 Well, thanks everybody for being here. The paper I'm about to present is called Motherhood and Flexible Jobs, Evidence from Latin American Countries. This is joint work with a lot of colleagues from Argentina. Inez Berniela, Mariana Marchioni from Sedlas, Universidad de la Plata, and Lucila Berniela and Dolores de la Mata from CAF. And I am Maria from Universidad de San Andres in Buenos Aires. Let me give you some motivation. I think nobody needs motivation in this room, but just in case. We're all very well aware of the great convergence in terms of gender roles for men and women during the last century. Yet we also know that very large gender gaps exist, especially in the labor market. And we know that these gaps are particularly large in Latin America. And the literature recently has identified motherhood as a key factor here. And basically, motherhood is identified as the last hurdle to be overcome if we are to finally achieve gender equality. And so basically, there has been a lot of papers that have been able to identify the causal effect of the arrival of the first child. And there are a lot of nice papers, but basically these papers are focused on developed countries. So we get a lot of information about the effect in labor market outcomes for developed countries in the country for Latin American countries or for developing countries in general. We have very little information. And the question is, does it matter? Should we care that we just have information for developed countries? And we think this is important. We think that extrapolating these results from developed countries to developing regions might be misleading. Why? Because as we discussed today, we have very different cultural and institutional backgrounds. And so it is important that we provide specific evidence for developing countries. OK, so this is what we tried to do in this paper. We tried to do three different things. First, we provide evidence of the causal effect of the arrival of the first child with comparable data for four countries in the region. These countries are Chile, Mexico, Peru, and Uruguay. And we assess this effect on a variety of labor market outcomes. We assess this effect on employment, also on changes in the occupational structure, so part-time employment, self-employment, and informal jobs. And lastly, we try to provide some suggestive, descriptive evidence on the role that social norms and family policies may have in these motherhood effects that we are estimating. So let me give you a glimpse of the methodology and data we are basing our analysis on. We are using an event study approach around the birth of the first child. This is the standard methodology that's been used in these developed countries I mentioned before, basically led by the first paper by Cleven and Coethers. And this event study approach requires panel data. We need to follow men and women across time. We need to follow their labor market trajectories. And we need to know the exact date of the birth of the first child. So we need all of this information to be able to estimate the causal effect of the arrival of the first child. And it's important to know that we need people to become parents at some point. So our sample is composed of men and women that become parents at some point. And another thing that's important to bear in mind is that we are only capturing post-birth effects. What this means is that maybe there may be some anticipation effects. Women may be making different choices anticipating the fact that at some point they will become mothers. We are not going to be able to capture these effects with this methodology. And of course, we have this to rely on an identification assumption. And this identification assumption is that the timing of the event is not correlated with the labor outcome we are looking at, conditional on having a child in the sample period, and on the included controls. OK, so this is the equation we are estimating. Let me walk you through a little bit. On the left-hand side, we have our outcomes of interest. So y is our outcome of interest for individual i in time t. y are going to be four different labor market outcomes. It's going to be employment, self-employment, part-time employment, and informality. And then on the right-hand side, the first term is the most important. Here we have a set of dummy variables that what they're doing is that they're indicating the relative distance to the birth of the first child. So basically, this beta tau that's there, this parameter here is our parameter of interest. And this tau, this sub-index tau, is measuring the time relative to birth. So whenever tau is zero, we are identifying the time of the birth of the first child. Whenever tau is positive, we are identifying periods after the birth of the first child. And whenever tau is negative, we are capturing pre-trans. And we are including a set of controls, age, and calendar months, and year-fixed effects. Data sources. As I said before, we require panel data, very intensive panel data. And this is a very unimportant limitation to try to explore this issue in Latin America because there are very few surveys that account for this. So we are able to do this for these four countries. I mentioned Chile, Mexico, Peru, and Uruguay. These are the sources we use. And all of these surveys are nationally representative. And they have the information we need. That is the exact date of the birth of the first child and the labor trajectory of men and women before and after the child is born. And these are the results we get. I am going to present all results in the forms of graphs, what is typically done in this literature. So let me walk you through the first graph. Let's concentrate in Chile here, so that you get an idea of what we're looking at. And then I will repeat this for the other results. So we are focusing on employment. This is the first labor market outcome we are looking at. And basically what you have on the x-axis here, you have tau. So you can see very well, but here tau is equal to zero. So basically here, we are looking at the birth of the first child. Everything that is to the right of tau equal to zero are all periods after the birth of the first child. And whatever is to the left of tau equal to zero is a region. And what we have on the y-axis here, we are plotting the pedas. So each of these dots here in green, we are plotting pedas for women. And in orange, we have men. So basically what these pedas are capturing is the difference in the outcome of interest, in this case, employment. And sorry, I forgot to say that we are using as an omitted category tau equal to minus 12. So we are looking at one year before the child is born. We are comparing the outcome of interest in each period to what happened one year before the child is born. So basically, we are measuring here the change in employment relative to the employment one year before the child is born. And so what we see for Chile, and then this is repeated for countries, is that as a child is born, we find a very large drop in employment rates for women. Nothing much seems to be happening to the fathers. And the same is repeated for Mexico, for Peru and for Uruguay. I didn't mention also that here for Peru and Uruguay, we have yearly data. Instead for Chile and Mexico, we have monthly data. That is why you're looking at graphs that look a little bit different. But basically in all cases, we see this sharp drop. For women, nothing much seems to be happening to fathers. And we see that this persists over time. On average, the magnitude of the effect is very similar across countries, between 17 and 20 percent. Yes? These are the veras, the change, the difference between the period, like that equal to like two years after the birth of the first child, compared as the employment one year before. Sorry? And yes, we're including age as a control. Yes, in Mexico, right? Yes, yes. Yes, yes. In the present, here in the present, right? Or after the child is born? Yes, okay, okay. That's before, do you mean this, right? Yes, we're focusing on this large drop that's happening here. Like after the birth, what I meant is after the birth, but actually in Mexico, we are seeing something. But we're concentrating. Sorry? Okay. Yes? Okay. Okay, so this is employment. Then we go to part-time employment and we find a similar pattern. You won't find Mexico here because Mexico does not provide hours of work. So for Mexico, we cannot try this analysis. And we see here that part-time employment is increasing for women, but it's decreasing actually for men. And this is significant for Chile and for Peru. We don't find significant results for Uruguay. Then we have self-employment. And here again, women seem to be taking up self-employment in Chile and in Mexico. For Peru and for Uruguay, we do not have the results are not significant. But in the case of Chile and Mexico, they're taking up, they seem to be taking up self-employment after the child is born, first child is born. And then we have labor informality, which is an aspect that maybe it's not so much considering the developed countries literature. And basically we find here that labor informality for women is also increasing after the birth of the first child. And this is significant for all countries. For Chile and Peru, the effect is around 16, 17%. And for Mexico and Uruguay, the effect is of the order of magnitude of 50%. It's very large. So basically what I've shown you is that we find a significant impact of motherhood on labor market outcomes of women. Many women job out of the labor force when the first child is born. And for those that remain in the labor market, many change their occupational structure towards self-employment, but time employment and informal jobs. This comes at a very high cost for women. For women, of course, they are giving up economic independence, they're giving up social security, they're giving up future career prospects. And of course these choices have also a positive side. They're offering flexibility to women. And so the question is, why are women willing to trade these present and future benefits for time flexibility? And so what we do in the last part of the paper is to try to explore, to provide very suggestive descriptive evidence of the possible drivers behind these choices or at least two possible drivers that are gender norms and family policies. And of course we know that this is very difficult to do. We have a lot of endogeneity issues here and so they are very difficult to overcome. And even further on we have very strict data restrictions for the Latin American region. And so we decide to go on a descriptive route and that's okay, but even if we stick to a descriptive analysis we find further challenges because ideally what we would like to have is to have these causal effects for all Latin American countries and make some correlations with gender norms and family policies for the whole set of Latin American countries. But of course as I told you, we are only able to produce these causal effects for four countries in the region. And we think that it would be a little bit misleading to base our correlational analysis only on four countries. So what we do for this, the exercise we do is to rely on an approximation of this causal motherhood effect. And we are not looking here at causal effects but rather we are computing the gaps between mothers and non-mothers in labor market outcomes. So the difference between mothers and non-mothers in employment, in self-employment, part-time employment and so on and so forth. And we use this as an approximation of the motherhood effect. We call this the motherhood gap to make a difference. And we correlate these gaps through the deep prevailing gender norms and family policies in the region. We take the data from SEDLAC, the database from SEDLAS. So what we do here is this is just a set of, as I said, very suggestive and descriptive evidence. We have made no claims of causality here. We're just connecting these two different issues. And again, I'm going to present this in the form of graphs. So let me walk you through them. So you get an idea of what you're looking at. We are here, we are relating gender norms to this motherhood gap. Remember the difference between mothers and non-mothers. And basically what we are looking at is one gender norm, which we elicited from the Latino barometer. And the gender norms is the level of disagreement with the statement women should work only if the partner does not earn enough. So what you see here is basically that in all countries, for all results, we find that the more conservative use the country held, the larger the gap between mothers and non-mothers is. So the more conservative a country, mothers are making choices that are really different from non-mothers. And the less conservative, they make similar choices. Then we do the same for childcare availability. And here we plot this motherhood gaps for the four outcomes of interest to preschool enrollment rates for younger children, children three to five years old. And once again, what we see is that it comes as no surprise, of course, that when enrollment rates grow up, we just saw it with Veronica. Basically mothers tend to make much more similar choices than non-mothers in terms of labor market decisions. And finally, we do the same kind of analysis for maternity leave. Basically we are correlating here the number of weeks of protected maternal leave with this motherhood gap. And once again, we find that the more generous a countries in terms of this maternal leave, the way women with children and without children make decisions in the labor market regards to this, regarding this four labor market outcomes is more similar. So just to sum up if you got lost a little bit in the presentation, basically what we try to do here is to estimate the causal effects of motherhood on four countries in the region. And we find significant reduction in employment and a significant change in the occupation of structure. This is per instant over time. Women seem to be moving on to less quality jobs. And basically we try to answer the question of why they're doing this. And we just provide some suggestive and descriptive evidence regarding gender norms and family policies that may be to some extent shaping the decisions of women. And let me conclude by saying we're pretty sure that there's a lot of room for designing better policies. We know that these policies are really tricky to design, but of course we need to design better and more policies for alleviating time constraints for family if we want to achieve gender equality in the region. Thank you very much.