 relationship between foreign aid and democracy in sub-Saharan Africa. And so the project is basically designed with a global group of researchers. We had about 10 researchers from Europe, Africa, and North America working on the research. And these are typically scholars who have both a background on the countries that they worked on as well as on the issues of foreign aid. And the main aim of our research was to understand, well, generally how does aid impact democratic trajectories in sub-Saharan Africa? And then more specifically, in what way does aid, you know, in the absence of aid would we see worse democratic outcomes? So what aspects of aid are absolutely fundamental to some of the trajectories that we've seen? So on the other hand, what aspects of aid, including modalities of implementation, actually exacerbate certain weaknesses that we see in sub-Saharan Africa. So our reason for focusing on Africa was twofold. One reason is because even though Africa is a really dynamic region, it's attracting a lot of foreign direct investment, we're seeing a growth of a middle class, it's still the most aid-dependent region of the world. So it still receives four times as much foreign aid as a share of gross national income as the next most dependent region, which is the Middle East and North Africa. And then the second reason is because a lot of democracies in Africa are still very fragile. So if we're talking about electoral democracies in particular, we saw a number of countries in the 1990s in sub-Saharan Africa transition from one party to multi-party regimes and become what are considered electoral democracies. So these are countries that have regular, generally free and fair elections that the opposition has a relatively equal playing field. And unfortunately, since that transition in the 1990s, we've seen 13 countries actually backslide and go back to one party regimes. So we only have, you know, about 10 or 12 countries in Africa that are considered electoral democracies, at least by Freedom House. Some other weaknesses that we see throughout the region are incumbent dominance. So institutions like legislatures and judiciaries don't play a large check, don't have a large check on what the activities of the executive branch. We also see entrenched incumbent advantage. So in Africa, incumbents win reelection by 85 percent of the time. This is much higher than in other developing regions. So this means that opposition parties have a really difficult time getting onto the national political arena. So we're seeing this combination of both high-aid dependence and still very fragile democratization. And so we wanted to understand what is in a relationship between the two. The interest in democracy came from the fact, from a first and normative perspective, that we think that following Amartya Sen's view that exclusion from one's political sphere and from political participation is equivalent to a human deprivation in the economic realm. So being able to participate in the political sphere is very important from a normative perspective. Then from an instrumental perspective, democracy might actually be a means to greater development ends. So there's a large body of scholarship that shows that democracies are at least more interested in providing goods and services for their citizens than autocracies. So they at least spend more on things such as education, health, and social welfare, even if the outcomes are not as good as maybe in some autocratic regimes. Then thirdly, from a policy perspective, this is really an important issue on the donor agenda. So we're seeing a lot of donors now say that along with everything else, promoting human rights and democracy and good governance are really an important part of their efforts and most of these bilateral agencies have democracy and governance units. And so we wanted to know in what way have they actually done this in their efforts and what types of trade-offs or contradictions might have existed between their democracy and governance work and other type of development interventions that they have. And then fourthly, from more of a research perspective, many of the existing works on aid and democracy have had a lot of mixed findings. This is typically because a lot of the work is cross-country econometrics, so it provides a broad perspective and generalizable findings. But the problem is they often use very different time periods that are often data-driven and often a theoretical. So there's no real reason for why you look at 1971 to 1990 or 1980 to 2000. And they often incorporate very disparate countries into their country samples, so countries that are at very different levels of democratization. And so we wouldn't expect aid to have the same effect across these countries given that they are experiencing different trajectories. And they also tend to aggregate all foreign aid together. So increasingly there have been some studies just focusing on democracy assistance and the impact on democracy, but a lot of them have tended to just look at foreign aid overall. When we know that the foreign aid community is very diverse, that practitioners within the community have very different objectives and modalities for delivering aid. So in terms of being useful for policy makers, aggregating all foreign aid together is not particularly useful. And finally, when they're looking at the impact on democracy, they never really clear on what they mean by that. So some studies do say they're looking at the transition from autocratic to democratic regimes, but a lot of them are just looking at the impact on aggregate outcome indicators like a freedom house indicator or a polity for index on political regimes. So democracy is not very well conceptualized and operationalized. And then on the other hand, away from the cross-country studies you have really interesting in-depth case studies, but they're sometimes very selectively chosen. And so people will talk about Ethiopia and then the negative impacts aid has had on Ethiopia, which gives very good kind of causal process tracing mechanisms. But the downside is that we can't really draw general conclusions given some of the unique factors of the Ethiopian regime, for example. So this project was very much aimed at trying to reconcile some of those existing research gaps. And we did that by using mixed methods, so combining the strengths of these two approaches. So we had first a cross-country analysis across all Sub-Saharan countries, looking at important indicators of democratic transitions and consolidation. So looking at, you know, did aid increase the propensity to transition from a one-party to a multi-party regime? What is the relationship with incumbent turnovers? Relationship with a number of times incumbents have tried to change the constitution. Relationship between whether elections are deemed free and fair. So a number of different, more disaggregated outcome variables besides just democracy. And then we combined that with a number of country case studies. So we had seven country case studies. These are all countries that are considered electoral democracies. But there is still some variation amongst them. So we were looking at Benin, Ghana, Mali, Mozambique, Tanzania, Zambia, and Malawi. But as I mentioned, there is variation amongst them. So Ghana, Zambia, and Benin are seen as being more robust electoral democracies, particularly in Zambia and Ghana, we've seen the ousting of the incumbent a number of times. So we've seen at least two turnovers of political parties, which some see as a major sign of consolidation. Then we have two very fragile democracies, at least at the time of doing the research. So certainly Mali, which has experienced a coup in the last year, as well as Malawi, which is in kind of a fragile transition period after the death of the president earlier this year. And then we have Mozambique and Tanzania, which are considered dominant party regimes. In both countries, the ruling party has been in office since those countries became democratic regimes. So we haven't seen the type of turnover that we've seen in our Ganas or Zambia. Dividing foreign assistance into both democracy aid and development aid. So trying to get at least one of these research gaps that I noted earlier from the other cross-country studies. So when we're talking about development aid, we're talking more about aid that's intended to promote growth, reduce poverty and inequality, promote broader social welfare. So aid that goes mostly for agriculture, education, health, transport, and other types of infrastructure. When we're talking about democracy assistance, we mean aid that's going more to support actors and institutions that are trying to promote greater political liberalization in their policies. So supporting elections, parliaments, civil society, things of that nature. And then the other innovation of this research is also trying to take a more sophisticated approach to what we mean by democracy, which is the second research gap I was noting earlier. And so here we've looked at both transitions to democracy, as well as democratic consolidation. And when we're talking about consolidation, we're seeing it as a spectrum with both what can be considered positive and negative polls or endpoints. So on the negative side, we see consolidation as you're trying to avoid something. At the most extreme, you're trying to avoid a transition back to authoritarian regimes or back to a cessation of elections. Just next to that in kind of a middle ground is you're avoiding erosion. So these are conditions where you have, you continue to have elections, but they're not necessarily free and fair. Potential civil liberties of certain minority groups are infringed upon. Potentially the government has tried to change the constitution or has imposed restrictions on other important institutions. And then on the positive side, you're trying to deepen democracy and promote certain elements that we think are important for greater consolidation. So in this regard, we look specifically at strengthening vertical accountability. And these are relationships of responsibility between elected officials and citizens. And there's various mechanisms that citizens can use to try to keep elected officials accountable, one of which is elections, another is using the media, another is working with civil society groups to put pressure and enhance transparency of elected officials. And then we also looked at strengthening horizontal accountability. So these are relationships of responsibility across various government institutions. So looking at the role of parliament, judiciaries, anti-corruption commissions, et cetera, and their ability to keep a check on each other and on the executive. So in designing the project, we were trying to do mixed methods, the cross-country and the in-depth case studies. We were trying to have a more disaggregated approach to foreign aid, the division between development and democracy assistance, which is, of course, could be much more nuanced and detailed, but for the sake of then having generalizable findings, we found it was really important to at least have that distinction. And then finally, having a more sophisticated approach to democracy, so looking at both transitions and then various aspects of the consolidation process. Did find some interesting differences between these two types of aid regarding transitions. And in particular, a lot of the differences for both transitions and consolidation is driven by the fact that these two types of aid very much see democracy in a different light and have different tools for influencing democracy. So development aid tends to see democratization as potentially being a long-term process that comes about through promoting greater social and economic well-being, perhaps the growth of an educated middle class that might have more cosmopolitan worldviews that enhance democratization, while democracy is more interested in short-term actors and institutions and their role in promoting liberalization. They also, importantly, have very different relationships with the government. So often the development aid community sees the government as being a critical partner in their interventions, and we hear this a lot in discourse about partnership and ownership. The democracy assistance community sometimes can see the government as being a hindrance to their activities, particularly if the government doesn't want them working with the opposition or civil society groups, for example. And then very importantly, they have different areas of leverage on governments. So development aid tends to have much greater leverage because it just tends to be a larger resource. So democracy aid tends to only be about, well, in Africa, as of 2011, it was 8% of total overseas development assistance. So it has been going up over time in Africa, and Africa is the biggest recipient of democracy assistance, but as a share of total ODA, it's still a very small share of total ODA. And so because of that, democracy assistance doesn't have much leverage over governments. They're not too upset if they lose support for civil society groups, for example. But development aid does tend to play a very big role in a lot of these countries' economies, and there is much more concern about losing those resources. The development aid community can also use two important mechanisms, one of which is coercion through conditionalities or threatening to withdraw aid. And then the other side is using incentives. So if governments meet certain targets, then they might get additional tranches of funding in the next year or the next funding cycle. And increasingly, as more donors are using general budget support, which is when foreign aid is typically not earmarked. Instead, it goes to support the Ministry of Finance and the activities that the government has delineated in its national development plan. And so many African governments prefer general budget support, gives them some freedom over how they're going to use the money. But the flip side of that is that it means that donors threaten to withhold budget support that has even greater influence than if they're just threatening to withhold project support to a specific sector of the economy. So because these two types of aid have different types of leverage, they then have different indirect impacts on the transition process and then by continuation on the consolidation process. So in terms of transitions, development aid plays a much more direct impact. What we've seen during the 1990s, but even what we're seeing more recently in regimes such as Uganda or Rwanda, is that development aid, because they're using that mechanism of coercion, in particular, has been able to influence governments to engage in some reforms. So they can do this directly by specifically saying, you know, unless you're going to allow multi-party elections or cease human rights abuses, we're going to stop aid. So a critical example of this was in Malawi in 1993, 1992, 1993, when there were a lot of human rights abuses under the dictator then, Hastings Banda, and they reached a culmination at least for the donors in 1992. And all the donors said they were going to cease all non-humanitarian aid until the government held a referendum on holding multi-party elections. And because there was a food crisis at the time, because there was a lot of protest and domestic discontent, Banda, Hastings Banda did hold the referendum and then subsequently held the first multi-party elections the following year in 1994. Kenya is often seen as another example where the donors were able to have that influence on Daniel Erepoit. The economic conditionality is that then provoke economic discontent that filters into demands for political reform. And so we've seen this in Benin or in Zambia in the early 1990s where structural adjustment and economic austerity provoked a lot of the trade unions and civil society groups out into the streets. And they often tied their economic demands to political demands as well for greater multi-party democracy. You know, once the government, for example, like Malawi, decides to hold elections, then the democracy aid community plays a really big role in funding those first elections and making sure that they happen successfully and that we don't have a democratic breakdown and go back to a one-party regime. And this is particularly important in post-conflict situations as well when you're trying to make that transition. Those first elections, such as in Mozambique as well as another good example after the end of the Civil War, holding those first elections is really important. So you need the donors to be able to fund, which is typically very expensive undertaking, to register everyone, create a voter's role in the first place, get an electoral commission up and running. So democracy plays a role in the transition, but it's kind of after the fact. It's after the government has decided to allow those to occur. We again see very different impacts as a result of these two types of aid. So democracy aid tends to play a much more direct role on the positive poll of consolidation that I was talking about. So on these issues of strengthening horizontal and vertical accountability. And it does this because it's specifically often targeting those mechanisms. So for example, as I was just alluding to, electoral assistance is a really important important aspect of democracy assistance. Many countries could not have held and still cannot hold relatively free and fair elections without the help of the donor community in some way. Donor community is really important, as I said, for helping to update voter registers, sometimes even doing an audit of the electoral register, which donors did in Senegal in 2011, seeing what problems exist, engaging in civic and voter education programs, and basically training electoral commissions on how to roll out an election, how to monitor the elections. So we found in a number of our case study countries that democracy assistance has been really key for elections, which are an important mechanism of vertical accountability. The downside is that they tend to be very much focused on the electoral period, seeing donors come in maybe a year or 18 months before the election. And so elections are seen as an event, not really as a process. And this can often inhibit long-term sustainable capacity training, particularly for electoral commissions. So often, not all countries, but many often just have an electoral register that's open right before the election. They don't have a continuous registration process because maybe the electoral commission doesn't have the capacity to do that. And so there's a high level of administrative burden right before the elections, which could have been more smoothed over the time period between elections and ensuring that people know about the electoral process that no one's excluded because of logistical or administrative constraints. In terms of other aspects of vertical accountability, democracy assistance has been really critical for a number of civil society groups and for media organizations that couldn't find resources elsewhere for their activities. The media in particular being able to, you know, just have, being able to afford newspaper, ink, or radio transmitters, just kind of basic expenses that otherwise there wouldn't be the resources for. The downside, again, has been sometimes civil society groups, strictly NGOs or sometimes competing with each other for donor assistance. So instead of maybe coming together and speaking with one voice on particular issues that they're all interested in, can be more of a combative atmosphere for that. And also democracy assistance can't really target the broader legal environment in which some of these groups are operating. And so if there are infringements on media freedom as particularly the independent press, media organizations might not know how to react to that or how to fight against that because they've been getting more assistance in terms of financial assistance from the donor community, not really kind of legal training. Another mechanism of vertical accountability is political parties. Political parties are really critical for aggregating citizen interest and communicating them into the political sphere. And as I mentioned earlier, particularly opposition parties in Africa have been relatively weak because of both financial and strategic limitations. And so party aid does certainly fall under the democracy assistance area. But the problem is that it's a very small area for most democracy aid donors. Civil society tends to dominate most of the assistance that democracy aid community gives, even though political parties could really benefit from having more support from the democracy aid community. And one of the key reasons given for why it's not a very popular area for donors is because it is seen as being more political. It might be seen by the government as supporting opposition parties or giving them an unfair advantage. But from our research, we found that there's some really important critical but technical areas of intervention that the democracy aid donors could be focusing more on. I mean, one is just helping political parties, both opposition and incumbent, better understand what a manifesto actually is, what it entails. Secondly is helping them considering mechanisms for internal democracy, so promotion of the rank and file within political parties so that they're not consistently revolving around just one particular leader or person who leads the party. Because often we see in many countries if there's a debate within the party, members leave and create a new party and creates this party proliferation, which isn't so good for creating party voter affinities over time. And often this is because there aren't very clear mechanisms for advancement within political parties. And donors can also help with just teaching about campaign finance, how to broaden political parties, ability to gain finances so that they're not, again, just dependent on the party leader or particular private sector interest to support the party. So what ways in relatively poor countries can political parties try to start getting membership fees and campaign finance fees that creates more of an affinity between party members and voters. We did find in our research that development aid also has an impact on vertical accountability, but it's much more indirect and not often intended by the donors. So I'll just give one example with regards to political parties. And this is the fact that in a number of our countries we find that the development aid community inadvertently reinforces incumbent advantage because they allow presidents to take credit for development projects, for example. So for example in Malawi again, the donors were partially involved in directly funding the fertilizer input subsidy program from not during the first year of the program but subsequently. And then they were continuing to fund the program through general budget support as well. And right before the 2009 elections in Malawi saw a huge spike in spending for this program, about $228 million just under the government spending on that. And the minister of finance has clearly said that at least two-thirds of that $228 million came from general budget support. And as a result of those elections, the president at the time, Bingu Matarica, he won overwhelmingly, got about 66 percent of the votes. It was the first time in Malawi you didn't see ethno-regional voting patterns because his fertilizer subsidy program was popular across the country, particularly it's a very rural country. And so it was highly popular and the increase in spending right before elections certainly helped his chances in winning those elections. And so it's just a small example of, of course, the donors were just supporting a program in the country's national development plan, but it very much reinforced the incumbents' advantage in office. We've seen other issues like this such as malaria medicines being dispersed right before elections in Vietnam. And you can often see many other examples going into our country case studies, but in other African democracies where a president might be pictured on a billboard showcasing a new road or a new well that the donors most likely finance. Again, here we see democracy aid has more of a direct impact, development aid having a more inadvertent impact. So democracy aid has been really important for a number of parliamentary strengthening activities, has been involved in anti-corruption commissions and judiciaries, also police and military training, but mostly on parliaments, that's where a lot of assistance has been focused in the area of horizontal accountability. So these might be training workshops for parliamentarians, it might be helping to finance a library or for them to have resources to go to at the time of budget discussions. The problem is that a lot of these interventions are not very well coordinated, so you can have different donors coming in at different times, you know, each having a five-year cycle program where they're going to work with different MPs. They might choose different approaches, so USAID and Malawi, for example, likes to work with individual committees, DFID might instead have three-week training courses for all MPs. So they're using different techniques, of course they're also hindered by turnover within parliament as well, but it's symbolic of a larger problem within the democracy aid community, which is not so problematic in the development aid community, which is that there tends to be more fragmentation and less coordination than the democracy aid community. The development aid community has some indirect effects, as I mentioned, on horizontal accountability, particularly with regards to parliament. And this is particularly through general budget support, was found across all of our country case studies to actually further sideline parliaments. So even as you have a democracy aid community trying to strengthen parliaments and get them involved in budget discussions, general budget support, because it's focused more on executive decisions in particular the Ministry of Finance, doesn't allow that same type of engagement with parliamentarians. A lot of them learn what donors are doing at the time of budget approval, and at that time they're often not given enough time to actually scrutinize the budget and see exactly what donors are doing. General budget support is also seen as a grant rather than a loan, so typically parliaments don't need to approve grants. This varies across countries, but a number of the cases we looked at, the parliaments were only required to approve loans, so they're often not aware of actually what the donors are doing in their countries. And that's a broader issue of potentially lack of communication between the donors and MPs, not having maybe dissemination activities that informs MPs about what they're doing. Here knows where we see the development aid community having more direct impact, and this is again because of this coercive mechanism. So if we're thinking about human rights abuses or changing constitutions, this is where donors can come in and either threaten or actually withdraw budget support in particular to get governments to change their reactions. We've seen this a lot recently with Diffred and USA talking about potentially withholding aid to countries that don't support, at least withholding aid to countries that are in prison or criminalized homosexuality. It's a very controversial mechanism, but it's been something that these two donors have at least considered recently. That's a big step forward because when we're looking in our case studies, most of the time, despite having this rhetoric about promoting good governance and human rights, has actually been issues around corruption or macroeconomic stability where the development aid community will be most influenced to act and to act in a concerted manner. So in Malawi in 2011, we're starting to see a large deterioration in civil liberties, further criminalization of homosexuality, repression of the independent consideration of greater repression of the independent press, an injunctions bill was tabled whereby people were not allowed to sue the government. And at the time, we were only seeing Germany, it was one of the first to react, saying it was going to withhold half of its budget support and we didn't really see the government reacting in any way. And it wasn't really until late 2011 where the government was still very intransigent about devaluing them allowing quacha, which had been a real big concern of the IMF, where we saw large-scale reaction by the donor community. The IMF halted its support and as a result, all the budget support donors had to follow suit and halt their support. And one of the reasons for this is when you look at countries performance assessment indicators or the indicators in their performance assessment frameworks, which details their disbursement criteria for budget support, they rarely have specific criteria on human rights and democracy. So it might be part of their memorandum of understanding with the government and their partnership principles, but when you look at the actual disbursement criteria, they'll be very erratic on human rights. In Malawi, for example, they focus on domestic violence towards women, that's their human rights criteria. And that's just by a few of the donors, that's not by all of them. And often when you look across the budget support donors, the one thing that they will agree on these macroeconomic and corruption indicators. So that gives them a platform for more concerted action in that domain. There are some really important complementarities between development and democracy aid, but that there are also some key tradeoffs. So I alluded to some where democracy aid is trying to strengthen, a little bit trying to strengthen political parties and elections where development aid might actually be reinforcing incumbent advantages, where democracy aid is trying to strengthen parliaments, development aid might actually be sidelining parliaments. So there's important tradeoffs between these two types of aid. That I think the research really highlights better than other research on aid and democracy, because by looking at these two together and their interaction, it's more useful for practitioners, specifically since democracy and aid units can't work in isolation of their colleagues in health and education and agriculture units. They should be working together and thinking about ways in which they're at least not undermining each other's initiatives. And you really, even if you're just looking at the impact of democracy aid alone, you're really not getting to see some of these interactive effects that come across in our research. And I think, secondly, the research really highlights that foreign aid is political, does have political consequences, even if it's not always intended to. So I mean, there have been some scholars, including Paul Huyer, who said that donors should not get involved in governance issues and in politics. But based on our research, it's pretty clear that it's really hard to divorce foreign aid from what's going on in the domestic political sphere. And so even though domestic actors will always be much more important in the democratization process, we do show that external actors play a really important role, both directly and indirectly, particularly in such an independent region of the world.