 Good morning everyone and apologies for not being able to join you in person. I had an issue at home but and I had to cancel my trip at the last minute but thank you so much for the opportunity, very happy to be here. Today I'm going to present our study on gender inequality, poverty and situations of displacement in Somalia. Basically this is a joint paper with Lucia Hanmer who is our team chair today and Juliette Santa Maria from the IDB. What is the motivation for this study? Well there is a growing majority of literature in development suggesting that safe displacement causes disruption in household and family structures which is related to specific gender disadvantages because of inequalities that we see in the different contexts. And this of course will have effects on differences in poverty that are not easily accounted for. In this context we basically aim to answer three research questions. The first one is whether female-headed households are actually poorer than non-IDP households, both male and female-headed and this is basically taking the traditional approach to analyze poverty using the headship concept which as you probably know has been already contested in the literature because it's not necessarily the best comparison. These households tend to be systematically different. The second question is whether there are specific households IDP households living in and outside settlements that are more likely to be poorer than the non-IDP counterparts and here we take a step forward to go beyond the the headship concept and propose an approach also building on previous studies which takes into consideration alternatives to get at the gender dimensions of poverty using the characteristics of the households that go beyond the head of the house the sex of the household head. And third we analyze whether these alternative classifications actually add value to the conventional analysis of poverty based on hatching. For the analysis we use the Somalia high frequency survey the second wave which was collected by in 2017-18 by the World Bank and includes a large sample of internalized place populations. And finally for the empirical strategy what we use are descriptive analysis of these household structures and for the more rigorous analysis we use linear probability modes. Why is studying Somalia? Well it is a very interesting case study both from a development and agenda perspective. From a development perspective what Somalia is is one of the poorest countries in the world. It is estimated that almost three million people have been forced to flee internally because of conflict and natural disasters and so on. And they represent approximately 17% of the population. As mentioned one of the causes of this displacement is the floods and droughts that are recurrent in the country but also an ongoing conflict and the situation of insecurity. Many IDPs settle in informal settlements which are characterized by poor housing conditions, overcrowding, lack of access to basic services, high risk of gender-based violence among others. So from a gender perspective it is also very interesting because Somalia is one of the countries where female labor force participation tends to be low compared to other countries in Saharan Africa and also compared to men in Somalia. But at the same time women tend to be economically active, particularly participating in family farming for instance. And there is also some other evidence suggesting that there has been a change in gender roles precisely because of the disruption in family structures. So women have adopted more male-oriented tasks such as lifestyle trade. And then as mentioned GBB is particularly prevalent and particularly in IDPs settlements and child marriage is also prevalent. So for the empirical approach we classify households using three different alternatives. So the first one is using the traditional headship concept which compares male and female-headed households using and then for poverty we use an income per capita income measure. Then the second alternative is to classify households according to their demographic composition. So here we're using the basically the household roster which identifies the relationship between household members their age and sex. And we come up with five different types of households starting with a male single caregiver which for instance consists of a household with a male adult which is in charge of either a children elderly or disabled and there is no spouse in that household. Then the same idea but with the female single caregiver followed by the traditional neutral household of a principal couple of women and men with children named households with multiple generations so to extend them households to put in another way. And finally families without children which tend to have the lower dependency effects. And the third alternative to look at the gender dimensions of poverty in this paper is to look at income profiles. So we classify households into seven different groups using the number and age and sex of the people who contribute to the household thing. So for instance we have the first family type of ignorance people who don't receive any remittances so basically they depend on humanitarian assistance. Then we have remittance recipients they don't have any people actively working in the labor market but they depend on these cash flows and followed by female single learners so basically it's an adult female adult with only one person contributing to their household income with many others which could be dependents, children elderly and so on. Then the same idea with male single learner then a group of households with an equal number of male and female learners which tends to be similar to the couples with children in demographic composition. And finally two categories which try to capture households that have multiple earners so in one case multiple female learners adults and then in the other case multiple male earners. So just presenting here are some descriptive statistics of these three groups. So the first one if we use the measure the headship measure first we see that's 70% of households in Somalia are poor. However when we look at the headship concept and this is based on self-reported headship in the interview and the survey survey we see that poverty rates tend to be higher among male-headed households than female-headed households regardless of their displacement status. And again we are here we're using a poverty measure based on income per capita so there are assumptions behind that but this is contrary to what we would have expected given the knowledge that we have about the context. When we look at the demographic composition though which is again a measure that captures burden of care dependency ratios and so on and it uses the relationship between gender and nature household members we see that some contrary to the headship concept what we see here is that it is female single caregivers that tend to be at a higher risk of falling into poverty. And even though poverty rates are in general high as I mentioned at the beginning there are some difference here for instance when comparing the two extremes right that the female single caregiver and the male single caregiver and as expected families without children tend to have the lowest poverty rates when comparing using the demographic composition. Then when we move to the income profile which tries to capture access to labor market and economic opportunities and using information on the number and sex of people who contribute to the household's income we see that as expected no earners tend to have the highest poverty rates but they are followed by female single learners again similar to what we are seeing with the demographic composition and at the bottom we have male single learners which even though again tend to have high poverty rates they are statistically lower than those faced by female single learners and other household types such as equal contribution households. So to conduct a more rigorous analysis of this the value added of these categories was to make linear probability models separate estimations for IDPs and IDPs here the dependent variable is because variable basically one in the household is below the poverty line and zero otherwise and we control for different individual and household characteristics within the household characteristics we include the family types using the demographic composition and you have two minutes left. Thank you so what did we find so for the demographic composition we see that it is a female single caregivers we confirm that finding that said tend to have a higher probability or odds of falling into poverty compared to families without children that tend to have the lowest poverty rates this is also the case for couples with children and with a generation here I'm just showing select resources that regressions then when we look at non IDPs we see that only statistically significant difference is the poverty rates between families without children and multiple generations with children. When we go to the income profile what is interesting to highlight here is that it is the fact that IDP households in particular benefit from having multiple earners and particularly the probability of falling into poverty is much lower for majority female earners when compared to no earners for non IDPs it is mostly for majority male earners and here again female single caregivers tend to be quite vulnerable poverty rates are similar to those experienced by no earners so what are the policy implications of our findings first obviously that it is very important to consider family structures and the disruption caused in those structures to analyze poverty and particularly for policy making so even though poverty rates are very high in this context still targeting remains important and I put here an example of the Baxnano program which is a national cash transfer and they take a very serious approach to targeting among those barriers that they use this is an index of distress and also the characteristics of of the households and the program so far has been quite good it's very important to know that had we stopped at headship we would not have known that single caregivers are quite vulnerable to poverty and then for the other policy implications I leave it here but basically considering that the restrictions that women face are quite relevant in the design of economic opportunity programs as well as considering the option of using mobile money to expand and scale up social protection and humanitarian assistance thank you so much for your attention thank you very much