Supporting Entrepreneurship: What Works in Developing Economies?

ISOBEL COLEMAN: (In progress from source) -- initiatives and opportunities, including, in some depth, microfinance. Last year our focus was really on the opportunities around mobile technology and development. And this year our focus is very much on looking at ways to invest in women and improve their economic productivity and their earning potential.

And as part of that, we are doing this session today, which looks at some very interesting emerging research. We have another session in December that's going to look at global supply chains and how corporations and other entities, NGOs and governments and others are looking at ways of bringing women-owned businesses more into global supply chains as a way of really investing in a very high-leverage way in building women business owners and entrepreneurs. And then in the spring we will probably revisit this research that we're going to talk about today as it concludes and look in some depth at some of the findings.

But I'd like to introduce Noa Gimelli, who is with ExxonMobil and the director of the Women's Economic Opportunity Initiative. Thank you. (Applause.)

NOA GIMELLI (Director, Women's Economic Opportunity Initiative, ExxonMobil): Good afternoon, and welcome. As Isobel said, I'm Noa Gimelli and I direct the Women's Economic Opportunity Initiative at ExxonMobil. I want to thank you all for joining us for what I think will be a very interesting discussion on a very important topic.

I'd also like to give Isobel a very special thanks for hosting us today at the Council on Foreign Relations. We've been a partner of Isobel's over the last several years and really appreciate all of the work we've done with her.

So as I said, we're here to continue the discussion on creating greater economic opportunities for women. So as I begin, I think it's appropriate to acknowledge the progress that has been made over the years. We're fortunate to have Markus Goldstein here, who is with the World Bank. And as he can attest, the 2012 World Development Report shows that the lives of women around the world have improved with gains in basic rights, education, health and access to jobs.

But we all know that there is much more to be done, as evidenced by the stark fact that women represent a full 40 percent of the global workforce, yet earn just 10 percent of the income. And we in the private sector are one player in the mix of organizations working on these issues. We recognize that we can make a difference by lending our time and expertise in such areas as business skills, training and supply-chain management, as just a couple of examples.

And of course, we can also make a difference with our financial resources. We recently looked at all of the programs that we've contributed to over the last five years and discovered that we've contributed slightly less than a billion dollars on community investment programs, and we're just one company.

So in reviewing how we were going to spend these resources, a few years ago we made the decision to focus a portion of our strategic investment dollars into women's economic programs, and here's why. We typically enter into a country in partnership with the host government to help develop that country's energy resources. Many of the countries where we have operations have great development needs, and to be successful at doing our job, we must be considered a valued partner by contributing to economic development.

We think the data on investing in women, and its impact on economic development, specifically, is undeniable. We know that a mother's economic status is one of the best indicators of whether her children will complete their education and enjoy healthy, poverty-free lives. It was this multigenerational return that was the turning point for us in making the decision to invest in women.

And so today we fund a variety of programs and initiatives, including microfinance, skills training, mentoring, technology programs and, very importantly, research to understand the most effective interventions. While the evidence is clear about the multiplier impacts of investing in women, more work needs to be done to better understand which programs are most successful and how best to structure these programs to make them even more impactful.

As Mayra will discuss in just a minute, through our collaboration with the United Nations Foundation and 15 research teams, we're partnering on a study to identify and analyze solutions that have proven to advance women's economic productivity and improve incomes for women. Specifically, we are examining various interventions in four main categories: entrepreneurship, agriculture, wage employment and youth employment.

So to close, we see today's event as part of an essential ongoing discussion that will help us all learn from each other and hopefully will lead to smarter and more effective programs. I thank you all for taking part in this.

Turn it over to Isobel. (Applause.)

COLEMAN: So we're going to do this a little bit differently than we normally do on our meetings. We're going to do two meetings today; you're getting two for the price of one. I'm going to have just a brief conversation with Mayra, who is leading the research that Noa just referred to, and then after that Agnes and Markus are going to join us up here and talk in more detail about the actual studies that they're doing.

So Mayra, who is a longtime expert in gender and development with years of experience at organizations like the World Bank, the Inter- American Development Bank, Vital Voices and a range of other organizations, and is now a fellow at the U.N. -- the U.N. Foundation.

And Mayra, tell us a little bit about this research project. And my understanding is that you're really focused on looking at ways to improve women's economic potential and earnings potential in four areas -- agriculture, entrepreneurship, wage employment and youth employment.

Just -- why those areas? And why the focus of this research now?

MAYRA BUVINIC: Sure. First, you know, let me thank you and the council and this wonderfully gorgeous room -- (chuckles) -- inspires us all. (Chuckles.) And to thank ExxonMobil, obviously, because they really were the ones who approached the U.N. Foundation to do this work. And it was just -- it's -- you know, I was just very privileged to be there at the right time because I had just -- I had just finished my work at the World Bank, and as part of the work of the World Bank, I saw the bank -- and actually, Markus Goldstein is one of the authors -- put together the first World Development Report. The Bank does one every year, but this was the first one on gender equality.

And the report is very clear on what Noa was already saying, you know? Investing in women makes economic sense. It's smart economics. And the report also -- it's very clear on what are the determinants of inequalities. Why is it that women are less productive than men and earn less incomes than men? But really, that's where sort of -- but then, you know, that's what the report does. And a year prior, the Food and Agriculture Organization had done another report also focusing on gender equality, and again it explains why women in agriculture are not as productive as men in agriculture.

But then the next question is: What --

COLEMAN: What do you do about it?

BUVINIC: What do we do about it.

COLEMAN: Right.

BUVINIC: And there's what I think both reports -- you know, I'm -- and in part because the evidence is yet not there, sort of raise the question that there was the need, now, to provide an answer. So I think that this project just came at the right time. It -- we hope to provide an answer; it will not be a definite answer, but we're trying -- we are compiling all of the best evidence and rigorous evidence.

And we are looking at something that is really much more difficult, in a sense, to look at than failure. Everybody, you know, understand why is it that women are not as productive as men. Everyone understands what are the determinants of inequality. But here, we're turning around the question of saying, well, you know, what succeeds? What is it -- how is it that you turn it around and that you really -- with what kind of programs you improve women's productivity and earnings?

COLEMAN: And I know the research is still in development, but what have you found that's surprising so far?

BUVINIC: There may be -- you know, I'm -- it's -- I mean, if I can go sort of the -- an area that is incredibly important is the area that we're going to be talking today, which is entrepreneurship. And in entrepreneurship there has been a lot of very rigorous research done recently -- you know, a lot of experiments that have been so -- that are really -- you know, they provide very good evidence that can really make a case: This program, this is the effect it has.

And a couple of things that maybe are not as surprising, but they are -- if you are trying -- I think what we are trying to identify cost-effective solutions, because that's one thing that, you know, ExxonMobil really (trust ?) us with saying what -- you know, what works but also what is cost-effective. And there is not that much data on cost-effectiveness, but then, you know, the very simple argument is the simplest the intervention, you know, you assume that other things being equal, is probably going to be more cost-effective, is going to be less expensive. And I think simple interventions are also very, very important because of the lack of capacity in developing countries to implement (boutique ?) programs or to implement complex programs.

So, you know, having this in mind, one of the things that the research and entrepreneurship is very clearly telling us is just simple interventions work, but really not for the very poor women. For very poor women -- which, in a sense, it makes sense, but for very poor women, for instance, credit is not enough. An infusion of capital in -- through credit or through a grant is not enough. You really have to pare that with very high-quality business training or training. And if you do both, there is some amazing resource, one, of a BRAC program in Bangladesh that very poor women with an infusion of capital -- that was in terms of livestock -- and training really improve their productivities and earnings dramatically, more than 40 percent.

Now, another catch -- well, you know, and that is for the very poor. Now, if you go up the ladder in entrepreneurship to something like small and medium enterprises, there you see the training alone makes -- can make a difference and that, in fact, it can be cost- effective.

So, I mean, in a -- in a sense it's a simple -- you know, it's a very common-sensical solution, but it poses a quandary if you're trying to implement very simple programs in very poor communities.

COLEMAN: Know your customer.

BUVINIC: Yeah. Know your customer. And that's, I think, one of the things that we are going to try to do with this research. Solutions, you know, depend on whom you're doing the solution for and your context. And I think we all -- we often enough tend to just generalize and say, well, you know, this will serve everybody, and it does not.

The other very interesting thing about this research, and I'm sure that Markus perhaps will talk a bit more about it, is things work in terms of promoting entrepreneurship for young women. They work sometimes better than for older women. And I think, you know -- and here is something that I had not said about this first research -- is a very interesting thing, though, with this very ultra-poor women, that the -- what the programs have done is they have selected women where husbands in the family are not the main income providers. So a lot of these women are women-headed households. You know, it obviously -- it immediately --

COLEMAN: Self-selects, yeah.

BUVINIC: -- self-selects, but it tells you that in a household where the woman is the main economic provider, maybe there are not that many intra-household conflicts about how you spend your money and how you -- what you do with your money. So women can be -- you know, can be more productive.

And I have the sense that with young women, maybe that is also the case. You know, younger women do not have intra-household conflicts or trade-offs that women have to make when they have children and husbands and there's more conflict.

COLEMAN: Interesting.

BUVINIC: So those are two --

COLEMAN: Well, great. Thank you.

So I think we'll bring Markus and Agnes up now. Thank you, Mayra. And Mayra will be available for questions, too, when we're finished. Thank you.

As you heard, Markus Goldstein is a development economist with the World Bank and one of the lead authors on the --

MS. : She let you pass.

COLEMAN: -- 2012 World Development Report, which is a really, really interesting document for any of you who are interested in these topics.

And Agnes Quisumbing is an economist and senior research fellow at the International Food Policy Research Institute.

And I just have to say, when people ask me, how did you get involved, how did I get involved in working on gender and development issues, and I sometimes tell the story that, well, it actually goes back to Les Gelb asking me to work on these issues and me saying, Les, I don't really know why I want to do that. (Chuckles.) I'm not sure I'm -- I don't know much about it. And he said, well, go read about it and figure it out.

And I started reading, and the work I started reading was -- that interested me was the work of economists who were looking at this. This is going to back a decade now. And it was Agnes' work on agriculture that really got me interested in the remarkable opportunity that was lying in front of us of investing in women as a way of increasing economic productivity around the world, because the gender gaps were so huge and the potential was so great, if you closed those gaps, to really increase economic productivity. So, Agnes, thank you.

AGNES QUISUMBING: Thank you, Isobel.

COLEMAN: And of course the World Bank reports, "Engendering Development," which your predecessors did at the bank.

So let's -- Markus, let's start with you, and actually, maybe you could just say a word about the 2012 World Development Report and explain a little bit about how -- why it was seminal, in a way, for the U.S. -- for the World Bank to look at gender in that report. This is a report that's been done many times.

MARKUS GOLDSTEIN: Right.

COLEMAN: And hone in on what it actually -- what the kind of key takeaway was.

GOLDSTEIN: So it's seminal in the sense, I think, as Mayra was saying, that, you know, we'd been doing it for 25-plus years and it's always on that topic; poverty comes up every 10 years or so, so on; and there had never been one on gender. So in that sense, for the bank, you know, the subtitle maybe should have been "It's about Time."

COLEMAN: Yeah.

GOLDSTEIN: And it's partially about time. So I think that's one of the messages as well.

With regards to economic empowerment, I think it's a half full, half empty kind of story. Half full: You've seen the huge surge in women's labor force participation. You know, 4 out of 10 workers now globally are women. Education -- we were talking about this at lunch -- education for women has gone up. The gap has closed and that sort of -- that explains, I think, part of this increase in women's labor force participation.

But women are still earning less than men. And I think here the report does a bit to -- relative to other big global, you know, sort of monitor stand kind of reports that you have, I think it a does a bit to sort of get down to the underlying causes.

So if you think about the fact that women earn less, you know, and it's -- and it's global, so in Germany women earn 60 cents on the dollar compared to men. In Malawi, farmer -- women farmers earn 75 cents on the dollar compared to men. So they're making less, but they're not worse workers, they're not worse entrepreneurs, they're not worse farmers. So why?

Two sort of proximate causes. One is the sector that you find them in. So in terms of the firms and the labor force, women are concentrated in the service industries and the agriculture. You don't see them concentrated in manufacturing. So the sector, I think, does a lot to explain it.

And then the second big sort of proximate cause is this lack of access to productive inputs -- technology, seeds, land, capital.

But that's sort of -- those are symptoms, and the causes behind that we sort of trace out to four things. And I'll be brief here (before ?) -- it's a very long report. (Laughter.)

I think there's issues in markets, issues of information, networks, access to markets. There's issues around norms. That's a second factor.

Then there's issues in terms of institutions. You think about formal institutions, service delivery. Even when service delivery is, quote, "gender-blind," it can have -- it can be sort of discriminating against women.

And finally there's a -- there's an issue of time. Women all over the world, from Sweden to Bangladesh, do more work than men, and they work longer hours than men. And they do more at home. You know, I was looking at the statistics the other day, and women in Pakistan -- men in Pakistan do more in terms of hours of housework per week than men in Sweden. So this is a -- it's a global thing. And that time means that they have less time to engage in the economy.

So I think those four things, taken together, are what drive this pattern. So we'll see.

COLEMAN: And the focus on entrepreneurship is really important in the work that both of you are involved in with the U.N. Foundation. And the focus on entrepreneurship makes sense, too, because so many women are self-employed and so they are entrepreneurs.

And they face particular barriers. What do we know about those barriers?

GOLDSTEIN: I mean, I think we know -- I'll go back to what Mayra was saying. We know a fair bit about those barriers, although there are some that aren't immediately obvious. But we know less about -- a lot less about how to effectively overcome them.

So let me give you an example from -- I'll go agriculture. In Ethiopia, we are -- we're working with a group of women and agricultural extension. And so this is the service provision story. The Ethiopian extension courses are held eight hours a day, one day a week, at the farmer training center, which they have to travel to.

Men were dropping out of these courses at the rate of 50 percent, women were dropping out at the rate of 90 percent. So, why? Well, mobility was an issue and time was a big issue, right? So they couldn't get -- they couldn't get eight hours in a row, they couldn't get eight hours. Those were both factors.

But they also said that they were being trained in cash crops which, given the government strategy, was what the extension was focusing on, and they weren't growing cash crops. So you -- then, you have a question, do I go for the transformative policy solution, which is let's take women and train them in cash crops so that they can break into the sector where they're vastly underrepresented? Or do I provide training to them in the crops that --

COLEMAN: In what they already do.

GOLDSTEIN: -- in what they're doing, so that the service actually is useful to them. And then if I decide to go the transformative route as the policymaker, who do I target in the community? Do I go after the wives of the most influential women (sic) -- so I'm not going after the poorest women -- and hope that it'll trickle down or change the norm such that this will be OK? So -- and that's an open question.

But -- so I think -- when we think about overcoming these barriers, we have a need for evidence and we have a need for experimentation and we have a need to sort of get those lessons out there and share them. And, you know, I think Mayra's project is interesting because we're starting to get those lessons and putting them out there.

But, you know, another example, on time constraints, when we were writing the World Development Report, there's very little evidence that -- in particularly in developing countries, that daycare is the answer, right? Because we're not going to get men to do more work. (Laughter.) We keep trying, all right? This is an issue consistently at my house -- (laughter) -- so I know we keep trying, but -- so there, it's about the norm and about circumventing it. So we think about daycare.

There's some evidence in Latin America, maybe we get peri-urban, that that really increases women's labor force participation. One or two studies. What about on the farm? What about within the plantation? What about in the marketplace? These are things we don't know and very few people are trying them. Some people are -- you know, Seiwa (ph) in India is trying -- is trying it for their members. So we need to document those things and sort of figure out which ones are working and which ones aren't.

COLEMAN: Agnes, in terms of entrepreneurship, I mean, many female entrepreneurs are in agriculture.

QUISUMBING: Mmm hmm.

COLEMAN: And tell us some of the -- from the work that you're doing now -- some of the insights you're gleaning on what works and what we don't still know.

QUISUMBING: OK, so I deal mostly with women, self-employed people in agriculture, and I would say that they're almost entrepreneurs -- almost entrepreneurs in the sense that they're on the brink of producing enough that they can actually market it, but they're not quite there yet. And I think some of the barriers have to do with the barriers of poverty to begin with, having much lower access to inputs, land, labor that they can mobilize, fertilize and create other inputs.

But even within poor households, women tend to have much less access to those inputs than men do. So there have been a number of attempts to try to overcome these barriers, and I think that the ones that have worked have worked because they have made specific adjustments to their social and -- social and cultural context.

So the area of the world that I know probably best is Asia because I grew up in Asia. I do most of my fieldwork in Bangladesh and the Philippines. And in Bangladesh, one of the issues was how to deliver resources to women who are -- traditionally have been secluded because of the female seclusion, who don't have equal inheritance rights, et cetera.

So one of the -- one of the things that has to work in Bangladesh has been using group-based approaches. And in some studies where we have looked at the same technology which is being disseminated through group-based approaches, versus an approach which is household-based but targets -- ends up reaching the male of the household, the group- based approaches have been better at building women's assets and reducing the asset gap between husband and wife.

In other places, the -- or in other socio-cultural contexts, the issue might not be social exclusion. It might be as simple as access to information or access to technology or even the issue of learning about more productive techniques so that they actually enter the value chain. So we have been -- the project which we are reporting on for Mayra's research effort actually looks at a number of different value chains. One is in dairy, and the other one is in horticulture in different countries. So we have two value chain projects in Africa and one in Bangladesh and one -- and two horticultural projects in Africa.

And I think -- as we talk about the horticultural projects, I think here it's where I think the socio-cultural adaptation is quite interesting. In one of them, which is the orange-fleshed sweet potato project in Uganda, sweet potatoes were -- sweet potatoes there are not are nice, yummy, orange-looking sweet potatoes you see here during Thanksgiving. They're white. They're -- or light yellow. And so HarvestPlus bred these to have orange that -- to be more orange so they had more beta carotene and can better address the vitamin A deficiencies.

But what they used in their extension messages was they targeted them to farmers' groups and then really emphasized the nutrition message in their messages to women. And they found that that was very effective, actually, in getting people to adopt OFSP, and particularly also because one of the most important ways of disseminating this information is through social networks. So the women's social networks are very important in disseminating information about OFSP.

The other --

COLEMAN: Orange-fleshed sweet potato.

QUISUMBING: Orange-flesh sweet potato -- (laughter) -- your regular sweet potato here, not your regular sweet potato there.

The other vegetable project was in Burkina Faso, where -- this is being implemented by HKI, which has been doing homestead food production projects all over the world. I mean, their program in Bangladesh is now nationwide and scaled up. What they did was -- I found this very interesting in Burkina Faso -- is they tapped older women leaders, their OWLs. They tapped OWLs to serve as people who could disseminate this message.

So FP (ph) is now working with HKI to randomly select communities which have OWLs and communities which have the traditional ways of disseminating this information. We're still in the middle of the evaluation, so we don't know what the result is. But I think it -- this illustrates that you do have to experiment and innovate with respect to what adaptations you need to make an intervention successful in its context and then evaluate it properly.

COLEMAN: I know we're talking a lot about entrepreneurship, but what about wage employment? Because not everybody's cut out to be an entrepreneur, and some people are seeking wage employment. How do these -- what are you finding in the -- in the research about wage employment?

QUISUMBING: I think the primary driver of wage employment is education and of course opportunities, but I mean, if you look at Bangladesh, for example, you're having a massive outflux of girls from rural areas moving to the garment factories. And part of this has been sustained because of the very heavy investment in female education both at the primary and the secondary levels. I mean, I think underlying all entrepreneurship programs should be really a very -- a very committed government or a committed policy to increase female schooling and to close gender gaps in schooling.

COLEMAN: There's been a lot of money invested in job training. What do we know about job training in these developing countries -- (inaudible)?

GOLDSTEIN: OK. So I think there we have more -- I'm going to -- I would say there we have much better evidence about what works for women than we do for business training. Business training, I think that one's still pretty mixed.

COLEMAN: Business training meaning --

GOLDSTEIN: For entrepreneurs.

COLEMAN: For entrepreneurs.

GOLDSTEIN: Yeah. So -- I mean, I think -- but still, you know, in regions like Africa, South Asia, we have less. A lot of this comes out of Latin America and the developed countries.

And so a couple things tend to work better. For the programs to benefit women, an extra subsidy seems to be associated with more success, extra subsidy for transport but also for child care. Those are things that -- or providing child care in some way. Those are things that seem to work.

And then in terms of the job success and retention, there's a -- the evidence isn't clear yet, but what seems to matter is either on- the-job training a lot more than classroom training and soft skills training, so not, you know, how to be a better accountant, but how to negotiate a workplace. And those --

COLEMAN: Show up on work on time, how to --

GOLDSTEIN: Exactly.

COLEMAN: -- work with other people and --

GOLDSTEIN: Exactly. So the stuff that's coming out now is -- I think it's still fairly suggestive, but it suggests those things matter a lot. You tend not to see programs succeed which don't do that. Now, they're bundled often with other things, so we haven't been able to pull apart the bundle. Some people are trying to do that.

COLEMAN: But in terms of business training for entrepreneurs, are we -- you said we don't know a lot about what works. Do we know anything that works in business training for entrepreneurs, because again, there's so many programs that are out trying to target --

GOLDSTEIN: (Inaudible.)

COLEMAN: -- entrepreneurship and creating entrepreneurial capacity in countries, and it's a little alarming that we have -- (chuckles) -- all these programs and we don't really know what works. So --

GOLDSTEIN: Well, and I -- my take on the literature is that we know a -- we have a couple of programs where we've seen them work for women and men together. We're pretty sure they work for men. We're not sure about women.

In terms of programs that -- so I think when we're looking for a program that benefits women, maybe one or two. They're small, because a lot of these studies aren't set up to look at the gender dimensions. So they'll tell you if they worked or not, but they won't tell you anything about the gender differentials.

Some of those that are suggest that the programs benefit only men, so there's some work in Sri Lanka, for example, that was depressing. And the root cause there -- part of the story was that the women were afraid the husband was going to -- they gave them cash, right, so it was training plus a cash grant for the business. And the women were afraid of the expropriation from their spouse, so they were investing in things that couldn't be expropriated but weren't necessarily the best things for the business. So there's a distortionary effect from the -- from the husbands, husbands as distortions. (Laughter.)

COLEMAN: Now, will the -- will the -- will the spread of mobile technology, for example, begin to quickly change some of these things where women can now get that cash in the form of electronic money on their cellphone, which is becoming more and more ubiquitous, and then not have that expropriation worry and distortion effect?

GOLDSTEIN: I mean, in the Sri Lanka case, I don't think a -- the cellphones would have solved it, because it was an investment in a physical asset for the business, right? So he's still going to come and take it if it's something he wants. So she was trying to pick things, as I understand it --

COLEMAN: That he couldn't take.

GOLDSTEIN: -- that he couldn't take. That he wouldn't want to take. He probably could take them. So -- but, you know, in Ghana they did a very similar study, and they found very different impacts. No impacts for women who were at sort of the bottom -- this is Mayra's point about the poorer entrepreneurs -- but the ones on the upper part of the -- sort of the scale of business -- I mean, here we're still talking small businesses, but the more successful ones, there there was an impact, right? So maybe the Ghanaian women have more bargaining power than the -- than the Sri Lankan group of women.

So mobile phones, I think, have a lot of potential. Again, we have yet to see how it plays out in practice. I'll give you an example. We're working with a bunch of traders on the border between DRC, the Democratic Republic of Congo, and Rwanda. And we were -- we're collecting data and providing them information through phones. Six months later, a significant fraction of the phones are now in the hands of the husbands, right? So we gave them the phones, but the -- you know, some of them were sold, maybe to their husband, but some of them were -- seem to have gone over to the husband. So I think there's -- you know, there's even this issue of control of that basic -- now, I think as they flood, you know, when they're -- when they're $20 each or $10 each, then you start to -- then everyone has one, but --

COLEMAN: Last question and then we'll turn to your questions in the audience, but what all of this is saying is that there's a lot more we don't know than what we do know.

And are we collecting data in such a way now that will begin to help us hone in on the right answers? Are we doing -- many of you may have come to the session I did with Dean Karlan here about a year ago, who -- an economist from Yale who has been doing a lot of randomized control trials, looking at works and what doesn't in development. Are we doing that type of thing in a sufficient way across lots of different interventions to begin to understand what does work and what doesn't?

I think -- I think we know -- we look at the massive youth unemployment in sub-Saharan Africa, in the Middle East and South Asia, in this country, in places around the world, and we know that entrepreneurship has got to be part of the solution. And yet, you know, for all the talk about improving the environment for entrepreneurship, the fact is we don't really have a good understanding of what investments are going to change the -- move the needle. And -- but are we at least putting the right programs in place to figure it out? And what's the time frame for doing that?

Agnes?

QUISUMBING: So I think your question, Isobel, has -- I think it has a two-pronged answer. One is, what kind of data are we gathering, and the other one is what is the design of the research that we do to evaluate these programs, and are -- will this type of data collection be embedded in that?

So I'm going to answer the data question first. I think there have been major strides made in terms of collecting data at the intrahousehold level, where we don't only look at assets of the household but also the assets controlled -- owned and controlled by men, women and other individuals in the household. People who gather data always say that, oh, it's too difficult, it's too much work to ask this. But actually it's not. People like talking about what they own, and they can usually identify, oh, that belongs to him, that belongs to her, or we own that jointly. I mean, even something which is as simple as that can go a long way in terms of identifying what are the resources that men and women each bring to a household enterprise.

Now, in terms of methods of -- to understand what works, I definitely support the experimental approach, but I don't think it should be the only approach to program evaluation. I think that it could be nice to have experimental approaches co-existing or even working together with some qualitative methods where you explore the perceptions of people who are involved in the program to see why they think something worked or not. So it's like they would experiment, when they asked the woman, why didn't you get this -- into this, you know, more lucrative business, and it turns out that she doesn't want the husband to take it away from her.

So I think that experimental approaches will help you get a good measure of the estimated impact. You may not necessarily understand very well why those impacts happened, and if you're trying to replicate or transfer the model to another area, you have to question some of the external validity and look at -- you have to be conscious of the adaptations which you have to make to scale it up or to transport it to a different setting, and do the necessary groundwork in -- wherever you're taking that intervention.

COLEMAN: Markus?

GOLDSTEIN: Yeah -- no, I -- this is something I think started when Mayra was at the World Bank; I -- we really, at least from the World Bank point of view. You have this big push on experimental methods and development economics, and you have all these gender programs that have been run on, sort of, well-informed hypothesis -- like we know this is problem, we think this can fix it, but we don't -- we still don't, at that point, don't know if it works, you know, in a very cost-effective or even in a positive return sense.

So I think, now, we -- it's growing, right? You're starting to see a lot more -- and hopefully more focus on these underlying causes: Is it time? Is it service delivery? Is it property rights? And going after these with experimental methods and, hopefully, with qualitative complements so that we can really understand what's going on.

So we've started -- you know, in the Africa region at the bank, we have this thing we call the Gender Innovation Lab, sort of following in the footsteps of the other labs and we're trying to experiment with these kind of things. And you know, we're doing work on property rights, we're doing work -- trying to get some work on time -- on day care, basically. So is it enough? No. I mean, there should be a lot more going on.

But I think the interesting that you're seeing from sort of the discussion that started with the gender action plan that Mayra was doing is more of these other economists' interest in gender issues, right? So there's this nice spillover and leveraging where you're getting -- when someone does a microfinance evaluation, they pay attention to the gender dimensions; when someone does a business training intervention, sometimes -- at least more than zero -- they pay attention to the gender dimension.

So I think there's a -- there's a kind of -- it's a good time in terms of expanding and looking at the really important issues. But we -- but it's going to take three, four years before we get --(inaudible).

COLEMAN: We do have microphones, so just raise your hand and we'll bring a microphone. We'll start with Betty right here.

QUESTIONER: Endeavor is probably one of the most successful entrepreneurship-building nonprofits, and their model is something that the country business leaders, when they go in, have to mentor the young entrepreneurs. And -- but the level there is much higher. You have to be able to write a business plan and have some organization.

But what I don't understand is, how are you all going to make criteria to be determined to be an entrepreneur? And how do you decide what programs, and how do you recruit the people you're trying to foster to become entrepreneurs?

GOLDSTEIN: So let me answer with an example. We're working with a program run by an NGO, BRAC -- Building Resources Across Communities. It used to be Bangladesh Rural (Access ?) -- but now they're global so they changed the name. And so we're working with them in Uganda, and this is a program for adolescent girls. What they do is they give them a space, sort of a club. It's a room -- nothing fancy -- in the village for them to hang out. Then they do some life skills training -- this is the soft skills I was talking about -- how to do a household budget, health, hygiene, everything. I mean, it's sort of a grab bag of important things that they should know.

QUESTIONER: (Off mic.)

GOLDSTEIN: Yeah.

QUESTIONER: (Off mic.)

GOLDSTEIN: I would not disagree. No, no, no, I wouldn't disagree. There is a BRAC USA office, but I don't know if they've started working. (Laughter.) And then they do training, and so here they let the girls --

QUESTIONER: Training for what?

GOLDSTEIN: No, no, I'm coming to that. That's -- that's the root of your question. They let the girls select into whether it's training for wage jobs or training for enterprises. Now, the training for enterprises that's offered is based -- they do an assessment of the local market, and so it's tailored to the surrounding area, right? And this seems to -- it seems to work.

QUESTIONER: And these are adolescent girls.

GOLDSTEIN: These are adolescent girls, 16 to 24. It's an older set of adolescent. And so after two years -- we set this up with them in a way that could be evaluated rigorously -- they did the program, we did the evaluation -- and after two years, these girls are more likely to be running their own business. Not all of them. Some of them have pulled out of wage work and are running their own businesses. They're making 30 percent more than they were before. They're 30 percent less likely to have a kid, and they're 75 percent less likely to have had sex against their will.

So putting all of these things together -- this safe space for girls, the soft skills training, this job -- you know, this enterprise training that's tailored to the local market -- on average is allowing these girls to do a lot better in a -- in a -- in a whole range of dimensions relative to girls who don't participate in this.

But your larger question about should -- I mean, I think the larger question you're asking is should everyone be an entrepreneur, and I think the answer's no. But we're still -- I think the tools for screening people, other than letting them decide on their own, which is what this program did -- the tools for programs to screen them are still -- they're getting better, but they're still not there, right -- in terms of figuring out who's going to have success in business. I mean, that's the magic bullet, right?

COLEMAN: Scott.

QUESTIONER: Thank you. Scott Carlson, Hutchins Family Foundation.

We've got over 20 years of experience and data from the microlending community. Why isn't that a good source of what works, what doesn't work data, et cetera?

GOLDSTEIN: It would be a good source. (Laughter.) It would. I mean it would. I think --

QUESTIONER: (Off mic.)

GOLDSTEIN: So -- I think there's a range of lessons that can be drawn. So I don't know your data, so I can't -- I can't pass any judgement on it, but I do want to draw a general distinction between two ways of -- or I think there's a continuum of the way we draw lessons.

And at one end, you have sort of randomized control trials. The value of these things is that people who participated in entrepreneurship programs, people who borrow money from microenterprises -- they're different than the average person, right? And they're different usually because they're more entrepreneurial.

So if we look at those people five years later and they're more successful having taken a loan from our microbank or having participated in our training, it's hard for me to tell whether it's because they're more entrepreneurial to begin with or because of the effect of the program. It's probably both. And I'd like to know how much was due to the program because that's the part that we control. We don't control people's abilities. So I think that's the value of that kind of learning.

But we also learn a lot from monitoring and evaluation of programs and what's going on in how they deliver things and who they're targeting and how they target people. So I think there's a whole range. So that's why I think you're data's useful. If I knew more about it, I could be more specific.

QUISUMBING: Can I add to that too?

COLEMAN: Yeah.

QUISUMBING:I think also that if you're talking about program data which was gathered by people who are actually implementing the program, they're useful in finding out what works within the context of a specific program. It may not necessarily tell you what the situation would have been in the absence of a program.

We're also finding that some of the longer-term data, however, have a value because you can take -- you can deal with long-term impacts. In some -- for programs which are trying to build assets, whether they're business assets or physical assets, the short-term effects may end up being very negligible, may actually end up capturing a lot more about -- you would find out a lot more about the long-term impact of these programs if you did have these 20 years of data on the same individuals. And so ideally, it would be nice if you -- you know, whatever data were collected would track the same individuals or households over time so you can actually see whether they have been able to make progress.

COLEMAN: Let's just -- I'm just going to put in a little plug for Peter Tough's (sp) new book, "Why Children Succeed." Everybody in this room should read it. But it's really interesting on this longitudinal data, where there were all these early childhood intervention programs that they had a big impact for a year, but then by fourth or fifth grade, they were negligible. There was -- there was no discernible impact. But 40 years out, those kids are now, at much higher rates, healthier, married, holding down a job, in remarkable ways that you wouldn't have known at fourth grade. And you know, it comes back to they -- these two control groups.

So I think this point on longitudinal data is really important. And what they thought they were doing with this early intervention was helping kids read at a better level. And by fourth grade, those reading improvements were negligible. But something else happened with this early childhood intervention that led these kids to have much more successful lives by, you know, any number of measures 40 years later. And I think one of the controversies around microfinance is, you know, you're trying to measure effects a few years out, and they're not always there. And a lot of the microfinance institutions are saying, well, you know, you're not looking at it over the longer term, or the impact on the next generation. And you know, I'm not the economist here, but it's complicated, and I think you do need those longer-term studies to be able to really draw these things out.

Sorry, Craig (sp).

QUESTIONER: You talked a bit about training. I was wondering whether there's been any research -- and mentoring as well -- any research on coaching and its impact on women. And by coaching, I mean something different from mentoring, but rather -- you know, it's a strategy which develops skills, which is fairly easily scaled -- (inaudible) -- a lot of success with businesses in this country and in some developing countries I'm aware of.

GOLDSTEIN: I haven't seen anything.

COLEMAN: Haven't seen anything.

GOLDSTEIN: I mean, I think it would be very interesting to look at. Yeah, I'm not -- that's not -- (inaudible) --

COLEMAN: Are you -- are you aware of any coaching programs in developing countries -- (inaudible) --

QUISUMBING:I am aware -- I am aware of a program which is meant to improve the career advancement of African women in science and technology. And this is (award ?). So it's set up like a mentoring program, but it's not an entrepreneurship program as far -- I mean, it's not.

QUESTIONER: (Off mic.)

GOLDSTEIN: Yeah. No, I think that'd be really interesting to look at, particularly -- sorry -- getting at this issue of helping women break into male-dominated sectors. I think that would be a fascinating thing to try because I think support -- from the qualitative work we're doing, support seems to be part of the story.

COLEMAN: Yeah. Over here.

QUESTIONER: Hi -- (name off mic) -- with JPMorgan. One of the heartening and really inspiring aspects of development is how grounds-up it often is in terms of the work.

One of the very challenging things about development, when you think of the scale of the challenge, is how grounds-up the work needs to be. Can -- since we're focusing today on the positive of what's working, can you talk a little bit about where scale has been most effective?

COLEMAN: Agnes?

QUISUMBING: So I think one of the -- one of the -- so first let me start with a not-so-promising idea, and then I'll move into the more promising idea. (Laughter.) So some of the not-so-promising ideas is that you have a lot of civil society organizations which are implementing their own little programs on the ground, and when it comes to scaling up they flounder because not much effort has been put into mechanism and design and all that stuff.

But the partner that Markus was just talking about, BRAC, is really -- has really been excellent in terms of scaling up. I think one of its first successes was in the -- oh, gosh. I wasn't even working in development, so I -- it's been a -- it's a long time ago that when they pioneered the use of oral rehydration salts and taught basically all their community health workers how to mix a proper -- (inaudible) -- solution and distribute it to -- you know, to families and sick children among that. BRAC has been a model in terms of scaling up and delivering interventions really at scale.

So I think what we might want to do is to look for those organizations which have been successful in delivering things at scale. These are the organizations which are not afraid to experiment. I mean, it's been very interesting to work with BRAC, because I work with them mostly in the Bangladesh context, whereas I think Mark was working with them in Africa. They're very -- they are not averse to tweaking their programs. They have their -- they have their own research and evaluation department, which does studies of its programs, how they work. If the program needs adaptation, they change it.

And so we would -- I think they could still be a little bit more systematic in evaluating some of these aspects which are -- (inaudible) -- but I think that idea that you're an implementing organization, you have an idea of experimenting, of documenting and evaluating your own experimentation and looking at (mechanism ?) design I think is what you're looking for eventually.

QUESTIONER: Thank you. First of all, thank you, Isobel, for putting this together. I'm Carole Brookins. I was U.S. executive director at the World Bank for four years a few years ago. And I wanted to refer to a report, "Scaling Up Poverty Reduction," which was a quite good report that was done. I was involved in it when I was at the Bank because it really did take examples. It wasn't gender-specific, but it took examples of things that had been done that had been scaled up like the KDP program in Indonesia.

I wanted to ask both of our wonderful speakers here, all three of them or both of them and -- who are up here, Markus and Agnes, about some of the geospatial issues that are there, the rural versus the urban. Where there's more population density, it's easier to reach more people. It's not the same kind of cultural issue, and it's not the same infrastructure issues.

And I -- every country I visited, whether it was Chad during the building of a pipeline or whether it was Bangladesh or Uganda or Indonesia or -- before and after I was at the Bank, so many women that I spoke with in villages who were in local areas, so much of their challenge was on infrastructure, where you could build a feeder road, or where you had actually an all-weather road, or you had household water -- that's where the KDP program was quite striking to me where I visited in parts of Java -- that women felt so much more empowered. They weren't -- they weren't dray animals anymore. They were actually having opportunities, or they could go to school, or they could do things more simply. And so I wanted to know how much of this issue of infrastructure it's -- is billed in, or is this kind of siloed out in your considerations?

COLEMAN: Markus? Markus?

GOLDSTEIN: I think infrastructure matters. I still think we're teasing out how it matters, because, you know -- I think when you're with -- so there's two studies of water, and they're not -- I wouldn't list water -- providing pumps as my number one intervention for gender equality. I still think it matters. I mean, both of these studies -- this is what -- Agnes' point about context. So one study is in Morocco, where there was a standpipe outside the house, so they brought the water inside the house. Big deal. I mean, it didn't lead to a lot of time savings. It led to some, but not a lot of time --

QUESTIONER: But what would they do before they had the standpipe? (Off mic.)

GOLDSTEIN: No, no. It came inside the house, so they were just going outside to get the water.

QUESTIONER: No, I'm speaking of the difference in -- (off mic) --

GOLDSTEIN: Yeah, exactly, exactly, and their -- and so that's my point, is we don't -- we don't have really good evidence on how that changes things and how much of it is for girls versus women, right, because with water, that's a big issue.

Roads and access to markets.

That's something, again, we think matters a lot, but I haven't seen a lot of concrete evidence that sort of shows this is what's happening. We're looking at two or three roads projects to try and get a -- at this and how much it matters.

But I think going back to Isobel's early point on cellphones, you know, the interesting thing there is market information -- there's a lot of interesting interventions around market information as well as transfer of money.

So some of -- some of these things will be -- you know, leapfrog the technology, and that'll help close -- and there's an interesting literacy -- female literacy programs in Niger using cellphones. So I think there is some things that the technology also help with, which seems to move much faster than building roads.

QUISUMBING: But I think following onto that, I think the geospatial issues and the cultural issues sort of come into play very interestingly where, in areas where there are traditions of female seclusion, so giving a woman a cellphone, assuming that she can keep it, can be actually quite liberating in the sense that if she's a farmer, she can get access to agriculture information; she can maintain contacts with her relatives, which is a very important source of support. And so I think particularly in areas where you are geographically isolated, having those forms -- having other forms of technology can be quite important.

QUESTIONER: Good afternoon. Binta Brown, Kirkland & Ellis. So two quick questions.

The first is, understanding that we've been talking about microfinance and smaller women entrepreneurs, I'm wondering how the data for success changes once you start talking about women founders of prospectively scalable businesses. That's the first thing.

And then the second thing is, we've been talking a lot about women it seems on the assumption that all women have children and are married. And so my expectation is that the barriers that exist for women entrepreneurs, women wage-earners, are extant no matter what their status is. So I'm just wondering what research has been done in that regard.

The comment before about interhousehold conflict, making it -- lessening and making it maybe possible for women to be a little bit more productive, I'm curious about that because it seems to me that maybe one of the sources is having greater support as opposed to a single woman who might not have any support, which would be an impediment to her productivity as well.

QUISUMBING: I'll tackle the second question because I don't really know much about women founders of prospectively scalable businesses.

So I think there is -- there is enough literature in terms of even ranging from areas of, say, labor force participation to -- ability to participate in all sorts of development programs that looks at the way family structure and demographics affect a woman's ability to participate and benefit from these programs. And while it is the case that -- I mean, sorry to say that in some cases, the husband may be a nuisance, and maybe she's better off without him around -- familial support is also important.

And so we are seeing in some -- so in this -- in this multicountry project that I'm doing now where we're actually working with implementing organizations and seeing, you know, what adaptations they have made to make their projects work, we -- a lot of women, they are saying that it's very important for me to bring my husband on board. So a lot of them are married. They have husbands. And they aren't trying to look for ways that they can bring their husband on board because the demands, the time demands of these new enterprises are quite heavy.

I think that a lot of development interventions may implicitly assume that women's time is -- costs less. Of course, everybody knows it is not. I mean, there are things to do. There are kids to -- there are kids to raise, you know, houses to keep clean, meals to cook, whatever. And then if you -- and if you now bring a new productive enterprise into the -- into the picture, there are time costs. And so the ones -- the -- in our -- among our partners, the ones who are doing better are the ones who are able to mobilize familial support.

And I think that -- and actually, the interesting thing is, as a result of -- partly because of working with us, the organization we're working with is now implementing -- or is -- actually, they're experimenting with a small evaluation, (which is ?) a little test whether communities where they have implemented programs to sensitize husbands to the importance of women's owning assets, succeeding in business, are better off -- I mean, whether the programs end up being more effective when you actually the husbands on board.

COLEMAN: It's just -- maybe Markus, you want to comment on this point too, that a lot of the work you do -- we were talking earlier about, you know, Markus' -- your title is the gender expert, not the women's expert, the gender expert.

And the work that you do -- you know, as you said earlier, you cannot be successful in doing that work without really including men in that. And you find that, I think, across all of the work that you're doing. Whether it's this -- you know, it's the husband sensitivity training you're talking about or something else, that at the end of the day it has to really -- men have to see the benefit, right?

GOLDSTEIN: Yeah. Well, I mean you have to -- yes, you have to deal with that in your programs.

We did some work on Rwanda, on land titling, land title registration, and so Rwanda passed a whole battery of laws which gave women better property rights. So I wouldn't expect a program which writes down these rights, which already exist in law, to have any impact, right? Because people know the law, and the husbands should be listening. But yes, it did have an impact to write down those rights and put them in place.

And so, you know, so that's one mechanism which is going in forcing the household to engage with this. And, you know, we saw greater investment response in terms of land fertility on the side of women, which meant that they were more insecure to begin with, despite the law having been there for a number of years, not a huge number, but enough.

COLEMAN: Do you want to comment on Binta's first --

GOLDSTEIN: On the scalability question, I think, you know, this comes back to this question of who's going to be a good entrepreneur or not. And I think we're still -- I think we've made some progress in the last couple of years, very recently, on better measurement, on who will be -- who's likely to be a good entrepreneur or not, particularly in these low-income settings.

And so some of the -- one or two of the recent studies have looked at the impact of programs on people who you would predict, based on psychological -- you know, this comes from the business psychology literature, comes -- and those programs tend to work better, not surprisingly, for people who are more likely to be an entrepreneur.

And in fact, sometimes you see -- well, you see two things. There's one study which shows that actually some of these business training programs cause people who are less -- who score less well on these entrepreneur measures to stay as entrepreneurs. So in the control group, those people stop being entrepreneurs. So maybe that's not such a good thing; I don't know. It depends on the effects of the training on those people. And some of them shows that those other folks exit, right? So I think it depends on the nature of the intervention. Some of those interventions are enough to cause people who realize that maybe they shouldn't be an entrepreneur to leave and other people to stay; you know, the good ones to stay.

So I think we need to do more work in that regard and we need to get those tools better. I mean, obviously, if those tools were great, I wouldn't be sitting here; I would be running an investment fund in a developing country. (Laughter.) And you all could sign up.

COLEMAN: Ellen and -- (inaudible).

QUESTIONER: Just very quickly. I'm Ellen Chesler from the Roosevelt Institute. This has been such an interesting conversation. But representing the Roosevelt tradition, I want to ask about law and policy. You briefly mentioned the issue of law and policy, particularly human rights law, by mentioning Rwanda and the success with land reform. But there are so many other examples of places where, even in countries without strong governments or strong rule of law, policies that donor countries can impose are making a difference. For example, with respect to entrepreneurship, we know from our own history in the developed world that affirmative action in contracting by the government with women-owned businesses was probably the most successful strategy at scaling up women's opportunities in the 1970s under a Republican government; you know, under Richard Nixon.

So I know that the State Department and others are writing contracts -- some of the new development approaches involve contracts demanding, you know, that governments essentially create affirmative action programs. What do yo think of that? Is that a mechanism for scalability? Does it work? Does it cause other problems in terms of bad reactions from those left out? What can you say about law and policy in this area?

GOLDSTEIN: Well, there's a whole -- there's a large section of the World Development Report devoted to this, and I'm not going to try and summarize it, because it would be a disaster in, like, five minutes. But just a couple of points.

One is, as somebody who deals with gender in Africa, one thing that I've heard -- and I can't attribute causality, but what I've heard is, because the Millennium Challenge Corporation has now made gender equality one of their criteria for getting a compact, a lot more governments are interested in reform.

Now I don't know if that is a directly causal relation. We are seeing -- across the continent you are seeing changes.

Two -- I mean, you have had for -- one thing that I've been following is gender quotas in parliament, and you have had it in a number of countries in Africa. Interesting fact: The highest fraction of female parliamentarians is a country in Africa.

But now you're starting to see it in the -- in the older democracies. So the -- it's a subject of discussion when we were in Ghana. It's a subject of discussion -- well, I think it's -- it's coming in Kenya.

So in both of these countries, they've taken affirmative action to the parliament, and I think that's encouraging. I -- that I wouldn't directly attribute to anything from the MCC. I think it's part of the larger dialogue, but it may have helped. You can't rule that out.

COLEMAN: And we -- we're going in -- the meeting in December is going to focus specifically on women-owned businesses and global supply chains and this whole of idea of contracting.

GOLDSTEIN: Yeah.

COLEMAN: So we'll revisit it then.

GOLDSTEIN: Which is -- that's another fascinating -- (inaudible).

COLEMAN: Do you want to --

QUISUMBING:I want to say something also about policy in the sense that we often underestimate the impact of policies which may not be related at all to entrepreneurship. So I'm thinking about policies which are related to property rights -- for example, allowing men and women to have equal rights to own assets; policies with respect to marriage law, which protects assets upon marital dissolution; inheritance policies. I think these lay the foundation for women to own assets and to own them equally as men do. And it -- we don't tend to think of this as interventions which can help entrepreneurship, but they do provide the foundation for it.

GOLDSTEIN: Absolutely. I mean, yeah, that -- (inaudible).

QUESTIONER: Thank you. I'm very grateful to have a chance to listen to this presentation today. My name is James Tunkey and I'm an entrepreneur, and I'm the son of an entrepreneur who's a displaced homemaker. And I think about what our entrepreneurship -- I think about risk taking, and I know that my mother's choices that -- and what influenced her decisions to be an entrepreneur had different sets of decisions around risk than my decisions as a -- to choose to be an entrepreneur.

And I haven't heard a lot of talk today about risk taking. I think it's something that comes across all the different factors that you've mentioned. And I wonder if you could speak a little bit about what the data is saying about what influences risk taking, how risk taking by women you see as being different than risk taking by men, and how do you see that influencing policy? Thank you.

GOLDSTEIN: Wow. Do you want to --

COLEMAN: Agnes?

GOLDSTEIN: That's a big question. (Chuckles.)

QUISUMBING: That's a very big question.

So the studies -- the -- so the area of the world I work in are poor countries. So it may be quite different from the United States.

But I think that many of the studies that have been done on men's and women's risk preferences suggest that women are more risk-averse than men. I don't know whether this is still universally true. I think there are still more studies which are going out to replicate this.

But what I do know -- and some of this is coming out from some of my own studies -- is that men and women are actually subject to very different kinds of risk. They may be vulnerable to different kinds of risk. And in the area of the world where I work most, which is South Asia, I've found for example that some of the risks that can push a household generally into poverty would be, you know, the typical -- death of the income earner, a drought or a weather risk, illness.

But what's quite interesting is the way that these risks are borne differentially by people in the household. So what we found, for example, in Bangladesh is that because their emergency assistance system in the case of floods works well, weather risks actually don't have a great impact on asset holdings, quite surprisingly. But illness risk, illness shocks tend to affect women more, and that's probably because women's assets can easily be disposed -- they're smaller assets -- they tend to be disposed of when somebody gets ill. So it shows you a very different risk profile for men and for women and probably different sets of instruments that can be used to cope with that risk.

So how would that affect risk-taking behavior? I mean, if you're in a country where there is an emergency assistance system that works but there's no health insurance, I could see a woman's, you know, willingness to take on risk be much less if she sees that a healthy -- this -- the health conditions in her community aren't good, for example.

COLEMAN: We've got a number of questions. I see three. So why don't we take these three together, if that's OK, and then answer them together, because we're going to run out of time.

QUESTIONER: I'm Bill Abrams with Trickle Up, and we're in the business of helping women start businesses. And by the way, we include the husband in that process and there's definitely a positive effect.

A quick comment, quick question.

I was really glad to hear discussion of the ultrapoor and the distinction between women at the very, very lowest levels of poverty, living in 50 (cents) or 75 cents a day, (down to ?) a quarter, which is often overlooked. And I guess my question is if you could define entrepreneur, because I've heard a lot of implied definitions here. Or more specifically, is that the right word? Does that create expectations and connotations, especially among Western donors, in things that actually isn't accurate or productive?

COLEMAN: OK.

And here.

QUESTIONER: Thank you. Thank you so much for the conversation. I really appreciate it. We started off the conversation talking about research and the importance of knowing what works and what doesn't and sort of where to direct our energies and our investments. And Mark (sic), you said something that just keeps resonating in my head. Think about a wage employment training where you have men and women. And I think I heard you say that for men, we know it works, and for women, we're not sure yet. And I'm just curious as to why we know what works for men, but for women, there's still this question. Is it in the assessment? Is there some external factor that we may not have captured yet that will help us better direct our energies?

COLEMAN: Galen (sp).

QUESTIONER: Galen Gingrich (sp). It's interesting, though not surprising, that in a program on women entrepreneurs in developing economies, we end up talking about husbands. We probably should mention that other major driver of gender inequality, which is religion. Any idea of what works in getting religion out of the way of women entrepreneurs, both in developing and developed economies, I might add? (Laughter.)

COLEMAN: OK. Three questions: definition on entrepreneurship, why do we have greater certainty on what works for men and not for women, and religion. And you can -- any, all.

GOLDSTEIN: I'll take the first two. (Laughter.) No, it's up to you.

COLEMAN: Go ahead.

GOLDSTEIN: All right, let me start with the first one.

QUISUMBING: I want to start praying. (Laughter.)

GOLDSTEIN: Yeah, exactly. I think we've been talking a lot today about self-employed, right? I think that -- you know, when we talk about women in agriculture -- and I -- you know, Agnes made that interesting distinction and good distinction at the beginning. When they start marketing -- and so they're self-employed, right, because they're farmers. And then when they start marketing it, then they're starting to engage with the market and so on. So it's -- it is -- it's a continuum. And I think you're right to point out it comes with assumptions, and that -- and those assumptions color how people treat those kind of programs.

OK, so my point -- I think on job training, we have some decent evidence. On business training, so training for small businesses, there that's where we see some of these things showing impacts for men but not -- but not for women, and some -- and this is a small number, like less than one hand -- and a lot of them that are evaluating programs that show they work for their target population, which is predominantly men, but don't have the statistical power to tell the difference between men and women.

So to do a proper study, to tell the difference between men and women, as a rule of thumb, you need at least four times the sample size. So that's expensive, and you've got to find a lot of entrepreneurs to do that. So I think we see a lot less of those studies, period. And that's something we're trying to fix, but it's expensive.

So -- but the two that -- or one that I was thinking about in particular is a study in Sri Lanka, which, you know, worked for men and not for women, full stop. They had the statistical power for it. And so I do think we have more of a knowledge gap.

In Sri Lanka, that case was about husbands' expropriation. Similar program in Ghana -- expropriation wasn't an issue, which surprised me, because I worked on intrahousehold allocation in Ghana, and -- but they have very separate economic class, husbands and wives in Ghana. So maybe it's pretty clear whose is whats.

You know, with -- so I think we definitely need more work there.

COLEMAN: Agnes?

GOLDSTEIN: The -- sorry, you can --

QUISUMBING: I'll take the first question and the third question. So I just wanted to say a few things about the first question. I think -- my definition of an entrepreneur is somebody who takes production decisions, takes decisions about sale and, most importantly, takes risks. And so in developing countries, this could apply to a lot of people, you know, your poor subsistence producer, the person which is about to start selling in the market, somebody who's selling a lot of stuff and making money. So it might be quite different from the Western idea of the entrepreneur.

I think that if you -- if you look at those who are self- employed, by this definition of taking decision of production, taking risks and taking sales decisions, that would -- how -- and taking them for yourself or your family, that would be my definition there.

Now, how to get religion out of the way? That's really a tough one. What do -- what if we change the word religion to culture? Because I actually think that even the same religion in a very different sociocultural context can manifest itself is very different ways. I mean, I'm Filipino. I was -- and I'm Catholic. And I can tell you Catholicism in the Philippines is very different from Catholicism in the United States. (Muted laughter.) We party all the time. (Laughter.) (Don't tell that ?) to the Pope. But anyway, so let's take -- let's take a look at culture and how norms are expressed and how there may be barriers to, say, women's empowerment.

I think that people tend to think that culture doesn't change. But culture changes. It may change slowly; it may change more quickly when you introduce new technologies like cellphones. But I think that people who resist change -- because it's always been this way, (because ?) it's culture -- and are in the development world are probably fooling themselves because you don't engage in development work without changing culture, whether you realize it or not.

And so there is attention in the gender literature between addressing women's practical gender needs -- you know, what she needs to survive -- and her strategic gender needs, what she needs to -- what you need to -- you can change power structures. And so some -- and sometimes interventions may have to take a very delicate balance because we have seen cases of interventions which succeeded in the short term because they met her practical needs, they respected social norms -- so for example, a project that we've studied in Bangladesh, which is a vegetable production project, where the NGO came to the household and bought off the market -- bought off the goods and so -- in the market so that she wouldn't have violate social norms. But that ended up not changing norms for her and possibly may not have been as transformative as one where they actually required women to go to market, probably doing in other ways such as having group trainings in out of town, or -- I mean, there are ways to attack barriers sort of gently or sort of stay away or not -- or not attack them at all.

And I think what we're seeing is that through some progressive and maybe, little by little, some incremental change, culture does change. So I'm not -- I'm not going to give up. I'm sort of -- I belong to the water-wears-away-stone philosophy of getting religion and the culture out of the way. Just keep at it, and it'll change, we hope.

COLEMAN: Well, I want to congratulate you all because you've sat through a very wonky discussion here at CFR. But I really think it was an important discussion because, you know, we've been talking about -- we talk about these big topics: entrepreneurship, development, women in development. And it's really important to be clear and have a clear understanding of what we know and what we don't know because otherwise, the tendency is to talk in abstractions.

And I think the work that's going on with this project is really interesting because as you can see, there is a lot we don't know. There is a lot we don't know. And understanding what you don't know is just as important as figuring out what you do know and being able to hone on what's going to work. And I think we're going to hear a lot more interesting things coming out of this.

And so I really applaud ExxonMobil, the U.N. Foundation, the economists themselves who are doing this work and helping shed some light and understanding on what we all know to be really important issues, which is how to improve entrepreneurship, how to promote women's empowerment at the same time and overall lead to more productive economies. I think that's what we're all after here. So let's thank our speakers today. (Applause.)

MR. : Thank you.

MS. : Thank you, Isobel.

Copyright © 2012 by Federal News Service, LLC, 1120 G Street NW, Suite 990, Washington, DC 20005-3801 USA. Federal News Service is a private firm not affiliated with the federal government. No portion of this transcript may be copied, sold or retransmitted without the written authority of Federal News Service, LLC. Copyright is not claimed as to any part of the original work prepared by a United States government officer or employee as a part of that person's official duties. For information on subscribing to the FNS Transcripts Database or any other FNS product, please email [email protected] or call 1-202-347-1400.