 Thank you very much. Good morning. The paper I am presenting has been funded by the expert group of A studies in Sweden and also in collaboration with wider, so I'm very pleased to be here to present the preliminary results. Yeah, so just just to give you an example of how the social protection system today has been expanding over the last decades. Well, as you can see from the graph in the early 2000s where we started to collect data, reflect a substantial increase in not only in the number of programs, but also by type of different programs they have expanded substantially, which also reflect different choices by countries and dependent on different factors. So one of the interesting things here is that specific types of programs have been dominant in the expansion of social protection in the global south. And here when I'm talking about social protection, I refer in particular to social assistance where we are focusing on this because among welfare institutions we find that social assistance has been much more dynamic than other types of program-like social insurance or labor market policies. So therefore the discussion of today will be focusing on social assistance in particular. However, if you look at the distribution of coverage across regions, you can see a substantial difference and in equality in terms of coverage, if you look at this table, the colors will give you a hint of the distribution of coverage moving from dark blue, dark green to dark red with dark green showing a higher share of the population covered by specific types of programs and by different types of vulnerable populations. And these dynamics and also these distributional characteristics of the expansion also again reflects a lot of policy choices. So the Peppers contribution is trying to understand the contribution of foreign aid in this expansion that I showed you. So we want to understand the extent to which in particular foreign aid has contributed to the development of these systems in low-income countries and also we want to understand the conditions that explain those dynamics and also the actors and what kind of other factors have been supporting or hindering this expansion. So in order to do that, we do quantitative analysis and also we look at aid flows in detail and also try to understand what is behind all those patterns that we observe in the data. And the reason why we try to do this is because, well at the moment, the literature is very, very scant. Most of the studies focus on specific cases. There are very few studies using econometric analysis trying to somehow identify the relationship between expansion of the systems and the role of external actors. There are also interesting studies using qualitative methodologies, but again, there are different challenges from both type of studies from the quantitative side. Many of the studies don't go as far as trying at least to establish some kind of causal relationship, whereas in qualitative studies also they face obviously a trade-off between identifying the process of tracing the channels and the mechanisms and also looking at different cases. Also, we are trying to contribute to this scant literature. So overall, what we can say is the literature highlights certain patterns that highlight the way donors have contributed to the expansion of certain types of programs, in particular the World Bank and other agencies have been playing a role. Again, some of these evidences more anecdotical or based on certain methodologies that we are not sure how it would extend to which we can claim some kind of generalizations. So some of these studies also highlight a number of determinants and we exploit the information that the literature highlights. It's also to control for specific dimensions that are reported to affect it or have been influencing the expansion of social protection in the global south. And these are related to historical legacies, for example, path dependency, the role of institutions, in particular democracy is highlighted, the role of demographic dynamics, for example, in the southern part of Africa, HIV, whereas in other latitudes, there are other factors that also relate to the demographics that explain policy choices. There are a lot of ideas and obviously coverage shocks, you know? Latin America, for example, financial crisis were determined in making certain decisions back in the late 1990s. And also those kind of factors are controlled in our specifications. Right, so one of the things that we obviously came when we were doing the analysis, how we measure aid to social protection. So we have two definitions, overall one we call the narrow definition that contains these type of activities covered by foreign aid and then we use a broader definition that also covers activities that relate to, for example, labor markets, policies, for example, social dialogue, which can be according to some hypothesis relevant. So we include two definitions for measuring social protection aid and overall the historical trends based on the existing evidence or the existing data shows this kind of pattern. As you can see, the aid flows towards the support of social protection through the activities that I just showed you have overall captured about 2% of foreign aid historically, which is a very small percentage of foreign aid. What is interesting here is to see how different types of donors, either multilateral or bilateral, show different patterns and also different levels of contribution to this type of flows. So we also explore this information in the analysis to show how different types of donors have been actively involved in the provision of aid. And you can see, for example, on this on this graph that global aid, which global means total aid, multilaterals, bilateral plus DAC countries and non-DAC countries aid, are dominant by multilaterals, so about two-thirds of foreign aid is channel via multilaterals to this sector, which is quite distinct in relation to other sectors of activity. If you look at total development aid, bilateral are much more dominant, whereas in social protection, multilaterals have been taking a much active role. And also you can see these spikes one in the late 1990s, another one around 2008, which also shows how multilateral aid responds to shocks. The last big spike reflects the financial crisis of 2008, which also shows how these systems can act as contraceptive measures in types of financial crisis. So if you look at the type of instruments or type of finance, you can also see substantial differences. So bilateral rely heavily on grants, whereas multilaterals rely on depth instruments, which also can work through different channels, can be conditionalities and also through a specific type of conditions that shape these kind of programs. And all this information is important for the econometrics. That's why I'm giving you this overview of data. And also one other thing that's very interesting is that Latin America has quite dominated aid flows over the last decades. And South Saharan Africa in particular has been increasing just recently after 2010. And this also reflects the new priorities of donor countries in terms of aid allocations more recently. So I was going to spend much time on the methodology, all the details of the methodology in the paper, but I just want to say that what we do is try to capture the very complex dynamics of these relationships. And we implement a number of approaches to try to at least address the issue of endogeneity of aid in this context. We are trying to capture the expansion by measuring coverage of the population in each country. And we use certain models that reflect, as you remember, the distribution of coverage. In the early days, usually trend is close to zero. And therefore the distribution of coverage is a sense sort towards zero. That's why we use these particular econometric approaches. We also use some additional approaches, the fractional response models, which in a way also capture better the fact that programs or the coverage is related to the size of the population. And also this is our preferred model for the estimates. But nevertheless, one of the things that I just want to briefly mention is the instruments that we use, we rely on some set of instruments that we have used in the past. And I can discuss in more detail the rationale behind those instruments, but overall try to capture some exogenous reverberation, in particular in donor countries, and also try to capture policy choices according to the constitution or the way parliaments and governments are instructed or organized in donor countries. And I can go and talk about this in more detail, but I will skip this. I want to just go to the results. And one of the interesting things that we find is across the board, we implemented several models trying to control for multiple factors to reduce the problem of omitted variable bias. And also after controlling for the endogeneity of age, we conducted some tests and, well, always instrumental variables are controversial. So whether you believe in instruments or not, but nevertheless, even assuming at the level of correlation, we find a very strong association in the contribution of foreign aid to the expansion of social protection across low and middle income countries. We look at different types of donors. But overall, the effect is about these are elasticities based on the fractional response models. So in a way, one percent increase in foreign aid leads to an increase about 0.25 percent of points of coverage, so which in many contexts is non-eligible. So also we find some interesting variation across work regions. So what we find is that in particular in South Saharan Africa, the effects are driven by multilaterals, as you can see in that top graph, which essentially shows points and estimates. And once you cross the line, you see significant effects at 90 percent levels. So one of the important things is that what we observe in the data is somehow reflecting also in the graphs that multilaterals in particular South Saharan Africa have been the main actors driving these effects. And obviously, when we look at other type of donors, there are a few issues. The confidence intervals are too big, so we have certainty about the accuracy of the effects. But nevertheless, we can see that these actors have been playing a significant role in the expansion of these institutions in different parts of the world. So to give you a sense of how we can interpret these point estimates, so let's say we take the case of Ethiopia. We can think that one percentage increase in foreign aid will, going from an average of $253 million, which is the average that Ethiopia received in the last five years, will have an effect in the coverage of about 260,000 people covered by this kind of season from a baseline of about 7.8 million people who are covered by these kind of programs in that country. So these are not knowledgeable effects and actually show the way aid has been contributed to the coverage of vulnerable populations. But also what we observe is that there is a quite significant inequality, let's say, in terms of the way donors allocate resources regardless of the level of poverty or the level of vulnerability among countries within those regions. So certain countries receive far less, for example, assistance regardless of whether they have lower or higher income per capita. And this also reflects policy choices in the North country and also reflect political economy factors that we also explore in the models. So we, as I said, we explore and test different theoretical predictions about the factors that influence those systems and we cover overall dimensions that are grouped into these kind of areas. And overall, we find interesting information that, again, it's in the paper. I just give you here a very quick overview of what we find. But overall, in just going to the conclusions, we find obviously that the data suggests that there is a significant effect of aid, particularly in these kind of sectors. And we don't find any evidence of detrimental impacts of aid. There may be situations in which, for example, competing policy choices may influence the way these policies or these programs or systems have been taken shape. But overall, we find some positive evidence about these relationships. The fact that 66% of these flows to South Africa in particular have been channeled via multilaterals and also through the instruments, which is important, in a way underscores the way governments have been taking play or getting involved in the expansion of these systems vis-à-vis what we observed in the early years, which I think is a positive signal. And also, we find that also bilaterals are heavily on certain types of modalities that although reflect the complexities of working in this area, because gradually countries have been moving to fragile states and it's more difficult to work with autocratic regimes, also there's the issue of state capacity and also state building in terms of helping mobilize resources. So there are obviously certain aspects that have been influencing the expansion of these programs, but also we find that there is no sufficient connection between the contribution of aid and also the extent to which tax collection systems have been evolving to finance those resources that are needed to continue sustaining these programs. So this is one of the areas where we think that based on the data, there is a need to increase aid allocations to improve tax collection systems in those countries. So, well, thank you very much.