 Good afternoon. I have been very much in doubt about how I should plan this presentation. I've decided that I'm going to be slightly provocative and I would like to say up front that this is done in all due respect, okay? So just I... But I'm going to try to tell you a story about what I consider a deep failure of journalists and a great success for an aide. Does aide contribute to economic development? Well, that's a question that's asked in very many places and Wider was asked a few years ago to do a research program about it. So we were contacted by Anita and Sita and we were asked whether within this framework our work program that we would try to come up with a series of answers to some difficult questions and as Wider is ranked the sixth international development research institution, we said, okay, let's now try and face that and from the very beginning we were perfectly clear that aide is very complex. It's a lot of things. No single individual can claim that he or she has all the insights. That includes me. I have some. I've spent more than half my working life in developing countries. My kids are born in developing countries. I've spent a major part of my life traveling, working and so on and so forth, but I have a slice. But it is of course adding a little bit up. So we initiated this so-called Recon program and there are some links here which you can consult. I leave the slides behind and I hope you will excuse me that these slides are a good example of academics who put too many things on their slides. Now we were motivated by four questions. What works? What could work? What's scalable? What can you scale up? Because it's not particularly interesting to have a small NGO doing something in a small corner very far away which if it can't be scaled up when you're looking at it from the perspective of the overall national or international development. It can be very interesting in its own right, but from our perspective scaling up is important and what's transferable. Now I'm going to give you one slice of this big program. I'm going to give you a slice which I myself have been involved in for the last 30 years. Erson academic and Erson 8 practitioner. And we started to start there because the so-called macro debate about aid and growth is the one that you journalists pick up and you write tons of pages, tons of notes on it and you write the wrong things because you take the scandal stories and you haven't actually sat down and try to see exactly what the empirical academic evidence actually has to say. Why is this debate elusive? Why is it complicated? Well it's complicated because there are lots of critical voices out there, lots of people have lots of opinions. I once did a count of what academics had to say and 9 out of 10 said that aid was working and had a sensible rate of return. I then counted journal articles in newspapers, this was in this case, in the Danish case, 99 out of the 100 were negative. Kind of interesting. Now these aid and growth studies, they very much frame the debate, the development debate about foreign aid and that's a pity. Why is that a pity? You have for sure heard about Dambisa Moye who says that aid doesn't work. She was picked up because she was provocative. Nobody has ever asked the question who actually paid for your education, who actually paid for Dambisa's education for her to become a senior Goldman Sachs employee for an aid. Did ever anybody ask whether she was about to flunk her exam, what she was? It's a great pity that these negative voices which are in many cases unsubstantiated that they catch your attention and when I say you please I started by saying it is meant in all due respect but I'm trying to get a point across. Now one of the problems in this is that there are stories about aid working at project level because we do know that there are some wells that have been docked, you get up some water, some kids get this clean drinking water and that's very good. There are tons of examples of that. There are also examples of things that didn't work. But what's the net on balance? What can we say? Well this is where people then refer to the aid and growth debate and they say oh economists cannot find it. Well some economists cannot find it. But I mean that does not necessarily mean that the impact is not there. But you see the problem has been that because some macroeconomists have said it's not there then a lot of people are saying oh but then it doesn't add up. Something goes wrong as we move up from the micro from the projects to the macro and thus it's useless anyway because if you can't demonstrate it in the big picture it's useless at the micro level also. I mean try and think about how many assumptions are made in that sequence of things and how stupid that can actually be. But that's how the debate goes. That's how the debate has been going for 35 years and nobody has broken it so far. Okay is it really true that the impact of aid, that the value of aid sort of evaporates as we move up from the project level and then up to the big macro? Can we say anything on balance? And let me say up front that yes it has been very difficult in the past. Why? Well because we had not very good data. You know aid started around 1960 and it takes time right and economists they have to have some kind of data set before they can start saying sensible things based on these data. There are also ideological debates. There are simply people who don't like that you give aid. But that's one thing. I have nothing to say about people who just say I don't like aid giving. I cannot as a professional economist have views on that. I mean people are allowed to have their views. That's perfectly respectable. I disagree personally. My values, my norms are that when people are in need well you better engage with them. But if somebody says well I'm perfectly happy part of this world is hungry. I mean my background is that as a relatively young man I got engaged. So I've seen children dying of hunger. I've seen them getting shot in war and not getting help. I kind of cannot kind of agree but that's a value judgment. So if people think that it's okay to leave them alone. Well okay then there's nothing more to talk about. But there is something to talk about when we are talking about it analytically and what should get out in that public debate. I mentioned the data that's another reason why it's difficult but I can tell you one thing. They have been building up. They are now longer. We have more countries so we can start saying things. Now let me try and get just one thing across which I know is difficult but it is absolutely fundamental that this is grasped in this discussion. It is what we call attribution. The problem is the following that when we are talking about whether aid helps economic development we have to compare to something. I mean we can't just sort of in the abstract say well does it help. We have to have something that we compare it to right. I mean you know I mean it's like is this vaccination program going to work. Well we have to then establish what would have happened had that vaccination program not been there. Anybody who uses the following argument oh it's going so bad in Africa at least that's what they did 10 years ago. So thus aid does not work. That person will flunk any first year exam because that's nonsense just because it goes bad in Africa does not imply that it does not work. You have to put on the table how would it have gone without the aid and then what you can do is that you can compare and it may well be that there is an impact even if you're still sliding down the road especially because aid is pitiless small. Do you know how big aid is? Dan Bissemoyer talks about it as if it is massive fantastic amount of resources Belize that he talks about this international huge business. Do you know how much money an average aid recipient country receives in aid per person in that country per year 25 dollars. So do you really think that that 25 dollars can do so much harm as we hear very often? That's less than the poverty rate for less than one month. So I mean isn't it okay to just at least pause and think? I think it is. Now it is very very difficult to establish that causality that you really can say aid causes or aid does not cause but one of the reasons why economists debate this is that sometimes we fight about statistical significance. In order to establish that thing we compare to what would have happened if no aid has been there? We are using statistical methods. We are using different tools. We use our computers to try to help us get that thing we compare to. Now statistical significance is very important because we are only allowed to make firm statements if we have statistical significance. If we don't have statistical significance then we are meant to shut up. Unfortunately there are lots of economists who don't shut up. There are lots of economists who use absence of evidence non-statistical significance as evidence of absence. Can you catch the plain words there? I think that's incredibly important. There is not one single statistical significant statement in Dambisa Moyo. There is not one in Eastern, not one. There is one in Eastern's other book and that one is wrong and we've shown it. By the way, when it says the white man's burden have you ever thought about how Japanese would want to characterize themselves? They are a major player in international aid. So it's not quite the white man's burden only. But anyway leaving that apart. So I want to really stress that just because there is a lot of absence of evidence and by the way there are lots of people who have just been doing this. They don't see the actual evidence. Just because we have been arguing about it does not necessarily prove that it's not there. Okay? And that's one of the reasons why we engaged in this program. So this slide I'm not going to say very much about. But I'm going to just make one statement and then you can look at it afterwards. What is development about? How does development happen? Does it happen within half an hour or a year or two? No. Development is a long cumulative process. Think about our own countries. God, I mean Denmark? Jesus where I'm from. I mean I know all the stories about our drunken king whatever in 1400 and this and that and he raped his this and that and then so on and so forth. I mean come on. Development is a long process. And what happens in the development process? You accumulate physical capital. You accumulate human capital. By that we just mean better skilled people. Just accumulate year after year. My kids are brighter than I am. Their kids are going to be brighter than they are. At least they have no more. We're going to have more roads, more infrastructure, better IT, institutions as well become more efficient as time goes when development takes place. So this means that development takes a long time. So the idea that you can capture the impact of aid over a five or ten year period is just rubbish. There's a lot of work that tries to do that. But that doesn't say anything. That does not prove whether aid works or not. Okay. So what can we now say? And here in a way I would want you not to look at my slides. I'm just having them there because you can then consult them afterwards. But I'm now going to give you four very specific answers based on four very specific studies that are up to date. The first conclusion, what can we say on average? Across countries in general. Well the average story is that aid works exactly as economists say in their economic growth theories. With the amount of money that we put in, we're getting exactly the impact that we would predict from theory. No more, no less. But remember how big was aid? $25 when it's really going high. On average in aid recipient countries, how much is it per person? $3.5. $3.5 per person per year in aid recipient countries on average. That's what we're talking about. But the average big story is we're getting exactly what we're paying for according to economic theory. There is no macro-micro paradox. It's a myth. Now let me just go once again. That does not mean, and I want to stress that, that does not mean that journalists cannot find white elephants. You can. They're out there. There are lots of failed investment projects in the North, or in developed countries. Can you remember the .com experience? That was pretty damn expensive. Right? Okay. Now that statement I just made about the aggregate, that's comparing across countries. We're trying to make a very general statement. We have then studied 36 countries in Africa, and we have looked at them, each one of them individually, over time. Why is that important? That's important because when you have time series evidence for different countries over a long period of time, then the evidence starts piling up. I'm working with Catalina Eusele, who's a Finnish leading international economist on this, one of the best econometricians in the world. She's part of the so-called Copenhagen School of Time Series Econometrics. One of them was about getting an Opel Prize a couple of years ago. You're talking about the best analyst in this type of data in the world. She has for 15 years told me, Finn, I've certainly been asking, but Catalina, can we please collaborate? I mean, I work on this aid and development, and I need you technical expertise to work with me on this so we can come out with what's the true story and say, forget it. It's too early. The data are not long enough. All we would get, she says, is statistical insignificance until the time series data is long enough. When the time series data is long enough, then we can try. And we started trying a year and a half ago and said, now's the time. In 33 countries where we studied them over time, there is very significant statistical evidence to say that aid has made a positive contribution to macroeconomic indicators. That's three countries. Well, we can't get that story with a simple model to work. Which are those three countries you can ask? Let's take that in discussion round. It's also clear that it's heterogeneous. It is as I stated in the beginning. It's a very mixed picture. But in general, across time for 36 African countries. The third conclusion, when you look into, when you try to take into this aggregate impact, and you sort of, okay, now I want to understand, what does this mean in more concrete terms? So what does this mean in terms of number of poor people? What does this mean in terms of the share of agriculture? What does this mean in terms of the share of industry? What does this mean in terms of number of children that have been vaccinated and so on and so forth? You can unpack the whole story and it works. It hangs together. With one still complicated issue. The data do not yet show that education, that aid to education, aid to education we know has an impact. We cannot yet show that education has an impact on economic growth. We know it has. Of course it has. But we can't show it with the existing data. That's the one foot note. But the rest, everything hangs together nice and statistically significant. Then there is something called a meta study, meta work. There are a couple, there's a Danish economist and Australian economist. They have gone out and they have collected, they say, all of the studies in the world that have been done in this. And then they say, oh, when I put all of this together, then they get, what do you think they get? Do you think they get statistical significance or insignificance? They get an insignificant result, they say. And then they say, oh, there's nothing. So they use absence of evidence as evidence of absence. Now, and then one of them cracked a comment five years ago and said, nobody's ever going to check us anyway, because this is so complicated. Now we did. And we discovered that these guys who are widely published have been in all of the newspapers in the Nordic countries, all of them, saying that they have proven that aid doesn't work. What they do is that they put bananas, apples, pears, so many different kinds of things in the same basket and say they are to be considered the same. But I kind of like to think that there is a systematic difference between an apple and a pear or an apple and an orange or a banana. And I would like to control for that difference. And when I do that, you know what happens? The results turns from being insignificant to being statistically significant. And with approximately the size that economic theory would predict. So the evidence is mounting and I'm putting it like this because I think it would be good if US journalists would sort of keep an eye on it because there is no question that the average citizens, and we know that from all kinds of surveys, they have a pretty negative view of aid in part because the reporting on aid is not positive. It is easier to make a story out of a bad thing than it is for this average thing that just says, oh, it works. I don't know. I'm not a journalist. How that, I mean, and I'm trying to, and I've decided to do it this way because I really genuinely would appreciate your thinking about how to crack that knot because I haven't been able to. I've been unable to kind of get people fired up. I don't know why and that's why I'm maybe a bit more direct today. We have a coherent set of evidence and it's sitting there on the website. We will by the end of this year having 200 studies on all kinds of possible dimensions of aid you can think of. 200 done on all kinds of things ranging from support to well-dwelling in whatever, Burkina Faso to the macroeconomic impact to the post-agenda to, I mean, so I would suggest if you are looking for good stories, at least take a look. And those papers have not been prepared with one particular target. I have not been sitting there saying you must reach this conclusion, of course not. This is research. So people do their research and they come up with a result. Now I would like to say that the fact that there's some aid that does not work and we know that is clearly the case, does not imply that nothing works. When you calculate the average return to aid over 35 years and you have to have that time perspective because as I said before, development is a cumulative process. Do you know what the average rate of return is? So that's like the return, like if you would put the money in the bank. What return do we then get? Let's say that you put that aid money into the bank each year and just let it stay there. What do you think the rate of return is? On an annual basis? It's 16%. That's pretty damn good. I mean, if I could get 16% of my pension fund returns, I would be absolutely happy because then I would stop working tomorrow. 16% is very good. The World Bank used in the good old, good old high interest rate days, they used the cutoff point of 10%. So the average is 16%. Something must be going okay. I am in no way trying to whitewash aid. I'm in no way trying to say that all relations between developed and developing countries are good. I am probably more critical than most but I do not think that one should argue for stopping something that does seem on the best available evidence to have some impact. So is that a story? Is that a news? I don't know. Thank you.