 Thanks, everyone, for being here. This is a joint work with Javier Jara from LSE, Maria Cecilia Deza from Inter-American Development Bank, and Mariana Dondon Javier Torres. Javier is from the University of Pacífico in Peru. Mariana is from the University of Rio Negro in Argentina. The idea here is we are trying to account for the tax-benefit system, how the tax-benefit system affects the gender income gap in several countries in Latin America. As Javier previously explained, we have been building a series of models, tax-benefit system models for countries in Latin America. Today we are going to present some results for seven countries, but we have models for nine countries in the region. The main motivation here is that when we talk about gender income gaps, we usually talk about wage or earnings income gaps. This is especially important in Latin America because, as we know, we have lower participation for women in the workforce. Additionally, if women participate in the labor force, they are usually working in the informal employment sector. We know that in the informal sector, they are going to have lower earnings than males. The other thing is that this is true for every country, I guess, that we have these gaping incomes in earnings between women and men. We have this problematic picture for Latin America, so we expect these income gaps to be higher here than in more developed countries, but the idea is we have the tax-benefit models for the region and we want to see how the tax-benefit alleviates or improves or reduces this income gap. What we are doing here is passing from original income or market income to disposable income by means of this tax-benefit system. The idea is that we compute the original income. We apply social insurance contribution. We deduct the payments that workers have to make of social insurance contribution. We add benefits and we subtract taxes, so we end up with disposable income for each of the people in the service that I'm going to present next. What is the motivation here? We expect that the tax-benefit system reduces this gap by means of two ways. In the region, there are several programs that are allocated by default to the model. For instance, as there are public here from Colombia, we have a family action. Family action is a cash transfer that is allocated to the model inside the household. This transfer is going to be received by the model, so this is going to reduce the gender gap. We have the same for several countries in the region, in Argentina, as in the Universal Porijo, the same is allocated to the model. Bolivia, one of these bonus that they receive is allocated to the model. Colombia, Ecuador is the same, so we expect this to reduce the income gap, sorry. In the region, if the gender income gap in the region, and the other effect that we want to test is the effect of income tax. We know that men earn more, so they are going to pay more in income taxes, so this is going to reduce their disposable income, and therefore the gender gap is going to be reduced. We want to see if these two effects are important, and how much they help to alleviate the gender gap in income. The idea here is we have models for Argentina, Bolivia, Colombia, Ecuador, Peru, and Uruguay. We have also the model for Mexico, but we are not going to include the results in this presentation. We are going to analyze 2019. For most countries, we have information for this year, except for Mexico, that we use a similar method that was described before to update incomes from 2018 to 2019, but for most countries, we have information, household survey data, that we are going to use in the micro simulation models that we are going to present. So the idea is the final result of this research is going to present results for the period before the pandemic and the period after the pandemic. I'm going to show you only results for the period before the pandemic. Okay, so what is the idea here? We have three strands of the territory in this measurement of the gender gaps. So the traditional idea that if you go to earnings, there are going to be discrepancies in the earnings between women and men, so we have a very old territory there. The other thing that we are going to consider, and I'm going to explain this later, is that maybe in intra-household, the resources are not going to be allocated equally for all members of the household. So the idea here is we are going to, and I'm going to show you that we are going to allocate the resources to the earner, and we are also going to show you some results assuming that we are pulling incomes for couples and also for households. So the idea is inside the household, we have to allocate these resources, and probably the best idea is not to allocate the sum of all resources and divided by the number of members in the household and assign this income to all household members, but probably to have the resources allocated to the person that is earning them, mostly. And most recently, we are also building on the tax benefit systems that we have created. We, Javier and Rodrigo, explained about Euro mode, we are going to build on that. So the idea is like trying to combine these ideas for analyzing the effect of the tax benefit system on the discrepancy in earnings between women and men. So previously, we were combining these three things, we can find papers mostly for Europe. We are going to follow mainly Abram and Popova, that is here. So we are going to follow mostly this paper, but we found other literature related. So for instance, Figuari analyzed the effect of the tax benefit system producing these gaps inside couples. Gallego Granado and Darlene Kinn analyzed in the first case for Germany and in the last case for Europe the same. So it's trying to see the effect of the tax benefit system. And Kinn, what they try to do is to separate these gaps, to decompose these gaps into the effect of the tax benefit system and the effect of differences in wages and also differences in hours of work between women and men inside the household. So we are going to follow especially this Abram and Popova paper, this is the methodology I'm going to explain a bit how this methodology works. So we have all these models based on the EuroMOD platform. So they are fairly available if you want to access them. Most of them are available through SouthMOD, to the SouthMOD platform. And they are based on nationally representative household surveys, for instance in Colombia we have a gran inquesta integrada de hogares and the idea is we have built all the legislation, all the taxes, all the benefits, all the social insurance contributions in top of this survey. So we can simulate for instance if we introduce a new cash transfer, for instance in the pandemic ingreso solidario here in Colombia or maybe with the new tax reform here in Colombia we can simulate how much they are liable to pay for each observation in the data. So we have these models, these are like a description of the data, household surveys. And the other thing is that the countries that we are using here are very different between them. So for instance the labor force participation is 46% in Mexico but 71% in Peru. So the idea here is there is a variety of countries in our sample. So I think this is important for external validity for our exercise because if you want to for instance take another country from Latin America and analyze, you can just compare with some similar country in this range of countries that we are presenting here. So the idea here is I'm going to present three sets of results. Well the first one is a minimum income polling. So the idea here is that suppose that we only focus on the earnings and the non-labour income of each individual and we apply just this tax benefit system and compute what is the disposable income of this person. The alternative is intra-household, we pull the resources between couples. So for instance if the male inside a couple of women earns 120, we pull this and the average is going to be 110. So and the alternative and this is the last option is more related with how we measure poverty in Latin America is we pull the resources from all members of the household and we compute some, so we found some per capita income inside the household, not a couple but inside the household, so including elders and including a child. And the idea is in this first alternative, minimum income polling, the idea is we are going to allocate the resources to the individual that receives earnings mainly or for instance taxes. Taxes are paid mostly at the individual level in Latin America so you don't have to, you cannot declare joint living between husband and wife for instance. So this is nice for us because the tax is allocated directly to the person that has to pay it and social insurance contribution is the same, you have to pay some percentage of your earnings as social insurance contribution. Benefits, this is a little bit more complicated, we assign for instance if it's unemployment benefit but the person that become unemployed receives this. If it's social assistance as we saw before, some of these social assistance is allocated to the mother for instance, so in that case we allocate them to the mother but in other cases these social assistance are received by the household but not by a specific member so what we do here is we divide the split equally the benefits among the parent and mother so inside the household. The other thing that we do here is we have these incomes and we have to take into account that the household composition affects what this income is going to be used. So what we use here is a modified OECD scale so the idea here is that we recognize that there are economies of scale in the provision of consumption goods inside the household so the idea here is that we assign different weights to different members of the household. The other thing is that I'm going to present you some statistics that measure this gender gap. We are going to use the income of women relative to the income of men so the idea here is that if this ratio is above one, women are in more than men and if it's below that is mostly a case the idea is that we are measuring with this ratio the gap. I already explained this so let's see the results. Okay so there is also remember we have a ratio of one that's the ideal so a ratio of one implies that men and women are equally this is disposable income so it's after tax and benefit income so what we see here is that and the other thing is that we have disposable income with circles and with squares we have market income so for instance this ratio of women to men income in the case of market income so before benefit and taxes is around 53% in Argentina but with the in the case of disposable income is above so it's 67% so it's the idea is here the message here is that the tax benefit system reduces this gap in income so this is the case for most countries the effect is bigger for Argentina worldwide in the case of Peru there is a negligible effect of the tax benefit system in the case of Colombia the effect is reduced but this is not as small as Peru so these are results for all population. Here we are trying to decompose to analyze this gap again but considering household income designs here we are using disposable income to construct these designs and what we see here is for the bottom of the distribution the tax benefit system reduces the gap considerably for instance it reverts the gap in Argentina for the first design before the tax benefit system this ratio is 0.6 but after the tax and benefit system the effect is the ratio is above 1 so it's what we find here is that women are receiving more disposable income than men on average and the main message here is that this effect of the tax benefit system is higher at the bottom of the distribution but in the higher part of the distribution as you can see in almost all countries except Uruguay than Argentina the effect of the tax benefit system is almost null the most important one is at the bottom of the distribution and this is possibly because of this idea that the cash transfers are mostly allocated to the mother inside the household in this next set of results I'm going to divide the population the families of the households the households in our survey and the members of the household so for instance here we have working age the effect is not as as as big as before the in the case of the elderly here the the effect is is is strange because the effect the ratio with disposable income is below the ratio with market income so the message here is that in the region pensions are allocated principally contributory pensions are allocated mainly to the men so and this is and this result is historical and the idea here is that in the past men work at men work more than women they work in formal sectors they accumulated so they are receiving now a pension that more often than women so the idea is before the the pension we observe this ratio and before the back the pension we observe this ratio and after the pension we observe that the gap increases between women and men here is another another take of this idea that cash transfers are moving our our results we see here the we divide the the population and we have here couples which without children and we see that the effects are of the tax benefits system are very small but we go through couples with children and we see that the effect is higher so this tell us that the fact is mainly driven by cash transfer that are allocated to to the mother inside the the household and in in this other in this last graph the idea here is that I'm going to take the same definition of income and the individual level in the second in the column see I'm going to allocate I'm going to join the income of the couple and in the last column I'm going to age I'm going to pool all the resources of the household so here the idea is that we have also in in this case we have poverty for women and men so this is not the ratio again but this is the incidence of incidence of poverty for women and men so we see for instance here for Argentina that if we take income into account just individual incomes we observe that the poverty incidence is higher for women and lower for men and if we join the resources of couples so this difference disappears and also if we pull all the resources of the household the the incidence drops and is the same for women and men so this is the this is the the main result what we what we see here is that if we analyze these gender gaps at individual level it's going they are going to be huge but if we pull the resources between members of the household this the incidence and the difference between women and men are going to be reduced so finally the idea here is that what we found here is that the main measures is that pensions are important they are allocated principally to men so to elder men and they are going to increase the gender gap cash transfers are going to be allocated to mothers so they are going to reduce the gender gap and the other thing that we found is that the the effect of taxes is very small this is especially true because in the region most of the people do not pay income tax so the effect of taxes that we discussed before is almost null for for the region and finally the next steps is we are going to analyze this in the context of the COVID pandemic in the COVID pandemic in most of the regions we have new cash transfers we expect these two cash transfers to reduce the the gender income gap because most of these transfers are going also to be allocated to the mother so but we want to quantify the the effect of these new cash transfers available only during the pandemic thank you