 Thank you so much for being here. This is joint work with Megan Lange from the World Bank and it's called the economics of women's Entrepreneurship. It's very much related to what we're speaking about now So this work is motivated by the idea that typically we know that women Besides having lots of other inequalities have very few employment opportunities in low-income settings So that raises the question. What can we do to close this income gap? We have evidence about saying that's changed gender norms about female labor force participation But often there's also the policy idea about saying maybe what we should do is allow women to become Entrepreneurs and specifically productive entrepreneurs now How as Kono was saying like what we find very often is that when women are entrepreneurs? There's a huge gender gap to to men in the sense that their businesses grow very little Additionally, we know that in general Microfirm micro firms grow very slowly or not at all and have very low profit margins So additionally What we care about is to say if we do increase income for women Does that also have impacts on households that might be potentially very large given what we know from the literature on gender spending patterns? But then we go to the practical results and the empirical evidence about saying we know that this is a concern But how could we actually solve this problem like what is the evidence that we have? Unsuccessful entrepreneurship programs and those are results are typically quite mixed So what this we do in this paper is like broadly asked the question Can we make women into productive entrepreneurs and then if we do what happens? Specifically the question is if we have a targeted entrepreneurship program that Ames at ultra poor women is that going to deliver on the dual goals of business development and gender equity? And then we are interested in this bill over effects of women's entrepreneurship on the broader household and additionally We care about not only the household in terms of expenditures and income But we also care about whether there are any effects on children that are affected by these entrepreneurship programs So what we have is a Study in central Uganda where we partner with a research organization that has developed an entrepreneurship curriculum Tailored to these ultra poor women and they have typically very little formal education So at baseline about 50% has a business But there was a far away from being productive and businesses in the other half doesn't So the program gives them eight two to three hour modules on business skills And they are also in the status quo three intensive one-on-one mentoring sessions where coaches would come home to your home Come would come to your home or business and like Intensify what you have learned Importantly what this program does is something that we call like a three pillar approach to firm or entrepreneurship training where we say They teach better practices They teach you how to be more entrepreneurial So how to identify business opportunities how to start new businesses and kind of like figure out How you can be this innovative person creating firms and then importantly it has a large Psychological component as well. So it works a lot with aspirations. So they are role models There's a big graduation ceremony where you go through the village and show that now you are like a successful business owner and it works a lot through increasing persistence and grit and Resolving psychological constraints So now what we were doing is to get at this idea and like to add that answer about saying can we make them into productive entrepreneurs? Is we have two different versions of the program in five locations? For one third of the women we have business as usual So you get the training and then you get the three home visits for another third of the program We do not have the three home visits, but we gave them office hours So effectively you could stay around after the training or there was like an office our session Where the coaches were there and you would travel to the training location and you could ask questions And then we have a peer control group So this is an RCT with lots of survey data Where we have outcomes on businesses households and children at baseline exit which is six months after the baseline and then at At end line which is 12 to 18 months post-exit It was supposed to be 12 months for everybody, but then covered happened and we couldn't go back. So that's why It's different lengths in terms of exit But we feel like that they were able to recover enough such that these time periods are Comparable and then we have some sort of administrative data where we have weekly SMS data We ask about revenues over these businesses So to give you a quick review of results because time is very limited What we find is that yes, we can make women into productive entrepreneurs with this program Because we have large and persistent improvements in business outcomes So we increase the number of women that has a business, but we also increase the number of businesses each woman has Importantly Making this distinction between the main business that you start with and then the additional businesses that you create Throughout the survey period you first increase your profits by a hundred and two percent in your business main business compared to the control group and 88% Compared to the control group in all other businesses. So this suggests a huge income increase Now what the program is also does is because these women become more entrepreneurial What they are doing is using these profits to reinvest So then they reinvest and they have a hundred and twenty two percent higher investments in the main business at midline So when they finish the program and hundred and thirty six percent in other businesses at end line So one and a half to two years after we started working with them now if You reinvest the question is well, we may be increased incomes great They are now Productive entrepreneurs, but is that actually going to contribute to poverty alleviation and what we have to find here is no These women remain ultra poor. Okay, so they still have less than one dollar a day in expenditures per household member At some point we even find increased food insecurity because their money is fixed in the assets of these businesses And then what we find in terms of children is in line with no results on the household being better off We do not find direct effects or spill overs on the children from having a mother Being now an entrepreneur, but what we do find is there seem to be a specifically for boys Effects on their leadership index through living in a community that now has a different business environment and where more women become entrepreneurs So we believe that with this we contribute to Four strands of the literature more or less the first one is Unentrepreneurship programs in general where we show like this is a successful entrepreneurship program Compared to too many others We provide novel evidence and saying like why would it be that we maybe do not observe Microfirms growing into SMEs it could be because firm owners actually do not have a Vertical investment strategy where they grow their one business, but there might be cultural constraints or risk preferences that make you Reinvest horizontally which then means you're actually not going to grow into one bigger business and then our household effects suggest That it might be important to also provide consumption support at least in the short to medium run if we want to Encourage people to make long-term productive investments and we have these small evidence as well Unlike what happens if we change business environments and contribute here to this role model effects literature for children Okay, so quickly about Compliance there's not Not many concerns in the sense that any non-compliance that happens that is relevant as relatively similar in both treatment arms And this is our empirical strategy where we try to recover the encova estimates of the intent to treat and Importantly what I would like you to focus on is beta one is going to give us the indicator or like the Estimate of being part of the program and then beta Two is going to be the additional effect of having had this intense mentoring home visits And for the children what we care about is we're not going to differentiate between the two types of treatment We're simply going to say theta one is going to be our direct treatment effect from living with a woman That has participated in the program and then it is the additional effect of being connected to a woman That has participated in a child's social network So importantly here this is data that we get from the kids directly So we had photo books and asked all the kids do you know this woman in the sample and how are you connected? Okay, so to show you results These are Inverse hyperbolic sign transformation. So these are percentages. This is you should think about this as like 200 percentage change Compared to the control group. So what we can see is that at midline Sales and profits in the main business are extremely high, right? Like this effect is is massive, but not so much happens in the business in other businesses yet So at end line Sales and profits in the main business are still significantly higher than compared to the control group And now we also find that they get profits from other businesses. This is coming from these dynamics and saying make They don't say if like they have no savings. This is basically a zero effect But they are business assets that they come invest in so at midline They first invest in these business assets of the main business Not in other businesses and then at end line. They invest much more in other businesses Keeping some something to note is that at end line the control group catches up significantly So we do some spillover analysis and depending on to how many women you are connected with that have been treated Through business networks at baseline. You actually benefit quite a lot. So These treatment effects are gonna underestimate the real treatment effect given that the control group Benefits from being connected with treatment women as well Then something important is this so daily household expenditures increases Barely at end line and not statistically significantly So and nothing at all at midline and then something interesting is that we even see a negative effect on remittances that these Households receive so it kind of seems like oh now you're an entrepreneur. I'm going to support you a bit less could be a potential story So what are the mechanisms of why we believe that this program is so? successful besides being Targeted to these women and developed over a long time so we find that they are in general in And specifically on grid to take significantly significant increases at midline at end line These are more or less there and also it improves of the locus of control and we have some additional Qualitative evidence where these people say because I participated in the program. I now believe more in myself I have more self-confidence and It relaxes these psychological constraints In terms of business skills the program does teach business practices better and we find this more or less in our data specifically on track price management and Effort in terms of work hours in the main business we find some evidence That this is what supports the potential like support Potentially supports the effects that we find on business outcomes and then as in line with the literature that exists on the long run Typically business practices kind of fade away But it's also because we know that business practices is knowledge and knowledge is simply very easy to pass on so it's much more difficult to Not have any spillovers here So quickly the effects on children driven by is rather driven by community changes rather than Changes in the households if you look at this It's for boys the direct effect and the indirect effect and then for girls as well So what we were hoping for is that we would find all these great effects on girls because now you know Your mother is a successful business owner. She's generating more income like you become inspired But unfortunately we find relatively little of that So some impact may be on leadership, but we cannot precisely estimate it nothing on schooling outcomes and And then for boys very interestingly We have this indirect effect of saying if you now know more women in your social network That have been in the treatment group so that are more likely to be successful Then that increases leadership among boys. So in general what this picture says is The story is from the household not much happens, but there seem to be some effects at least from changing the business environment and locations Okay so to wrap up and to conclude what we learn is Yes, we can make women into productive entrepreneurs with a mix of business girls entrepreneurial skills psychological Skills or like from constrained reliefs And that is highly effective of improving business outcomes and therefore generating higher incomes for women Something that I haven't shown you here, but that is in the paper is What we also find is that there's more intensive mentoring where you get home visits is actually not a great idea So in the sense that it does not augment the effect of the core program and if it's like more costly to do home visits then You would want to cut this And in some cases we actually find that it's counterproductive because it has a negative effect on the psychological components We do not like despite higher income We do not find that they are being lived out of poverty and that the households are better off and then Importantly profits are reinvested because they are more entrepreneurial and I have to stop here