Print
Share
Is Microfinance A Schumpeterian Dead End?
May 15, 2010
Last month, Lant Pritchett, the invariably instructive scholar of development, had to sit twice through my spiel about the different notions of “development” I employ in my book to evaluate microfinance and my curiosity about extending this analytical thread beyond microfinance. Lant serves on both our Advisory Group, a group of leading academics, and our Board of Directors (ex officio).
The thing bothered him. If you’ve ever seen him in action, you know what I mean, and I mean that in the best way. I had a good but truncated conversation with him during a break in the Board meeting. I’ll represent his skepticism, or what I gleaned from it, as well as I can.
First, the whole development as freedom riff rubbed him wrong. I think he sees the emphasis on outcomes such as health and education as ends of development as having led to muddy thinking: the Millennium Development Goals, the idea that development can be bought with more aid ($50 for each child in school, say), and the obscuring of the importance of industrialization in reducing poverty. I reply (or replied) by arguing with none of that, but saying that an intervention that is giving millions of people an increment of control over their financial lives is doing something right, not bad by the standards of foreign aid. Perhaps he would say that he holds aid to a higher standard: it is wasting its potential if it does not contribute to economic transformation—industrialization—which promises to not merely ameliorate poverty but reduce it. Of course, accelerating transformation with aid is easier said than done.
Lant also disputed my characterization of microfinance as a Schumpeterian success, an example of aid building a thriving, disruptive industry that enriches the institutional fabric of nations. Rather, I think he argued, microfinance should be seen as an unfortunate work-around for the failure of mainstream financial systems to serve the poor. It’s like the private water industry in New Delhi that sells water to slumdwellers at far higher prices than the rich pay for piped supply. I said, “So it’s a Schumpeterian dead end?” He said, “Yes!” He told a story of visiting a self-help group (an example of a distinctively Indian microfinance) while working for the World Bank. After he and other visitors had finished asking questions of the members, all women, someone asked them if they had questions too. One member asked, “So how do self-help groups work where you come from?” The answer: we don’t have them.
It’s an interesting idea, microfinance as a Schumpeterian dead end. It strikes me that the claim imposes a pretty high burden of evidence though. Economic developments often takes circuitous routes. I recall Jane Jacobs’s thumbnail history of Detroit, I think in The Economy of Cities (I can’t check my bookshelf because I am travelling). It began as a tiny copper mining town; then moved into flour milling; then, using its accumulated expertise in machinery, boat repair (it being on the shore, connected to America’s internal maritime transport network); then boat manufacturer; then, in time, almost inevitably, the hub of the American auto industry. Only when it became a single-industry town, seemingly extraordinarily successful, did it economically stagnate. So the question is, how can you tell with reasonable confidence whether a new industry represents a dead end? Several big financial institutions today—Bank of America, Metropolitan Life, the Prudential (in the U.K. and its namesake in the U.S.)—started by serving the poor, arguably the microfinance institutions of their day. It’s not clear to me that work-around-for-deviation-from-the-ideal is a useful criterion for Schumpeterian dead end. Maybe Lant or others can chime in to continue the debate.
Update May 21: I just returned from Kenya, where I saw M-PESA, the mobile phone–based money transfer system with 9 million customers. It began as an idea: to use phones as an add-on to microcredit. Thus did a great new business start with a spark from an old one. More to come.
17 Comments on “Is Microfinance A Schumpeterian Dead End?”
Post a Comment
We value frank and constructive exchanges and encourage you to use your real name in your comments.
May 16th, 2010 at 8:00 am
We may not have SHGs per se in the west today but the emergence of credit unions about 150 years ago perhaps sprang from a similar impulse.
May 16th, 2010 at 9:26 am
David
You are to be congratulated for posting something, from a major world authority on development economics, that challenges the very fundamentals of microfinance. Professor Pritchett is dead on the mark here: the issue of ‘bottom-up’ industrialisation is crucially important to sustainable development and poverty reduction. This is especially the case when one looks at the great success of China and Vietnam of late, both of which rapidly developed and reduced poverty because they were able to promote a rapid ‘bottom-up’ rural industrialisation trajectory through a local financial system miles removed from what is understood as either Grameen-style subsidised or commercialised microfinance. Yet though Professor Pritchett clearly ‘gets it’, the fact that microfinance blocks such an important dynamic process is rarely understood – still less advertised – in the new world of microfinance.
The second point made by Professor Pritchett about Schumpeterian ‘creative destruction’ is also hugely important. The world of microfinance is almost entirely devoid of such important economic development dynamics, meaning far too many local communities now saturated with microfinance have been stopped in their tracks from developing and so eventually escaping poverty. They are caught instead in a very powerful, and growing, ‘microfinance poverty trap’. As I write in my forthcoming book (Chapter 4):
“It is important to note that, in general, displacement effects [when new microfinance-induced microenterprise capacities and employment simply displace existing microenterprise capacities and employment to produce no net impact, if not negative impact, in the local community overall - MB] are relatively much less problematic with regard to larger enterprise structures, such as SMEs. This is because SMEs typically locate a much higher percentage of their customer base from outside the immediate locality. So while displacement effects still arise – a new SME getting started in New Delhi will take some business away from an incumbent SME operating in Mumbai – the pain inflicted is generally diffused over a wider geographical area than just the local community. Moreover, much of economic theory holds such a displacement process in high regard, because it constantly displaces other less productive enterprises. This is an important industrial upgrading process captured by Joseph Schumpeter in his notion of ‘creative destruction’. So, even if some SME closes down due to the establishment of more innovative competition elsewhere in the region or country, such forms of displacement might have a positive efficiency-enhancing effect at the national economy level. However, this positive impact is generally not experienced in the microenterprise sector. A microenterprise displacing another microenterprise in the locality will likely mean nothing more ‘transformational’ than one trader or café or mobile phone time seller taking business from, or forcing the closure of, their equally simple counterpart located in the same street or neighborhood.”
Of course, this point is linked to the first point made: microfinance produces nothing more than rafts of the very simplest, non-industrial and overwhelmingly temporary microenterprises, an outcome that overall produces no real ‘transformational power’ whatsoever at the local level. The high rate of turnover and displacement typically seen in most local economies involving microenterprises therefore achieves nothing at all in Schumpeterian terms compared to, say, the comings and goings of SMEs or larger firms; in fact, it egregiously wastes the community’s scarce resources. Especially when these valuable resources are mobilised from the poor themselves (through savings, remittances, etc) this is a quite dramatically destructive outcome for the poor.
The comments of Professor Pritchett strike right at the heart of the microfinance paradigm. Thank you for sharing them with us. And three cheers to Professor Pritchett for expressing his thoughts to someone willing to listen and reflect upon them!
Milford
May 16th, 2010 at 10:24 pm
Thanks for the very interesting post David. While intellectually interesting to debate, are we not at too early a stage to say whether microfinance is the Deteroit of the 1950s, 1990s, or Detroit at all? There are creative forces and destructive forces at play, and we will continue to debate which is more important. I think the power of your book, David, is to say authoritatively that after looking at all the evidence we just don’t know.
May 17th, 2010 at 10:02 am
I agree with Pritchett’s comment that microfinance is an “unfortunate work-around”. It is a work-around in societies that lack the rules required to create jobs and income opportunities, in order to put food on the table.
A work-around is something we do when the preferred option is not available. So we could test pretty easily whether this is indeed the case by asking microfinance participants whether, given the choice, they would rather be doing what they are doing or working in a paid job. I’ll bet most people would prefer taking a job working for someone else.
Microfinance might be a useful safety net, but it is surely a second best solution to having in place rules and structures that reliably encourage business development and job creation.
May 17th, 2010 at 11:52 am
Neil Saravanamuttoo: even a family in which one member has a garment factory job that pays say $28/month, as is the case in one Banlgadeshi family showcased in Portfolios of the Poor, is likely to cobble together a wide variety of sources to raise money for major investments, emergencies or life cycle events – whether or not microfinance orgs are in the area. They are likely to have a strong need not only for micro-loans if available, but for a safe place to keep savings, since they are also likely to save a surprisingly large portion of their income. Microfinance may be a “workaround,” but it remains a highly useful one for the several decades it may take a developing country to get from the point of being able to offer factory jobs to a significant portion of the population to the much later point of having a banking system able to offer core financial services to the bulk of the population.
May 18th, 2010 at 5:10 pm
Thanks for sharing Lant’s skepticism on this. I’ve had similar conversations with him on COD Aid and other topics. I think he’s fundamentally right (as I interpret him) about the key issue being how social institutions transform rather than whether a particular beneficiary improves his or her condition.
In this more historical light, microfinance could be a dead-end if it actually displaces other more effective financial institutional development; could be an inefficient way to progress if it is a work-around in a second-best world; but it could also be the kernel of some new way of piecing together financial institutions that could be hugely efficient but just doesn’t look like it today or when compared to the way the Germans, British and Americans have conceived finance.
This debate extends well beyond Microfinance. I remember debating whether “social investment funds” – which were all the rage in Latin America in the 1990s – were work-arounds for dysfunctional public ministries and weak municipal governments. By supporting these funds, perhaps, we were actually stunting appropriate public institutional development?
The same issue is central to (but rarely discussed) in terms of health insurance initiatives. I am very skeptical of programs that promote “community-based” or small-scale health insurance for a number of reasons which I won’t go into in detail. But the broader question really comes down to one of institutional development. Regardless of how inefficient or highly subsidized such schemes might be, people who favor the idea of promoting small-scale health insurance schemes often argue that most Western European universal insurance systems evolved out of small-scale voluntary sickness funds or mutuelles and conclude that this is a stepping stone toward well-developed health insurance institutions in the future. Might be. But I also can imagine that – having seen the experience of Western Europe and Latin America – we’ve learned a thing or two about health insurance and could help low-income countries leapfrog a few steps.
On these kinds of historical institutional questions, I’m not sure how we make the right judgments or offer the right advice because it is hard to see very far into the future when the pathways are so contingent.
May 19th, 2010 at 3:04 pm
What I worry about development is that there are two ontologically different categories to which the word is commonly applied. One is an individual human being for whom when his/her well-being is higher we say they have more “development” and when we add people up we say this group of people have more “development” than another group of people. In this respect the HDI is one of many indicators of “human development” and no one defends GDP per capita or any narrow measure as being the ultimate metric and “development as freedom” I suppose is a good a starting rhetoric for groping towards measures of this as any other.
But tadpoles become frogs through development, acorns become oaks through development, that is it generally refers to a dynamic process in which things do not change their fundamental nature. In this sense most of “development” is really about some larger aggregate which casually some might call a “country.” There is something to the “development” of a “country” (some social aggregation) that is more than the adding up of the well-being of the people in it and this development has multiple strands (of which I identify at least four, economic, political, administrative, and social). The old fashioned view was that the dynamic and systematic improvement in the economic,political, administrative and social capabilities of “countries” was what led causally to the betterment of the development of its people measured as well-being. But the two are not the same. One can, as external actors, do things that better the “development” measured as human well-being that does not lead to “development” of the “country.”
So one question might be, “given that person X lives in a ‘country’ Z that has failed economically, politically, administratively, and socially and that nothing can be done about those processes, what can be done to improve the well-being of person X (and hence improve ‘development’?”
I just wish we had a sufficiently refined vocabulary to sharply distinguish the answers to that question from the answer to the question “how can development (in the sense of the larger economic/political/administrative/social transformations of societies) be promoted?”
There are many things that are “development” promoting in the first sense and not in the second sense, either in the sense of being neutral in the long run promotion of development or perhaps even negative. On the other hand, there are also things that are “development” in both senses–they both improve human well-being and reinforce processes that significantly improve the long run economy/polity/administration/social relations.
Humanitarian food aid is something that everyone agrees is, in some circumstances and done well, development in the first sense (raising human well-being above the disastrously low levels failures or crisis would have let it be) but not the second (just how neutral to negative it is to the second sense is the debate).
So, some “work-arounds” of failed development (in the second sense) may both help development (in the first sense) and unleash dynamics that ultimately lead to more development in the second sense. These are very good work-arounds. Others may just be good work-arounds around failed devleopment in the second sense that raise development in the first sense (by a modest amount usually) but are neutral for development in the second sense. I think private water sellers from the tanker trucks are a pretty good example of the latter.
Now, I have nothing in particular against those kinds of things–they are usually better than the bad alternatives–and even nothing so terribly strong against calling them “development”–as long as it does create confusion that they promote development in the second sense or worse, make people think that these are a complete substitute for development in the second sense–a society acquiring a productive economy, a responsive polity, high administrative capability, and just and fair social treatment of its members.
As for what “micro-finance” is on this scale of development in the first or second sense and in what mix, I have never taken a public stance as it is not an area in which I have done any research or carefully considered the evidence. My gut instinct and from what I have seen, I argued with David privately that it seems like development in the first sense (and no amount of refining the first sense with different measures, etc. turns it into the second sense as they are about ontologically different categories) and just not that important in the second sense one way or another. Certainly not necessary (too many countries have had development in the second sense without it), probably not sufficient, probably not actively harmful, probably just not that big a deal one way or another.
Since David is telling my stories, let me conclude with another one. I was living in India and discussing arrangements for household water supply with some development colleagues of mine. After about half an hour of pretty fruitless discussion I said, “let’s step back. tell me your long-run vision of the household water sector in India” They said “Our vision is that India meets the target that every household lives within half a kilometer of an improved water source capable of providing 40 liters of safe per person per day.” I said, “I see the problem. My vision of success is that every Indian can take a hot shower inside their own home.” The difference is that one can imagine meeting the first goal “programmatically” or with a series of “interventions” while the latter clearly requires endogenously functional systems. No one I know wants to have to go to a group meeting to take a hot shower. They want to turn the tap and it works.
(A much more minor, I would never use the word “industrialization” I would always refer to a high productivity economy, which may, or may not, involve industrialization in the conventional sense).
May 19th, 2010 at 11:30 pm
Hi David,
I really like the discussion you’ve started here and I wanted to throw in my two cents.
I’m working with a research group at the University of Kentucky, whose founder is profiled in your book. We’re trying to offer low cost, experimental micro insurance to farmers in the developing world against catastrophic weather risks.
I wanted to point out (and this may be an obtuse and academic point but I’ll say it anyway) that from an economic standpoint we fully expect that our insurance will actually DECREASE the expected incomes of poor farmers in the short term. That is the nature of all insurance…insurance is a very inefficient means of saving for low probability but terrible outcomes that may overwhelm your individual ability to cope. Insurance is inefficient in the sense that we all hope we never have to use it, and most of us don’t end up using it. It is one of the few financial products on which we are collectively happy to lose money.
Thanks to this inherent inefficiency, micro insurance is, at least in the short term, a net drain on individual development. Farmers who buy our micro insurance products actually LOSE money (we hope), as do most people who buy flood insurance in the US. If the Poverty Action Lab were to evaluate micro insurance using randomized evaluations of relatively short time horizons, we fully expect them to find that households who bought our insurance have lower incomes then those who did not.
And yet, I think there are very few development economists, developing world politicians, or individual poor people who would fail to see a thriving insurance industry as a sign of individual and collective development. Insurance lowers the variance of development outcomes (apologies for another nerdy academic idea) and that is unquestionably a good thing for individual poor people and for a country as a whole, even if many people lose a little money on the deal.
For economists, savings, credit, and insurance are three ways of moving resources across time. In some sense they do the same job of smoothing the potholes in our economic lives, not radically changing our lives for the better. This is true for individuals and true for countries.
If you really strip away all the hopes and good intentions that we’ve tied to the back of micro finance, I think its clear that micro finance as a whole is not really meant to help individuals in the short term. It is meant to help communities and countries allocate resources more efficiently, and to provide some small boost to individual incomes over the course of decades.
Bellyflopping 3-year long randomized trials is not a cause for a deep crisis of confidence, unless you’ve built micro finance up to be something it is not.
May 19th, 2010 at 11:31 pm
I think this post and the comment stream flowing from it illustrate another area with a similar problem of ontologically different categories being described with same term: blogging.
May 20th, 2010 at 4:46 am
The discussion is very enlightening. In a situation where a conventional financial market fails due to asymmetric information, missing collateral and so forth, microfinance is indeed only a second-best option. However, for some people it might be the only available alternative for the failed first-best solution. Hence, microfinance is second-best but might nonetheless improve the situation for the beneficiaries.
Leapfrogging or piggybacking in development processes might be charming. However, an evolution should be allowed for, to prevent overheating and overhastiness, which might in the end result in a failing developmental process. Microfinance is a bottom-up strategy that evolves on site.
My opinion is that microfinance can only help on a micro level for financial development. It is not the solution for all huge challenges of overall development in developing countries. Hence, it shouldn’t be claimed it would or blamed it wouldn’t. Being effective on a micro level will not prevent other development paths (SME, infrastructure, education, and so forth) to be successful. Nor will it substitute other developmental approaches.
May 20th, 2010 at 7:55 am
I agree with Joerg. We have to evaluate microfinance based on reasonable criteria of a financial service in often very harsh conditions. Microfinance might be second base but at least it offers many people the opportunity to absorb shocks in their financial flows and also to start up a business. These small scale loans ofcourse will not launch all borrowers out of poverty, the same as we dont expect that all loans in Europe to start up a business result in succesful businesses. The impact of microfinance most likely is higher when combined by other service providers with other services as BDS. the industrialization of a country is important but in most cases not evident, as for instance a huge country as Congo has a budget of a 1mln inhabitants city in Europe. Other investments do not contribute to local development but mainly to cowboy capitalism. Bottom up development strategies are vital, still if institutions are not in place things can be done by increasing the capabilities of the poor
May 20th, 2010 at 12:14 pm
An earlier posted mentioned that the discussion of microfinance and institutional finance development is similar to the debate about health insurance. In fact, it is also related to an ongoing “fight” in the public health sector. There is a subset of public health professionals emphasizing “health system strengthening” as opposed to interventions like those to battle AIDS, malaria, TB, etc. There is some evidence that piecemeal interventions actively undermine the health systems of developing countries, and are stop-gap measures at best. In one sense I can accept that, since people’s lives are being saved. But taking an institutional approach would almost certainly be more fruitful in the long run. The problem is that we may not know the best way to go about this type of institutional development, whereas we know the process for mere interventions. Providing malaria drugs is (relatively) simple. Building a functioning health system is a more useful goal, but is certainly harder.
May 21st, 2010 at 2:35 am
A very thoughtful discussion here, started off by a very thoughtful post.
I’m currently in Andhra Pradesh, evaluating a project that uses microfinance to extend water and sanitation in the slums. It’s co-ordinated by a local NGO and financed by an international foundation. Indeed, the standard here is quite low: to build a pit latrine and a water tap into every house (if there is enough space), and to operate a communal drinking water plant in every neighbourhood. This is miles away from a hot shower, and visions of hot showers aren’t even on the map. Of course, however, compared to the present state, this project (if successful) will represent a significant improvement.
But the women involved are spending a lot of time and effort, through SHG meetings and Slum-Level Federation meetings, on the project. It is also their money being spent. A first point here would be that we need to learn to value the time of the poor more like they do, to prevent situations where they have to “go to a group meeting to take a hot shower” (Pritchett). The poor are very busy ensuring their own survival all day; so while from a donor’s point of view it may look like a “cost-effective” means to have the poor self-organise and self-implement projects, from their perspective it is not.
In a world of second-bests, the question should not be “does this project work?” but: “does this project work better that any feasible alternative?”. To specify, in this case: “Is the time and money spent on this project more effective than time and money spent on an alternative?” Given, for instance, the fact that the municipal water supplier (through whom the new taps will be served) presently only manages to provide the existing public standpipes with half an hour of water every other day, would the time and money of the poor (and of the donors) not be better spent on improving the capacity of the municipal water supplier? Would it then not be able to provide the poor with the water and sanitation in their homes at a lower price in terms of time and money, than if they do this themselves?
These aren’t rhetorical questions, they are puzzles to be solved.
It seems to me that underlying many development interventions is the unshakable assumption (simply put) that “microfinance is the best way to do it”. My colleagues here are constantly looking for ways to expand microfinance into this, or that, sector. To apply microfinance to the water and sanitation sector is merely one outgrowth of this logic, where the starting point was “We need to bring microfinance into the water and sanitation problem”, not: “We need to improve the water and sanitation problem, so let’s find the best way to do it”.
Microfinance proponents always argue that it is a second-best solution, but it seems rather that it has taken the position of a first-choice, a default, the standard solution: “If there is a problem, let’s see how we can bring microfinance in.” The assumption that microfinance will “solve problems” (whether income poverty, gender relations, local development, lack of education, or water poverty) has always been a sketchy one. So, given the urgency of all these problems, the growing skepticism about microfinance should not be prompting the response that we need another 50 years to evaluate it; rather, we need another 50 alternatives to evaluate it against. Time to start exploring, again.
May 21st, 2010 at 9:03 am
Are the two types of development defined by Mr. Pritchett really “ontologically separate”? Even when countries are in the full flush of rapid economic development, it can still take decades before everyone has access to hot showers or conventional banks. Mr. Pritchett does acknowledge that some forms of development in his first sense — advancing the well-being of individuals — may feed development in the second sense. But to what extent do they catalyze it? Might “the old-fashioned view” — that “a dynamic and systematic improvement in the economic,political, administrative and social capabilities of “countries”…[leads] causally to the betterment of the development of its people measured as well-being” be in some measure reversed? Would not a service or activity that gives individuals more control over the conditions of their lives–making it likelier that they will educate their children or provide for their own retirement or improve their homes in ways that reduce risk of disease — make those individuals likelier to demand and create better governance, administration, etc.?
The degree to which microfinance fosters the benefits outlined above is of course open to question.
May 21st, 2010 at 12:35 pm
@ Grant. I agree that insurance may lead to a short term reduction in income, but this is not always the case in the short-to-medium-term. As you know, micro-insurance reduces risk. Lower risk means higher probability of access to credit for an enterprise. Credit that can in turn be used to expand the enterprise and thus increase income. For example, if a farmer insures her crop, a local credit org is MUCH more likely to lend to her. This credit, in turn, can be used for further investments in crop productivity (fertilizer, processing techniques, packaging for the value chain, etc.) So, I would argue, even in the short-medium term, increased income is likely if the package includes both insurance and credit.
May 22nd, 2010 at 11:10 am
@Andrew. Whether the two descriptions of development are ontologically different or not I don’t know but they are certainly independent and there are examples of where individual level development (eg increasing life expectancy) happens in the absence of any notable development at a country level and development at one level can regress while the other level moves forward. As it is development at the individual (following the wider development as freedom description of development) level that is the real goal (as countries are means of social organisation and not ends) the question that needs to be asked is whether higher level development is a necessary condition for individual development and if so what aspects of this macro development are necessary. If only we really knew.
May 26th, 2010 at 8:00 am
I’d like to get back to the hot shower problem for a moment. While living and working in India, I came across a watershed management project in Maharashtra run by NABARD. The engineers were doing all the usual development things—a tank here, a gully plug there, wells, walls, ditches, silting ponds and cisterns, you name it. There was a little bit of money left over in the project and the question was what to do with it. NABARD let the women decide. They chose to install hot water showers. It was a fragile-looking, jury rigged system of pipes, buckets, taps and spouts but it worked. The women said they looked forward to their shower after an especially grueling day of harvesting wheat. When I asked them what they would do next, a few replied, they wanted to tap the power lines for electricity and get a TV so they could catch up on bollywood movies. The whole discussion taught me that development is dreary and sometimes people would like something other than the bitter medicine we force feed them in the name of “development”. If we made interventions a little more fun, a little more thrilling, with a little less like drudgery, we might see better take up. Surely, we would observe a little more fun.