulate so economic activity it is a multiplier, and millions of small bets may prove better than a big one.

    I don’t think that the industrial countries have shown that useful credit tools develop for the poor. Interest rates at check cashing places and the like are often excessive. This is even truer in the third world. Many small vendors pay as much as 40% per month. This guarntees reduced incomes and little savings while the traditional feudal elite continues.

  10. Thank you for this great discussion. We started the weblog to provide candid information on what investment options, not subsidies, exist in microfinance. Currently, we list 67 microfinance investment funds, 13 of which provide a return on investment. You do a masterful job of telling the story of damaging subsidies. We try to tell the story of how to fix it. Thank you again, David Satterthwaite.

  11. Pingback: » Blog Archive » Microfinance brings hope

  12. I think microfinance should be viewed primarily as a charity rather than an instrument of regulating the economy. The fact that such instruments tend to have returns make them more effective (they get a bigger bang for your buck). And please let’s not forget that microfinance exists because top-down development may not be sustainable (especially in developing countries where no system of checks and balances exist to curb government corruption). 

  13. Just come across this good discussion that has been dormant for 11 months! Pleased to note that microfinance seems to be regarded by most contributors as ‘a good thing’ on balance. From my limited reserarches I am not convinced that microfinance is exactly subsidised – not at the point of delivery anyway (assuming a commercial interest rate on the loan). I acccept, however, that large corporate funds and donations can be considered subsidies.

    The surprising thing to me is how difficult it is for an individual or service club (such as Rotary) to invest in microfinance (rather than simply donate). Oikocredit seems to do a good job, but has no UK representation apparently. I’d expect considerable individual / service organisation microcredit funding to be forthcoming if investments of say €1000 or $1000 were easier to make.

  14. Oikocredit UK -Andy

    Oikocredit are focusing on building up 1st and 2nd tier MFI’s. When they reach a certain size , these MFI’s can cover their costs and the need for subsidies falls away. 

    I do not beleive that the market is going to satisfy the need for financial services of the rural poor in Africa, it is just too expensive to provide intially though when MFI’s reach a certain size it becomes more economic. A very exciting development in Afria is the advent of mobile phone banking which may radically lower the cost of delivering MF.

    Oikocredit have just received permission to sell their shares, which yield 2% p.a., in th UK. They are setting up their first UK support association with which I am involved. Anyone interested should contact.

  15. we are operating a credit lending unit in uganda east africa . however our major challenge is over whelming demand for credit services yet we have limieted capital. how would you help us access a capital boast.

  16. The Financial Times had it right. Microfinance can serve to solve a market failure. Theoretically, capital should flow to poorer entrepreneurs and less to richer entrepreneurs because the poor would yield greater returns, with the rich experiencing diminishing returns. If there was perfect information (as is assumed in a perfect market), the poor would experience an influx in capital and convergence. This doesn’t happen because of an information asymmetry (banks have a lack of information about the poor borrowers), along with other issues, such as the poor not being able to provide collateral. Information asymmetry is one reason for a market failure, and microfinance firms work to get to know their clients and correct, thus helping capital to flow to the poor.

    Read Aghion and Morduch, 2005, widely held as one of the best sources on explaining Microfinance.

  17. Pingback: Microfinance brings hope |

Leave a Reply

Your email address will not be published. Required fields are marked *