 Thanks to all of you for joining us this morning. I know it's pretty rare for you to have time to spend your morning listening to a seminar, so I very much appreciate you taking the time to join us this morning. As Augustine mentioned, I'm going to present the overarching preliminary findings from this 18-month project we've been doing on foreign aid and democracy focused on sub-Saharan Africa. And then we're going to give some specific case study examples from our research. And the second session will then focus more specifically on the Ghanaian experience. I think it's important for me to start off by saying that this is kind of a global research project. We have a global network of researchers involved. So besides myself, we have Mamadou Gazebo, Carrie Manning, Eji Mabode, who we'll talk later today, and Lisa Rackner, Eileen Tripp, Nick Vandeval, Joseph Wright, and Simone Dietrich. So this has really been kind of a global network of researchers focused on this topic. And really, our motivations were fourfold. First of all, of course, it's kind of a normative motivation for focusing on democracy based out of Amartya Sen's belief that prevention from participation in one's political community constitutes a major human deprivation equivalent to other economic and social deprivations. But as Augustine was alluding to, there's also instrumental reasons to be concerned about democracy. There's a large body of scholarship which has indicated that since in democracies, you do need to get as many votes as possible. Democratic regimes compared to autocratic regimes tend to be more concerned about public rather than private goods. And there is evidence suggesting that democracies spend more on education and health than autocratic regimes. And then thirdly, of course, the topic of democracy and human rights are really important on the policy agenda. We saw this just at the end of last year at the fourth high level forum on aid effectiveness in Busan, South Korea, where at the end of the meeting, the donor community issued a joint statement noting that promoting human rights, democracy, and good governance are an integral part of our development efforts. And then fourthly, it's kind of a research gap in this area. If you look at the research on aid and democracy, it really falls into two categories. So on the one hand, you have a lot of statistical analysis across countries. And some say aid promotes democracy. Some say it hurts democracy. But really, this research, they look at very different time periods with very little theoretical justification. They include countries that really are at very different stages of democratization so they don't really belong in the same country sample. On the other hand, you have qualitative, in-depth case study research. But it can often be cases picked very selectively. And the disadvantage is that you don't get the comparative analysis. You don't know if the results from one single case study are representative more broadly. And then finally, the existing research tends to aggregate all foreign aid together, even though we know that foreign aid has many different objectives, many different actors. It's a very diverse community. And particularly when you're talking about democracy, it's very important to make a distinction between development aid, which we think about as being the resources specifically targeted at supporting greater economic and social well-being, and then thinking of democracy assistance, which is typically resources aimed at promoting those actors and institutions who are concerned about promoting greater political liberalization within their communities. And it's very important to disaggregate these two types of assistance, because these two communities have very different views on the importance of democracy and the means to achieve it. So for example, the development community often sees democracy coming through a long-term structural transformation through the growth of the middle class and its commitment to democratic values. While the democracy promotion community is focused on the short and medium term, promoting those actors and institutions in the here and now. They often have different views towards the state as well. The development community sees the state as a partner. Sometimes the democracy community sees the state as a potential hindrance towards promoting greater liberalization. So with this background in mind, we also thought it's important to disaggregate these two types of assistance, because development aid is really the overwhelming share of assistance that goes to sub-Saharan Africa. And so the development community has a much greater leverage to influence government behaviors vis-a-vis democracy assistance, which, as you can see here, democracy assistance has been increasing steadily. If you're looking at the solid red line, it's been increasing steadily since the early 1990s when we saw much of Africa transitioning to democracies. But if you're looking at the dotted line, you can see that it still comprises a very small share of overall development assistance. It's only about 4% of total overseas development assistance. So our approach in this project was to take into account some of these previous weaknesses in other research. So first of all, we do disaggregate foreign aid into democracy and development assistance. And particularly when we're talking about development assistance, we look at a particular modality that's become increasingly popular called general budget support, where donors are directly supporting the budgets of recipient governments. Now, this is admittedly simplistic. We would like to have much more nuance, because as I said, the aid community is extremely diverse. But this, we think, is a good first cut in terms of getting beyond, seeing foreign aid as just being a monolithic resource. Then secondly, we use mixed methods. So we're trying to draw on the strengths of both the statistical and the qualitative research. So we do this by first having a quantitative cross-country analysis starting in 1991, which is when Benin and Zambia led the way to democratization on the continent. And then we have case studies of seven electoral democracies. We focused on electoral democracies to basically move away from the problem I noted earlier, where you see a lot of research aggregating countries that are at very different stages of democratization. And so it's very hard to isolate the actual role of aid. So when we're looking at electoral democracies, we're talking about countries that have all transitioned to a multi-party regime and have subsequently held at least four rounds of internationally deemed free and fair elections. But even though we're focusing on countries that are relatively similar in being electoral democracies, we still have a great deal of variation amongst them to exploit. So for example, we have what you could consider stable multi-party regimes, which include Zambia, Benin, and Ghana. Then we have two dominant party regimes. These are countries where the ruling government has been in power since the country achieved independence. And those include Tanzania and Mozambique, which Kara will speak to. And then two countries which we could consider fragile democracies, including Mali, which of course, since the coup in March, has really changed dynamics in that country. And Malawi, which since the death of the president in early April, has really created a very fragile unity government in that region. And then finally, when we're talking about democracy, we're looking at both transitions from one party to multi-party regimes and then consolidation of democracy. I think transitions is a very easy obvious term, but consolidation is a much more complex term. So just to elaborate, we're looking at consolidation as being a continuous concept that has both what you can consider negative and positive poles to the ends of the spectrum. On the negative side, we have the avoidance of something. We're trying to avoid at the end of the day breakdown a transition back from a multi-party to a one-party regime. Kind of occupying a middle ground is avoiding erosion. So here, we still have a multi-party regime, but we're seeing a deterioration in civil liberties or human rights or presidents trying to change the Constitution for additional term limits. Those are things that we'd like to avoid. And then on the positive pole, we're trying to deepen democracy. And here, we focus specifically on in three domains. We talk about strengthening vertical accountability. So relationships of responsibility between elected officials and citizens, reinforcing horizontal accountability. So these are relationships of responsibility across government institutions, and then creating competitive party systems. Given that we know in Africa, opposition parties often face a number of constraints. And actually a lot of research shows that incumbents win elections 85% of the time in Africa, which is very different in other parts of the world. So now let me just get to the findings. And I'll just divide these as I've divided the term. So we'll first just briefly look at transitions and then focus the rest of my time on consolidation. So what we found with regards to transitions from this research is that development aid has played a role in some democratic transitions on the continent. From our statistical work, which you can see on the left-hand side, is a figure showing that higher levels of development aid have been associated with a higher probability of transition from a one-party to a multi-party regime. But we've only found this in countries that had a non-military regime at the time of transition. And the reason for this, we believe, is because the cost of transition is much lower for a non-military regime because military regimes tend to have a much smaller coalition. So they tend to have a higher cost to actually transitioning away from a military to a democratic regime. Sorry, this keeps automatically going ahead. From the case studies, we've then tried to disentangle what are some of the mechanisms for this statistical relationship. And to begin with, if you're getting higher levels of development aid, the donor community tends to play a higher role within the country's political economy. And we found that there were two mechanisms driving this relationship towards transitions in the early 1990s. One was through economic conditionalities imposed by donors that then, in turn, precipitated domestic discontent that was associated with greater demands for political and social reforms. So this was, of course, the example in Zambia in the early 1990s, as well as Benin. The second mechanism, though, was straightforward political conditionality, where donors demanded that governments reform and have multi-party elections before they would resume aid. The quintessential case of this is Malawi, where the donors put pressure on President Hastings-Banda. Now, this research largely confirms what has been found looking historically at the role of donors and transitions. But the main criticism of that literature is that, of course, transitions aren't enough. We need to kind of think about consolidating democracy once we've moved to a multi-party regime. So when we switch to looking at consolidation and we first start looking at our negative pole at avoiding breakdown and erosion, what we've found from our research is that democracy aid and development aid actually have very interesting synergies with each other. So democracy aid tends to discourage elites from making certain decisions in the first place. From our statistical work, we found that higher levels of democracy aid was associated with a lower likelihood of multi-party breakdown. And this is particularly true in countries that had a history of party institutionalization. So countries that, even before they allowed multi-party elections for the executive, they still allowed multi-party elections for legislatures. So Senegal, for example, allowing multiple parties, including opposition parties, and the legislature before its major reforms in the 1990s. And the statistical result was also showing that higher levels of democracy aid were associated with a lower likelihood of presidents trying to change term limits. So we actually saw the risk of term limit violations dropping from 40% to less than 5% at higher levels of democracy aid. When we looked at our cases, we were trying to better understand this relationship again. And one of the key mechanisms is actually promoting civil society, which is very vigilant, for example, about presidents changing term limits. So we had the example of Zambia in the Oasis Forum in 2001. This was a forum of over 100 NGOs that were very active in protesting against Frederick Chaluba's bid to change the Constitution for a third term. It was heavily supported by the donor community. And at the end of the day, Chaluba decided not to put forth the bill in Parliament to change the Constitution. Now we find, when we turn to development aid, we find that development aid could be used in the same way to prevent breakdown, particularly if it was used to address structural inequalities that can be key sources of cleavage that cause a society to implode. And the example we had here was Mali, where our case study author for that case heavily felt that if the donors had actually addressed a number of concerns that were cropping up in Mali over the last 20 years regarding inequalities between the North and South, torii grievances, then potentially we might have avoided the coup in Mali. But of course, without the counterfactual, we can't say that for certain. What we can say with more certainty is that development aid has often been used as an ex-post tool to reverse elite's decisions. And this has been mainly through using leverage through the donor community, threatening to rescind or suspend some of its aid unless elites reverse some of their decisions. So some examples we have, which Kari might talk about is the group of 19 donors, the budget support donors in Mozambique going on strike. And then also in Malawi, which I'll talk about later today, threatening to withhold aid if but Kuli Malusi went forward with his bid for a third term. But what we're finding overall is that it's corruption and electoral violations that are the main issues that spur the donor community to react. Violations of civil liberties or human rights are addressed by donors, but it's much more erratically. We're seeing certain donors, like the Nordic donors, might be more concerned with these issues, vis-a-vis some of the multilaterals, like the African Development Bank or the World Bank. And so the issue of coordination is really key for effectiveness in that domain. So if we turn now to the positive poll of consolidation, trying to deepen democracy, if we're looking at vertical accountability, elections have really been the main mechanism of accountability between citizens and politicians. And we really have found that from the statistical work, democracy aid was significantly associated with promoting free and fair elections. And this was really key from our case studies as well, that many of the researchers felt that without some type of electoral support, elections would have been less free and fair in many African countries. But there's a number of key concerns. Elections still tend to be seen as an event rather than a process, so lack of capacity building for the electoral commission in between elections. Some aspects of elections are deemed too costly for the country to actually afford and believe there's been some debate in Ghana over the biometric system. And then most problematic has been the lack of electoral commission autonomy. At the end of the day, donors don't have too much influence in making sure that there's an impartial leader at the head of the electoral commission. I think in the interest of time, I'm just going to skip over the civil society component, but suffice it to say that we're also seeing civil society as an important mechanism of vertical accountability. It's the area of democracy assistance that receives the most attention from donors. And we think that's because a little bit of resources actually goes a long way with civil society. And we've definitely found that without democracy aid, many NGOs, for example, would have lower capacity or fundraising abilities than in the absence of this assistance. But at the end of the day, something that the democracy aid community really can't affect too much is the legislative environment in which civil society operates. So across basically all of our cases, we found that there's been an increasingly repressive media environment, whether it's explicit in terms of new laws being introduced or whether it's self-censorship amongst journalists. And this is something, this type of environment of freedom of speech was something that's not finding that the donors can influence too directly. I'm gonna quickly move on to another component of our positive poll of consolidation, which is horizontal accountability. And again, here we found that democracy aid has been really pivotal for supporting institutions of horizontal accountability. So here we're talking about judiciaries, parliaments, anti-corruption bureaus. And there's really been some notable successes. Perhaps Jima will talk about the public accounts committee in Ghana, but we've also found in Zambia, the anti-corruption committee has been very much strengthened and was actually a whistleblower in key corruption cases there. But a consistent finding across all of our cases is that this modality of general budget support tends to further marginalize the role of parliament. And this is particularly because budget supports typically negotiated between executives, ministries of finance and the group of budget support donors. And parliament is informed of these decisions when it receives the budget to approve. But often we found that parliaments are not aware of what budget support donors are doing and feel very much sidelined from the entire process. And then finally, if we look at competitive party systems, party assistance is the area of democracy support that receives the least attention. And we can even see that back here from this graph here. We're seeing that legislative and party strengthening is really small compared to support going to civil society. And this is a bit unfortunate because it's extremely valuable for opposition parties. A number of our cases pointed to the fact that opposition parties wouldn't have just basic resources for campaigning, offices, access to a telephone without this type of support. And also, as you're seeing from our statistical work, the figure on the left shows that a higher level of democracy aid has been associated with a higher probability of party turnover. So a higher probability that the incumbent party is ousted by an opposition party. By contrast, we found that development aid actually has the opposite effect, that it tends to reinforce incumbent advantages, particularly in dominant party regimes, which I think Carrie will talk a bit more about. And this is because incumbents can basically use this assistance to kind of justify that they've been doing something good for the broader populace. And so the kind of lines of accountability between what the donors are doing or what the government is doing become very much blurred. So for example, in Malawi, we saw agricultural subsidies being distributed right before elections. In Benin, there was an example of malaria medicines released right before elections. So ways in which incumbents have used development aid to reinforce their own advantages. So just to summarize the key findings, at the end of the day, as we all know, democratic consolidation is ultimately an internal affair, a domestic affair, but external actors in the form of the donor community do play a very important role, whether intentionally or unintentionally. And I think it's important that we recognize it because increasingly I've seen development economists saying that donors should not get involved in governance. They should not get involved in internal affairs. But our research is showing that even if they don't want to get involved in internal affairs, they do actually have often a direct, but many times an indirect impact on political trajectories. I think the research has also shown that democracy and development aid have different levels of influence over the democratization process. So development aid has had a much clearer effect on transitions while democracy tends to have a more direct impact on consolidation. And I think given Africa's changing development landscape, just considering Ghana, graduating to middle income status and findings of oil or same in Zambia becoming a middle income country, I think this development landscape is sure to change. And so we could think that democracy still remains pivotal given the importance of maintaining some type of mechanisms of accountability between government and its citizens. And I think lastly, the main point that's coming out of the research is that there's clear trade-offs between the democracy and development community. Whether it's the democracy community focusing on strengthening parliaments while budget support tends to sideline them. Whether it's democracy helping opposition parties while development aids indirectly reinforcing incumbent advantages. And so I think really for the policy community, it's thinking a bit more about how to reconcile these trade-offs to make sure that these two types of assistance are moving towards the same objectives.