 Oh yeah, my name is Timothy Shipp, I'm with UNU Wider, and my question is for Kunal, and also I'd like to ask the same question to Hema's work, in relation to Hema's work. So it's about this concept of the motherhood penalty, and that's really popularly known as a contributor to the gender pay gap, but I'm wondering how this idea that increased child care responsibility for women influences the productivity gap that you've seen in India, and if in that paper that was examined or looked at, and then furthermore how the transmission mechanism of this from the gender pay gap to the gender wealth gap in property ownership might, how that might have an impact. Thank you. My name is Paula Herrera, I'm a professor at the Universidad de Adriana, thank you very much for this session, it was really interesting on the papers that you presented. So I have one question for Kunal, it's related to the questions they were raised before, but how much of the labor productivity that you're measuring is being affected by the fact that maybe women are having more home-based production in informality activities, and men are having their activities, although they are informal, they may be doing in factory or in the street, I don't know how it works, but maybe the place where these activities are taking place can affect this labor productivity. So I was wondering if you had this data in mind or if you did something related. And I have a question for Julia, which is, I know you said that profits increased by a lot, it was amazing, but I was wondering how much does that mean in terms of minimum wage or in terms of the spending of a household, because as far as I know for instance in Bogota you can have an informal or an entrepreneur who is making some profit, but if they are too small then these large increases might not mean too much and could also explain the fact that the nutrition facts for the children are not improving. But also other expenses, as you were saying, they use their profits in investing, again, I was a bit lost in that, I don't know if it's in their own business or in other business, but again how much of that will be related on actually the absolute value of what you are finding, not in percentage terms. And for Emma, I was wondering, you had this entrepreneur, well, line of your regression, but what were you hoping to get with that, what was the mechanisms between land ownership and being an entrepreneur in terms of the, I guess at the urban, at the rural level I could see it a bit more, but at the urban, what were you trying to, it was more financial access, I don't know, that was my question for Adriana, I don't have too much questions because the data is really rich and I hope we can take a look at it, so thank you. Hello, I am Mariana Villalias from Sedlas, I wanted to ask Adriana if with this service it is possible like to construct the intra-household story, for instance, you know if a woman answers the service, she tells you that she lost the job, she increased the time allocated to childcare, can you for instance separate these patterns depending on whether the husband lost the job as well, like to see what happens within the household, I don't know if that lost the job as well, so like to know what happened within the household, but I don't know if you have the information considering only one person answers the service. Hi, I'm Estefania from Universidad de la República, I want to ask also Adriana, because you show data for like for aggregate of Latin America, many information, and I think that if you look like at different groups like heterogenities across different groups, they may tell you like different stories, because for Argentina and Uruguay, the informality in fact decreases, so because there was like a change in the composition of employment so that the informal ones are the ones who are losing their jobs, so the formality increases, so maybe you can do like more heterogeneous analysis by different countries, and it will be nice if you could like relate this with the policies that were implemented in each country, like those countries that were where the formality was higher, they also do a lot of policies like related to formal labor market like unemployment insurance and this kind of things, so I think that this analysis will be interesting to incorporate. It's kind of related, but okay, thank you. I am Martina Garejeta from Uruguay, I have also two questions for Adriana, it was a really nice presentation and the data seems like amazing to start working with. You show us there is huge heterogeneity by country in this variable of shovelosses, the gender gap in shovelosses among countries, and I wonder whether are you analyzing like some correlation considering the stringency of the COVID lockdown restrictions in the country and how that affected the shovelosses for female and men. I'm also related with Estefania's question, if you are analyzing the correlation of these measures by country, considering the policies that were more effective in mitigating this increase in the gender gaps. And also I have some curiosity regarding the time use data you show us that you like show us that not only females but also fathers increase the time dedicated to caring activities, and I wonder whether you already analyzed the second round of the second phase survey and show if this change in the time use is persistent over time and only responded to the more hard time of lockdown and then that effect banished over time. Thank you. Thank you all very much. Thank you all for the questions. I think there was one from Tim that was common to both me and Kunal. In this particular paper, we didn't really look at motherhood penalty. There are legislations that you can look at that say whether you have mandated maternity leave or and if you have mandated paternity leave. And certainly that would affect women's ability to participate in the labor market. We have another paper that's looking at motherhood sort of penalty and how that is. And I think definitely that is one that can cause a labor market impact at a longer period and there's very good evidence that has come out recently that's looking at tax records, et cetera. And I'll just quickly also answer the second question before turning it over in terms of entrepreneurship. So what this entrepreneurship variable is, I think it was in a summary model. It's actually picking up four different aspects of the law in terms of whether a woman can open a bank account in the same way as a man can if she can get access to credit the same way a man can if she can sign contracts like a man can and something else. There was a fourth variable. So it was a summary of all four. And we used entrepreneurship only in the summary variable really trying to see if there was it like to ensure that we weren't omitting out certain significant characteristics in some ways. You can think of it as her ability to be self employed because women are more likely to be entrepreneurs right, whether small or big. And if this is really affecting her ability to start her enterprise and then maybe move up in her enterprises. I think that was sort of what we were hoping to see, but we didn't really find much significance coming out from that variable. So I think the questions that came from Tim, and I'm sorry you didn't get your name. It's a parlor. All right. I think those are kind of related questions in context of our paper. Because the first question is about essentially the time use and so childcare, which we don't have a measure of, but I mean the dependent variable is productivity, which is value of sales over a number of workers in the enterprise. Again, I really imagine that if you are a woman entrepreneur spending time looking after your kids, you obviously can't sell as much. You can't be involved in the market as much. So that is an observed characteristic in our data because we don't have that in the data set. What you could do, the data set that we have doesn't have that, but there are other surveys in India, the employment survey, which does have information about childcare or what we call extra domestic duties, anything like unpaid work or paid worker or childcare. So we can use, match that. I'm not sure how much we can do that on that. It's a very important point. It is not then in this data set. On the special location, I mean in a sense the industrial effects capture a little bit of that. If you think about a street vendor, the street vendor has to get out and sell his or her stuff, right? So it does capture what I think. So within industries, there is a difference. But within industries, someone is working from home, but somebody has to go out to sell the goods, right? Then I can see that might be the case. And clearly they would be within industry. Difference in whether you work from home or whether you go and work, whether you go outside the household. And that is the data set. So actually, I think I'll go and check whether that makes a difference even if we control the industry. Thanks, effects. Thank you so much for that question. Yeah, thank you so much for these questions. You're absolutely right that I mean these are ultra poor women, right? And I mean you might say going from zero to like something positive, you know, is relatively easy, but that's something that I would debate. Like in the sense that typically in these contexts and what we have about firms is that they make negative profits. So the fact that they even make something is still a very positive effect. Like it's still a massive income growth in this context. And I think what we want to show in on top of that is to say this is a population that is very comparable with other programs or like other papers that have been written, but the evidence is not even close. So in that sense, like what we want to say is yes, we can make them into productive entrepreneurs. It is a way of increasing income substantially. But, and this is in line with other papers as well, like from BRAC evaluations, what they find is to say in this short to medium run, we cannot use entrepreneurship trainings to lift women out of poverty. Like that's that's basically what this paper says. When we say like they are still living on less than a dollar a day in the long run, maybe like it's probably going to be a good program to foster private sector development and economic growth in general. And in the long run, they might be lifted out of poverty. But if you would, you know, like want to say, like, do we make them into productive entrepreneurs to counteract the pandemic and the impacts of that? Then at least this paper would say not necessarily because it might take longer to materialize because this population seems maybe less present by us that we think they think long term, they become entrepreneurial and they want to grow and they even even take it that they're going to have more in food insecurity in the short run. And then basically what you're asking about the reinvestment is an accounting exercise that we have been asked about a lot. The problem is that these are measures that are very, very different right, because I mean, we can ask about total savings and like how much did you spend on food yesterday and how much did you reinvest in general, but to make this comparable? I think it's like really difficult to go from a measure about saying how much did you make an income last month to then like divided up into where this money is going because then you have shocks, you need to have this money. But in general, what we what we find is some of the money that they make go back into the household because like household expenditures do increase by like a bit. It's just not high enough to actually, you know, say like there's a substantial increase in consumption that would make them and like not poor. And like that's basically what we're going with. But like, of course, some part of it does go back to the household, just not enough to actually make a difference. OK, so thank you for the questions. I'm going to answer whatever I wrote here and maybe have you if you want to compliment something. Regarding the inter-household story, we have questions. So we are not asking everything about the partner, but we have a question if they have a partner or not. And we know if they are working on it as well. We don't know. We know if they have a higher income as well, I think. So so that's something that we that you can control for if you're doing this is just descriptive and these are just tables and tabs. So if you're going to do regressions and things like that, of course, you can include all those controls. And so as we know about house and, for example, in the household, we know if they have kids which are not their children as well. So some of the things you can you can you can look if the kids are not the children or and how many other like like we have the composition of the of the household in terms of if they are older people in the household or not and things like that. So it kind of constructed, but it's a phone survey and it was like 30 minutes and we needed to ask as much as we could. So, yeah, you can construct a little bit of the story in terms of of Argentina and informality decreases and things like that. Well, as you might know, Argentina has in a very different way informality compared to other countries, right, in the household service. Like there are differences in terms of like I remember that we were there and they told us that they cannot ask if they are if the firm is informal, if you are not paying social security because you will have to tell that they are doing something that it's illegal. Yeah, so in the in the in the way they are getting data in the way that household service are being collecting data is not asking other countries that you could ask, are you paying for pension or are you not paying for pension, but we can talk about it later and you can tell me about this. And yes, we have notes by country. So we have but for the first round, we have the notes that we did with the with the World Bank and we have a second wave of notes that we also processed for the second round by countries. And you can see a big difference, for example, as I was telling in Argentina, the opening of schools was really was really big, it was a big change between the mid and the end of 2021. And you can see there that the entrance, so the inactive or the ones who didn't have a job are entering a lot into in the formal or informal sectors, but employment was activated by the starting app of school in in-person school and not a virtual school. So you can see all these details that are different country by country and it's interesting to look at it that way. And it's also easy to do it because the data sets are representative by country. So you could just take your own country and go ahead and do any analysis. And we have like the data is between 1,100 and 1,200 observations per country, but for example, Mexico has a much higher sample size or Haiti has a big sample size as well. In terms of job loss and gender gaps and the lockdowns and things like that, these are all the questions that we have to exploit and that you can exploit as well with the data. Like what we would want is this is a project which was really long, involved a lot of people, involved lots of money. So I think the best way to exploit it is that as many people as possible can use the data. So you just go there and download the data and you can do as many things as you want. I have been doing some kind of correlations, for example, with education I have been looking at the numbers. And the thing is that I believe that the countries started like in a block, like politicians needed to take decisions which were very difficult. So a lockdown, so everybody did the lockdown almost at the same time and everybody opened almost at the same time. So for us it was like, yes, a week, a month, but in the data you cannot see it as much of a difference because they were just like started in March and ended in September, some October, some Brazil, for example, had a smaller lockdown. So you can try to test, but there are like three groups, let's say. So there's not much variation in terms of the policies just to be able to do analysis of differential policies compared to the outcomes that you got. But the data is open to do whatever and like very interesting questions are there. And regarding the second round of service, yes, we have done work on looking at the information and we did check the question about the time use. It's not that we have, you spent how many hours doing this, but if it increased, decreased, yeah. And what we see is a slight change in the increase, in the report of increase of time spent doing household activities. So it increased a little bit for household in general and it decreased for childcare and education and assistance to education. And in the countries that you see the most reduction in assistance to education is when the school in person is more than 90%. So you see there like a big drop in terms of how much the time spent assisting kids is taking time from the individual that we're analyzing. I think these are all the questions. Javier, do you want to say something? Something else? No? Yeah? Okay. Thank you. I think we have time for a few more questions. I actually don't have a question, but just to add. Hi. All of these analytical work is already available online at the UNDP website in addition to the data. All of them like the joint work with the World Bank and some of the notes that are not joint work. Anyone else? If not, I have a question for Julia in terms of how you measure food security. And you said it had not really increased and I think also child, the high school, the school index, right? And the time between your end line and your baseline was, I think, what, 18 months? Yeah, so I was curious to see how that was measured and isn't it a little optimistic to expect that the school index would actually improve? Is that not too short a time? Because we don't know what, if they are very ultra poor, what kind of deprivations or what it means in terms of cognitive development. So it seemed like you drew true stronger conclusion and you said there's been no improvement. So food security or like insecurity is just the factor of like, have you experienced not enough food in like the last week, right? Or like the last month, I think it is. And it's not about them experiencing more food security but that they have experienced more food insecurity. So like the people in the treatment group at midline, so in the time where they make heavy investments, they're like, yes, we had more periods where we did not have enough to eat. And then school, the schooling index, I agree, like is this not necessarily like improvements in your school and education performance, but like, do you go, like it's an index of a bunch of questions of like, how much schooling you get? And I think combined with like the pandemic, it's likely not to find anything there. But we're still saying like, okay, so like we have these high incomes, nothing happens with the kids, like it's not because it also has measures about like how much you aspire to in terms of schooling and things like that and we don't find anything there. But yeah, it might be, like all of the household and children's outcomes are on the caveat about saying like, maybe it just needs more time to materialize, but we say as one and a half to two years is not enough to see anything. So like, yeah, but you're right. That is, yeah, that that's the statement for sure. Yeah. Any last question? How many much minutes? Maybe I have a couple more minutes. Okay. One quick question. Yeah, first, Adriana, I was just curious about the ownership of cell phones across these countries, right? Because poorest workers probably don't have cell phones. I don't know what the ownership rates are in Latin America and they are usually informal sector and of course informal sector rates varies across the countries, right? And that will lead to kind of a major problem with representation of the data. So I just wonder whether you've corrected for that. And if there is a difference in ownership rates of cell phones, especially among informal workers across these countries, we do have data for that because the problem I have with phone service is that you have to have a cell phone, right? And at least in South Asia, South Africa, that's not the case that everybody has a cell phone and usually the poorest don't have cell phones. So just wondering about the representative nature of the data because using cell phones and not always national-represented household service. So. Okay, regarding the cell phones, we have like a list or a census of all the, the random digital dining what it's doing is that it's using a robot and it's just styling all the numbers, all the possible numbers so it's getting the whole thing in the country. But what you're saying is if there's a lack of cell phones in a part of the country, I know this about Colombia, which is really high, the number of cell phones, but I don't know for the rest of Latin America, I think it's also high. But the other thing is that we also have landlines. And the other thing is that we have, for example, some countries where they're like, like the initial number is guiding you into a region, yeah? Like a zip code, let's see, like a number that it's directing you to a different geographical area. So we also have that the sampling is adjusting for that information to be like well allocated or distributed between regions, but I don't know the coverage. I think there's no problem in terms of coverage in Latin America, right? Javier? Yeah, for older than 19 with a cell phone. Yeah, that's the restriction that we're imposing initially as well. Yeah. Data against what we know from household surveys. And so that's a reason why we don't have a rural urban analysis for all countries whenever like we didn't get the correct numbers. So we did this check because we were, we had that concern even though the sampling was done so that it would be representative with no problem. Let's say the number of observations would correct a little bit for the mistakes that you have in terms of the sampling. All right, thank you all very much.