 Good afternoon now, I guess. So I'm going to be speaking today about AIDS impact on democracy, and especially drawing on joint work with Miguel Niño-Sarasua and Anna Horigoshi and Melissa Samaran. So this work was supported originally by the Swedish expert group for AIDS studies, and we produced a report at the end of 2020, and then subsequently several working papers. We continue to work on this. So hopefully we have one or two more papers and some articles coming over the coming year. So the study begins with sort of two motivations. And the first is that democracy aid is a core component of development cooperation. This has to do with arguably support for fundamental freedoms, for the role of democracy in development, as well as strategic foreign policy considerations. If you look at OECD DAC countries, we find that roughly 10, 11% of their overseas development assistance has gone to this area in recent years. And for Sweden, which was a country of particular focus for us, it was about a third. So then of course in this context, understanding and demonstrating if democracy aid works and how and the context in which it works best is important. And it's important to understand that I think Azanka was discussing that the evidence from prior studies is really quite mixed. And in fact, a key criticism of foreign aid more generally is that it's actually bad for democracy. So if we think about Bill Easterly's work on how development experts have empowered dictators or about the body of research on the impact of aid on local accountability, weakening local accountability, governance processes and state institutions, we would also be concerned generally about the impact of aid, even democracy aid on democracy. So I'm thinking here about work, for instance, by Deborah Brodegem and Stephen Knack, by Todd Moss, by Tandika Makandawiri and so on. So it might seem a trivial question, is democracy aid actually supporting democracy? But in fact, it's not a trivial question in the context of the previous literature. A second motivation, so this question is given new urgency by the fact that global democracy is in decline. And this has been shown in a number of studies by the Variety of Democracy Institute, Freedom House, the Economist Intelligence Unit. And the figure here is from the Varieties of Democracy Institute, it shows data from their Electoral Democracy Index, averages in the Electoral Democracy Index over time for different world regions from about 1800 until 2018. And we can see clearly there've been improvements in democracy over time, but there's a concerning trend toward democratic decline from about 2000, more precisely from about 2005. This has implications, of course, for the people in backsliding countries for civil rights and political liberties. It also has more broader, broader, it raises broader concerns for the international liberal order and arguably then for economic prosperity and peace around the world. I think the latest report from the VEDEM Institute suggests that 68% of the world's people now live in autocracies. So this is quite concerning and it raises the obvious question, what can the international community do? Is there a role here for democracy assistance? So does democracy assistance work? To cut to our conclusions, our simple answer is yes. Our analysis points to a modest but positive impact of democracy aid on democracy. The impact of democracy assistance is clearer than the impact of developmental aid, but we in fact find no evidence of an overall negative effect on democracy for either. Aid targeted to support core democratic institutions, civil society, free and fair elections, media freedom and human rights, seems to work best in terms of impact on democracy. How do we get here? So we have a sort of three-part study. We start by developing an analytical framework. Interestingly enough, there's been a huge amount of work or a large amount of work on aid and on aid effectiveness, but it hasn't been very well-married in our view with the literature on democracy and democratization. And so we try to do that in this part of the study. And here's the simple analytical framework summarized here. Just a couple of points to highlight. Broadly, we're trying to address the question in the framework, how should democracy assistance in theory support democracy? And what would we expect to see in quantitative cross-country studies of this sort that we're focusing on here? From this framework, focus first on the columns. So these are the sort of three broad camps of theory of democratization that we distinguish. Many scholars distinguish, summarize the literature in a similar way. We start with structuralist theories of democratization. These highlight how macro-structural factors, like development, like modernization, influence processes of democratization. Secondly, we can distinguish institutional theories or institutional explanations of democratization that highlight how formal and informal institutions, for instance, political parties, for instance, Stefan Lindbergh's theory of democratization through elections, how institutions influence democratization processes. And then finally, a third camp of theories highlights the role of agency, of individuals, in particular of leaders, in democratic transition, survival, and consolidation. Of course, when we think about particular cases, there's always some elements of each in the story. But I think it's useful analytically to distinguish among these three camps. Then if you look at the far left of the diagram, we distinguish building on work by Thomas Carruthers between two broad types of aid. So first, there is democracy aid or what Carruthers calls political democracy aid, which is targeted at pro-democracy individuals and leaders. And it builds implicitly, sometimes explicitly, on institutional or agency-based explanations for democratization. And then secondly, we can think about developmental aid, which pursues long-term incremental change in a wide range of political and economic sectors, and it's much more consistent with these structural theories or explanations of democratization. So the second key part of the study was a systematic review of the literature, of the quantitative literature on aid and democracy. And systematic reviews follow a clear, transparent, and hopefully reproducible method of identifying and then summarizing the literature. And to our knowledge, this was the first sort of systematic review of this literature on aid and democracy. We identified 91 studies looking at English as well as Spanish and French sources that are relevant to summarize here. The table here provides an overview of the findings on the effects of aid on democracy, by aid type and modality. There's more information than we can process in this table. But just to focus your attention, very simply on the two bolded rows, so on developmental aid and democracy aid. And we can see clearly here, looking just simply at directional effects, that the literature on democracy aid is a bit clearer in terms of, as compared to the literature on developmental aid, in terms of having a positive aid, having a positive effect on democracy. So 26 compared to nine studies, having a positive effect when we think about the literature on democracy aid, 39 compared to 30 studies when we look at the literature on developmental aid. And we can cut the literature in different ways. So we also, for instance, looked at studies that we thought were sort of better done or looking at more recent data, and we still find broadly the same sort of finding. There's obviously a lot of contradictory evidence in this body of work, and there's a lot more that needs to be teased out, but this is a simple way to summarize the literature. So the third part of the work then is new quantitative analysis for 148 countries for the period 1995 to 2018. And this part of the work I should notice, especially led by Miguel Niño of Sarasua and Anna Horigoshi. For the more technical among you, if you have questions, I can do my best to answer, but I can also get answers for you from them. We measure aid similar to using the same data that Anka was talking about. We use the OACD-CRS data. We distinguish between democracy aid using both limited and extensive definitions and developmental aid. We measure democracy in several different ways, but primarily we use the varieties of democracy, electoral democracy index, and we also use some sub-indices of that index as well as the civil liberties index. We include a variety of controls and we use multiple advanced econometric methods and checks building on some of the issues that we identified from previous analyses. In particular, we pay some attention to the endogeneity problem. So the issue that aid allocations aren't exogenous to the level of democracy in aid recipient countries. So it might be that countries that are democratizing receive more aid and this will then obviously cause problems for our analysis of the relationship between aid and democracy. So we focus on three questions in the quantity of analysis. I'm gonna just summarize two. The first is, does international democracy aid contribute to democracy? How about developmental aid? Again, the results here suggest that the contribution of international democracy as well as developmental aid to democracy is small but positive and statistically significant. So the rope ladder plot on the right shows markers for point estimates and the spikes show confidence intervals at 95%, sorry, 90% levels. And we can see that the spikes crossing the reference line at zero show coefficients that are statistically different from zero. Focus on the top part of the diagram for now for simplicity here. So very simply, the results suggest in the preferred model that a 10% increase in democracy aid using the limited definition over a five-year period is associated with an average 0.14% increase in the electoral democracy index. This isn't huge but it's positive and it's worth keeping that positive point in mind. It also holds over multiple, multiple specifications. I can show you the packet of results from the analysis. We then look at democracy aids impact on lower level indices of democracy. Again, a similar sort of chart here and again, the impact of democracy aid seems to be positive with the possible exception of aid for elections where we can't really distinguish the impact from zero. So second, does democracy aid better support democratization sort of upturns in democracy scores or does it help to avoid democratic backsliding downturns? So the chart on the left shows results for upturning countries, on the right for downturning countries and the simple result here is that results suggest that democracy aid better supports upturns than downturns. So we can unpack this a bit more and look at some of the countries where we've seen the greatest declines in the electoral democracy index since 1995. For instance, Argentina, Nicaragua, Russia, Turkey, Venezuela. And broadly speaking, there seem to be a couple of differences to think about. There are differences in regime type between the upturning countries and the downturning countries. Upturn countries tend to be mainly electoral democracies and some electoral autocracies. Downturning countries tend to be more likely to be closed in electoral autocracies. And we see also a pronounced decline in democracy in middle income countries in Latin America, Europe and Central Asia. And these are, of course, countries that are increasingly excluded from international systems because of their middle income status. I'm gonna skip that and go to concluding thoughts. So very obviously, there are very good reasons to be skeptical about democracy promotion and the impact of aid on governance. I myself have done some research on this. But it's also, I think, important to be wary of knee-jerk aid pessimism to just say aid doesn't ever work. And I think it's very clear both in other studies as well as in our own that there does seem to be, that there's evidence of democracy aid having some modest but positive impact overall on democracy. So in other words, democracy assistance can help to address this global decline in democracy if it's used in the right way. This seems to be clearest for aid that targets core democratic institutions, so civil society, free media, human rights and maybe elections. It also is clear that in our analysis, at least the developmental aid doesn't have a negative impact on governance and democracy overall. But its positive impact is less clear and this really shouldn't be surprising when we think about how democratization works. We would expect developmental aid to take longer and because of this sort of long delay to make it more difficult to detect the impact of aid on democracy through the period. And finally, a pressing question remains how to respond to backsliding countries where aid seems to be the least effective. Thank you.