New research is debunking myths about microfinance and showing how organizations can effectively address problems associated with poverty. Yale faculty Dean Karlan, Tony Sheldon, and Rodrigo Canales discuss the problems and the promise in the field of microfinance and the lessons for other kinds of social enterprise.
Q: Let's start with a definition. What is microfinance?
Tony Sheldon: Microfinance is financial services for poor and low-income communities, people who have been excluded from the mainstream financial system.
Dean Karlan: I would say that's the correct definition, but it's not the often-used one, which focuses only on small loans to poor individuals.
Sheldon: During 2005, the United Nations' "Year of Microcredit," there was a big push to expand the public understanding beyond microcredit because credit is often not the most important tool for poor households. But it's so much easier to talk about microcredit than the complexities of microfinance that the two have become blurred.
Q: Could you talk about the excitement around the idea of giving loans to poor people? Why have people seen that as a powerful tool? Why is it something so many people have invested in? And what are the main criticisms?
Rodrigo Canales: Part of what makes microfinance, or microcredit, rather, so appealing is this idea that we allow poor people to work themselves out of poverty. We loan them the little bit that gets them going; then they do it themselves. I think it's an especially American narrative.
When I started doing fieldwork in microfinance, I remember one of the most powerful things for those receiving loans was that somebody saw these poor individuals as creditworthy. All of a sudden they're thinking, if this organization sees me as creditworthy, maybe I'm not seeing something in myself. That has a big impact.
It used to be that when you visited clients' homes or businesses in Mexico, you would find the original letter from the bank approving the first loan—they would frame it. I think that's one of the things that microfinance should do, when it's operating at its best. It should help create new expectations for destitute populations, expectations that they can live up to. I haven't seen that for a while.
When you look at what's actually happening on the ground, there are a lot of tensions that you cannot get rid of. The notion that you can help poor people without having to engage in important trade-offs is very appealing, but a lot of it ends up being false. We have made an assumption in microfinance that profitability is not at odds with having an impact, but in many cases it has to be. In many places it's very expensive to provide microcredit, so the interest rates that you have to charge in order to get the sustainable machine going end up negating a lot of the reasons why you even started doing it in the first place.
Sheldon: Microfinance has, in some ways, been more successful than a lot of people ever thought it could be. Before the term "social enterprise" had been coined, microfinance was the first field where the end clients were the very poor and the business model aimed to both be profitable and have a major social impact.
The idea was that by being profitable, you create a scalable institution that would be there in the long term, could borrow or attract investment, and would not be reliant on either the whims or the limited capital of donors. Microfinance has been grappling with those three key factors: financial sustainability, social impact, and significant scale, for 30 years. And those questions have been transposed onto social enterprise, more broadly.
Canales: We think more scale equals more impact, but we made up that equals sign in our minds. More scale does not necessarily mean more impact. Scale is going to mean lower cost. Lower cost does allow you to reach more people. But if you have a lower cost in your business model, you cannot provide more costly services. Then you've constrained your business model in a way that if there's a population that requires a more costly service, you're opting out of that.
It's a legitimate decision. But you have to be clear. And for the most part, microfinance companies haven't been clear about opting out of this entire population or service because of a decision we've made in our business model.
Sheldon: There are many who would still claim that the larger you are, the more profitable, and the bigger the impact. But I'd argue there are unintended consequences.
Tensions and trade-offs have erupted over the last several years as microcredit has become profitable and attracted private investors. We're not talking about social investors or foundations; we're talking about hedge funds and investment banks who want to buy stock, because where else can you get a 50% return on equity? But that means the nature of microfinance changes—who is attracted to do it, and who is attracted to invest in it—and that has consequences for the end client.
There's been a crisis of conscience within the microfinance community: how do we own up to our role in creating this, manage it now, and figure out a way forward.
Q: What are some of the specific issues being addressed?
Karlan: The question of who to reach. Despite the rhetoric, microfinance traditionally is not reaching the poorest of the poor. There are some exceptions, but those are just that, exceptions to the rule.
The question is, why? Is it a matter of cost? Micro-lenders or microfinance institutions are not willing to go that far downscale because the loan sizes and the savings amounts get to be so small. In most situations, working with the poorest, we also see unwillingness among borrowers to even participate—when people really have absolutely nothing, there's a fear of the formal institution. That's particularly true when there's a preference for those who are part of a group loan where people already have some sort of enterprise going on. The poorest people are individuals that don't have an activity that would even count as a microenterprise.
Tony and I are working together on a series of randomized trials in seven places around the world to evaluate the impact of a program that works with the people who are truly at the very bottom in any sort of poverty ranking.
They receive a package of services that provides them training and savings accounts—common components of microfinance. But it also provides them perhaps the most important component: a transfer of assets to start a livelihood. It might be something like four goats. The program also provides consumption support—bags of rice—in order to make sure that they don't have immediate incentives to turn around and sell the goats.
Some people thought of these projects as a way of preparing participants to graduate into microcredit. But the measure that we care about is whether this produces a safety net. Does it sustain itself afterwards or do they fall back? That's the question we're looking at now.
Q: Is the for-profit or nonprofit approach more effective? Does the size of the organization play in?
Sheldon: I wouldn't say it's for-profit or nonprofit; I think it's a focus on the client or a focus on the institution. There are nonprofit organizations that are extremely profitable, and there are for-profit organizations that are barely profitable because they're focused on the client.
Likewise, there are very large programs which I would say take advantage of their clients, and there are very large programs that are totally devoted to their clients. When there is a choice between boosting profitability or expanding services, they will choose expanding services. It's really a philosophical approach.
The idea of microfinance as a panacea is such a problem, because if you look at the websites of the different organizations, you would have no idea of their underlying philosophical approach. There will always be wonderful stories about Maria the Baker or Raj the Rickshaw Puller. And I'm sure they're true. But what's underneath that? And what really is motivating and entering into the decision-making process of the institutions and of their funders, whether it be through grants, loans, or equity investments?
Canales: This is where Dean's work is so revolutionary and so important. Especially at the beginning, microfinance organizations used to promote and measure themselves by looking only at Maria the Baker and Raj the Rickshaw Puller, meaning they would look at people they gave loans to who succeeded. That would show that they had huge impact. But they wouldn't look at people who, for example, had taken on a loan, didn't do well, and dropped out of the program. There was an inherent selection effect.
The first wave of randomized controlled trials in microfinance helped figure out the impact on everybody who receives the treatment, regardless of how they did. And what's started to come out is a more complex story. And, in some cases, the impacts can be very negative.
It raised questions about how we mitigate the potential negative impacts of giving loans to people who haven't had loans before. What do you do if they get in trouble because the business is not working out, especially when they have the pressure of the rest of the group that is expecting them to pay? That can lead them to make really bad decisions for themselves.
Q: Dean, can you describe your work and the key findings?
Karlan: I've worked on a series of studies aimed at questions of how the money is used and the impact.
Of the completed studies, two look at traditional microenterprise lending organizations—one in India and one in the Philippines. Both were randomized trials, and both found that there is no aggregate increase in household income or consumption.
We did discover some important patterns. For instance, in India, we saw an increase in business starts for households that had no business to begin with. If they already had a business, though, we didn't see any increase in what they were doing. There was definitely a ceiling above which they just did not excel. Microenterprises are not growing into the small and medium enterprises that are engines for growth.
In the Philippines, we don't see investment in enterprise. They actually consolidate their enterprises, which I tend to think of as a good thing, if they are diversifying their activities in order to manage risk. This tells us that credit is helping to manage risk. We also see that they buy less insurance. If the credit is basically is acting as an insurance device, well, then it replaces the insurance product.
The third published study was done in South Africa. Rather than traditional microcredit, this looked at loans to salaried individuals. The loans had 200% annualized interest; it looks and smells like payday loans in America. Something bad was happening in people's lives and they needed credit. Those who got access to credit were more likely to keep their jobs a year later.
One thing we wanted to see was whether the loans made some people better off but left others stuck in a nasty debt trap. And we didn't see any evidence of debt traps from credit scores. We also looked at people who had experienced hunger in the 30 days prior to the final survey. That percentage went down, not up, presumably a direct result of poor people having jobs.
Now, I'm not saying that the loans created more jobs in the economy, but access to credit at the household level allowed people to smooth shocks that they otherwise could not smooth.
It goes to this point: We need to get away from thinking about microenterprise lending as the model and just recognize that individuals that are in low-income households, whether they have salaried jobs or microenterprises, have many things in their lives they have to cope with. For some, insurance might be the best device. For some, saving might be the best way. And for some of them, if they don't have access to savings or insurance, then a loan is better than nothing.
Sheldon: For people who live on the edge of poverty, it's a dynamic process. And to say "we have brought x million people out of poverty" is misleading and deeply inaccurate, because it's really more a question of resilience. How vulnerable are these people, still? Are they able to sustain a livelihood? Are they able to cover their school fees or doctor's bill or funeral expenses? Has microfinance helped them be more resilient?
Even if the household's economic pie is not bigger at the end, if they manage to survive more stably, that's an enormous accomplishment. And that's what we have to look at.
Karlan: I think it is important to also remember the fact that if there is a school fee that has to get paid, and if there's a way of getting people to save up rather than borrow down, there's no doubt that we would prefer the savings from a social-planner perspective. If they're saving up, they have a lot more money for consumption, because they're not paying interest. The challenge is, how do we do that?
We have a project in Ghana right now that is the best example of how little can we do and still generate more savings. All we did was give people who already have a savings account a second account labeled "savings." It just had a name on it, and their savings went up 30%.
The thing holding people back from saving toward goals is not always the obvious—that they don't have the money. If they had more money, they could save more, but, given that, are there things we could do to increase savings? If giving people a labeled account increases savings, that tells us that there's a lot more that we could be doing.
The second thing I wanted to mention is despite the fact that most of the world's poor are in rural areas, most microfinance has been focused on urban areas. The types of lending and savings models that you need in rural areas are very different. We did one experiment in Ghana on rainfall insurance, which has some striking results on whether or not credit constraints are the issue for poor farmers. We took small okra farmers, who are under-investing—they could be putting in fertilizer, but they're not. They could be farming land, but it's left fallow.
To figure out why, we did two things. For some, we gave them money. For others, we gave them rainfall insurance, just the promise of money if there's bad rainfall. It they're credit constrained, getting the rainfall insurance is nice but it's not going to help them make a bigger investment, because there's no money attached to it. Likewise, if you just give them money, but it's the risk of rainfall that's preventing them from making an investment, then they're not going to invest the money you gave them.
It turns out that in both situations, they invest. That's a striking result and tells us that the issue is not simply credit constraints. There's a risk factor holding people back and scaring them away from making investments.
Canales: We're finding that the mechanisms driving the results in microfinance are very different from the narrative we've been telling. This is not about us creating an army of micro-entrepreneurs who are creating businesses and building ladders out of poverty. That's not what's happening. But people are able to manage the risks and vulnerabilities of being poor in a more effective way. We're giving them a very blunt tool, and they use it as best they can.
What's emerging from all these studies is that the tool we're giving them—credit—is not the most effective one. It has a lot of drawbacks; it creates a lot of unnecessary complexities for them. You can say, maybe it's the simplest tool and we should keep on using it. But the problem is sometimes it puts them in a situation where they're making decisions that are actually not helping them.
There have been some famous microfinance people who said, my grandchildren will have to go to a dictionary to see what poverty was, because through microfinance, we've solved poverty. That's a really dangerous thing to be out there saying, because we haven't, and the trade-offs are huge. We need to be much more nuanced and careful about what we're doing to make decisions on those trade-offs in a wiser way.
Sheldon: If you think about it from a management perspective, we figured out a business model that can be profitable, have scale, and assume impact because people are coming back for more loans. They're repaying them. But if you look deeper, the impact is questionable.
If you then say, all right, let's look at doing insurance and savings—those are more complex to design, implement, offer, and scale up. How do you think about an organizational model that can do that and still be sustainable? That's what people are grappling with now. And it's what a lot of other groups are resisting, because it's much harder. They say, it took us 20 years to figure out how to do microcredit well. Let's keep doing that, rather than rethinking the organizational design and service delivery and who our clients are.
Q: Does microfinance as a pioneering social enterprise have something to say about other areas of social enterprise and what the challenges are?
Sheldon: I think microfinance is mostly looked to as the success story that sustainable energy, healthcare, education, all the newer realms of social enterprise aspire towards. But to me it is also a warning sign of what happens when you're so-called successful.
I think some of the misleading metrics that have been promulgated in microfinance are now being applied much more broadly in other spheres of social enterprise. For example, the rationale of the way in which you keep profitability and impact running in parallel is scale, because the more people you reach, the more impact you have, the more efficient and profitable you are. And so scale is now what everybody is looking to optimize.
The danger is, if you're optimizing scale, is you are not optimizing something else. You are not optimizing for the depth or complexity of services delivered to the end clients. You can make that choice, but you shouldn't fool yourself that you're not making a choice.
Solar energy, in particular, is a great example of that. If you look at India, there are a lot of companies that have something that sort of works better than not for a lot of different households, and the goal is to push the product out the door and maximize the number of installations.
There are other companies, including one that we've done some work with called SELCO, that look at each household and what's the appropriate mix of technologies and non-technological solutions (such as putting in skylights) that will serve that household's needs. They've grown more slowly at first, but have proven more successful in the long run. SELCO has over 100,000 households after 15 years. But their focus is always on the clients. And anything else—scale, profitability—comes after that.
I think the lesson of microfinance for social enterprise more broadly is, if you want to stay true to the mission, you have to stay true to the mission, and that has implications for the level of profitability you can reach. I think you can be profitable and genuinely serve your social mission. But to think you don't have trade-offs is a fantasy. And to think you don't have to face those trade-offs on a daily basis is a dangerous and lulling frame of mind.
Karlan: Maybe one other thing we have to remember about social enterprise is choosing the right metrics. If we use the metrics of how many people we reach, and whether they coming back and buying more, and if we are making money—and let's even add in, are whether we are reaching the very, very poor. If those are our metrics, then the two most amazingly impactful enterprises out there would likely be alcohol and tobacco. They certainly meet the criteria: They've got scale. People come back for more. It's profitable. It reaches everyone—it's nondiscriminatory.
You also want to know, what is the impact on your clients' ability to have more consumption, earn a stable income, to live longer?
Sheldon: Anybody's free to call what they're doing microfinance, but if you're going to talk about a commitment to poverty alleviation, then you have to actually collect client-level data on poverty level, education, and health of clients as they come in, and what changes over time. You cannot talk about being a transformational force in your clients' lives unless you have reliable data and you're willing to be transparent. So there is another bar that's being set up in microfinance now.
That is a good lesson for social enterprise more broadly, but it's an expense. It's complicated. It means you're not meeting, maybe, some of the goals you said you were. But it starts to set up a firmer standard to talk about the impact side, and not just scale and profitability as proxies for impact. That, I think, is very constructive.
Canales: I have nothing against starting any company and having people invest in it because it's profitable. But some investors are willing to pay an extra cost or lose a little bit of return on an investment that they think is socially responsible. Having the label of a social enterprise buys you something in the marketplace. It buys you leeway. It buys you more capital. I worry that a lot of organizations out there are saying they're social enterprises when, in fact, they're not.
Some are doing it naively. They don't realize that what they're doing is not a social enterprise, because they're not collecting the data they should. But some of them are doing this strategically. And they're really just creating this amazingly profitable business, but because they call themselves a social enterprise, they get to lure in money and they get to get away with some things that otherwise they wouldn't.
I care about this label and making sure that we're using it properly. And that's where identifying the trade-offs and identifying how to measure the kinds of impacts that we care about so that we're clear about what trade-offs we're making becomes really important, because we don't want to give a free ride to people who are not using the labels properly.