 All right, good morning. Last conference day, last parallels, last presentation. So thank you for still being present. I'll try to be extra entertaining. So my presentation is about social protection floor gaps and pandemic relief, a case for universalism. I decided to not go very much into the details of the paper, but more present sort of the debate that it links to more broadly. Talk a little bit about the findings and the data that I've been working with. So I will start at the reverse order. What did I find? That universal policies are more prominent when social protection gaps exist. With social protection floor gaps, I mean a specific measure that looks at the implementation of social protection floors in line with the ILO social protection floor recommendation 202 that has been ratified in 2012 by I think all of the 187 member states. And where there are sort of financing gaps in terms of achieving those social protection floors, you could see more universal approaches in the way that countries responded to the crisis. There are then of course also variations across high income countries and the rest of the world. What I mean there is like any country that is classified as non-high income. Overall, globally, using the database that I've been working with, you can see that the targeted responses are more dominant. I think for anyone who worked in social protection, that's not surprising. But the share of universal policies in crisis relief in lower is lower in high income countries, right? These countries are also those that have lower financing gaps. So the debate that it links to, I think is like something that Yucca also allotted to in a way, is the targeted versus universal debate. I think it has been around for a couple of decades. And of course there are like benefits to both, but also costs, I would say. I'm trying to summarize a few of those here. So when it comes to targeted, like of course we have an element of selectivity in that we kind of try and isolate the groups or like identify the groups that are most in need. It depends on sort of like the political foundation that is underpinning the way countries formulate their programs. It is also the dominant method, as I mentioned. It was also the dominant method in crisis relief. And it's been argued also that it's fiscally more feasible, right? If you think about, okay, we have a part of public funds. We want to give it to those most in need so it can be a bit more efficient than having broad-based welfare. That's also the counter argument towards more universal methods that are in a broad sense available to all. As Yucca mentioned, the most let's say radical example of that is the universal basic income or the universal provision of basic services. It can go from either direction. It seems to be a bit more contested and it hasn't really been implemented. It has been piloted in different spaces with different results. As shortcomings of the targeted, I think everyone is familiar with that in terms of inclusion and exclusion arrows. Also sometimes in terms of political support. I don't know in how far you're familiar with the literature that's more like from the element, like the perspective of social psychology and economics where sometimes, I mean in terms of from a voter perspective, you're more likely to support programs where you stand to benefit yourself. In terms of targeting this, also social costs and certain stick-mounted, colleagues of mine, former colleagues of mine at IDS have been working on. For the universal one, you could argue you have lower targeting costs, but I would say from what I've seen in recent studies, the way that it is assessed whether they are politically supported is a more broader understanding of fairness and how inequalities are justified or reasoned about in a country's context. And there are also first studies that look at whether it can be a tool for creating greater social cohesion. However, there are still a bit of mixed results, I would say, very interesting if you're looking at that in the post-crisis area as well. So why does this debate still matter or matters in terms of looking at crisis relief as has been mentioned many times, I think, and particularly in this panel, there was a very rapid expansion of social protection as a consequence of the pandemic. And there's also been that debate sort of that sparked in sort of looking at social protection sort of at the critical moment or seeing it a little bit at a crossroad, I would say. Certainly a moment for institutional learning to revisit some of the infrastructure that is in place, where are the gaps and what are the best ways forwards. I'm also talking about debates on creating adaptive systems. I think it's very much what the first presentation was trying to show also in terms of you have shifts in beneficiaries during the crisis, but also shifting needs at different times and often at a very ad hoc or unpredictable manner. So how can you adequately reflect that in the system to scale up or expand it accordingly? So I think the broader question that I would like to bring to the audience rather than giving a definite answer, I think it's an ongoing dialogue, is whether we should kind of walk towards ever more fine-tuned systems, learning for example from climate change, having early warning indicators and so forth, or whether we should go rather like in a different direction of saying possibly we'll never get it exactly right that we are able in these kind of very rapidly unfolding events that we always find those that are most in need and what their needs are, should we just have a broad-based kind of landscape of social protection provision in place. So a little bit about the data that I've been working with and there are certainly different databases that you can use, housed by different institutions. So the ones that I used was the COVID Stimulus Tracker which has been published by, I'm just gonna pronounce it as a written UNESCO, which is an acronym. So as you can see on the left side, I think it's a nice illustration that they capture a lot of countries and that you can see that it was truly a global response because all that blue areas had more than 10 policy measures. The green areas less than 10 and the red areas so less than five but nevertheless some policy measures, right? And the other, the pie chart that I'm including is just to highlight that it has also information on the specific beneficiaries. So this is on a bit more high-level basis. There's also variables that have it in more detail as I will show you here. So in order to classify all these policies that are captured in this database into universal versus targeted responses is a little bit of an arbitrary exercise I would say. In part also because there is a lot of policy innovation and gonna come back to this point a little bit later on. So I don't think this table there is very readable for you but I nevertheless included to show you have the first column which are policy measures that classify as universal in line with the definition that they reach every citizen based on a basic criterion. I think Yooka was saying that it's like sort of broadly categorical like universal pensions for example that go to any citizens of a certain age group, right? And then you have the column after that one which are the targeted measures and there you have like very some of the targeting that we have heard like proxy means tested and people with disabilities as was also in Miguel's presentation but then there were also indigenous people there were people who were formerly in prison or are in prison so it also perhaps a new way of setting up beneficiaries in this space of crisis relief, right? And there's a question of course of what we take forward from this. So overall also you can see that targeted dominates as I have mentioned before. So this data I bring together with the social protection floor index so unless unlike that Miguel presented which was more about coverage this can be more understood as a financing gap for anyone who has, who's not familiar with the recommendation it basically sets out three basic income guarantees for children, people on active age and older persons and access to universal health like basic health coverage, right? So what I've been part of kind of designing this index I think almost like eight years ago now, Jesus. And what we do is like we look at basically an income gap and a health gap so I'm not going too much into the technical details but if you would like to hear more about it please ask any questions. The important thing is to understand what this measure basically expresses is like how much of a country's GDP would be needed in order to establish social protection floors in the country. With the recent addition they also published it as a percentage of total government revenue which I think will be very interesting for the audience of this conference. I also use for the income gaps, I use a relative poverty line obviously to make it applicable across the global north and the global south rather than absolute poverty lines, right? So to come back to the findings that I showed you in the beginning so what as I said before so if a country has the financing increases across countries you have more universal responses in the crisis relief. If you think about it it makes a lot of sense because say your system is not as expansive in terms of coverage and you need to quickly respond to people's needs it's easier to kind of implement a broader category and perhaps work rather have inclusion errors than exclusion errors during these times. What I also wanted and I think that's something that can be explored going forward whether this is an ad hoc sort of once off design shift that was found in particularly in non-high income countries or whether there is a sort of institutional momentum to go forward in a different direction in terms of like rethinking the particularities of targeting methods. As I highlighted before as well there's a lot of new beneficiary categories and like Miguel I also worked with two different definitions of social protection so one that was perhaps a little bit more traditional I'd say and one that was expansive in that it also included programs that were targeted at SMEs you know so part of social protection active labor market programs that was targeted at employment we can make an argument of course like programs that are geared towards businesses that are also equally relevant right. So there's kind of I would say there's even like Meredith looking at sort of new and old beneficiaries and that crisis response and perhaps rethink whether in which ways we could shift the conceptual boundaries of social protection and then of course in financing those crisis responses I think there's a big question of what to keep which is a very country specific question and not something I can answer on this very global level analysis but what I would like to sort of conclude with is still a question right because I've done some first level high level explorations or you can see also not even in the like sort of econometric estimations it's an association it's not a causal explanation that you can have but nevertheless I would say as far as the results goes like the existing system of course matters for the design of crisis response I think for various reasons if we think about vertical and horizontal expansions oh great I'll be quicker than that. There's also that question and that's something that I'm working on with colleagues to bring more politics into the space of social protection because of course there's a lot of merit in exploring like sort of the technicalities of programs but there's also a way of looking what is the social fabrics of political support in the countries who stands to benefit which is also relevant for like the collection side what in terms of fairness and taxation and lastly I mean of course in terms of fiscal feasibility I'd be very interesting to hear because I'm not really working on the collection side as much it's like okay what is considered more fiscal feasible knowing that targeting is a costly exercise and if we go forward with like these adaptive systems that have say more elaborate early warning indicators and like trying to make targeting more fine tuned is that more feasible or is it more having like a broad based measure in place perhaps a like narrower version than a basic income grant also the other differences sort of that we could look at in terms of costs for horizontal versus vertical expansions okay so I think I will leave it at that so we have some more time to discuss yeah thank you very much