 Anna actually did such an excellent job that I probably have to skip some of my slides because some of the literature side the same and especially for the motivation of some of the variables. So thank you for that. My name is Mina Baliamoulout. For those of you who were not here yesterday when I introduced myself, I work at the University of North Florida where I am a full professor but I am taking a leave and I'm going to be director of research at the African Center of Economic Transformation in Ghana. The real starts on August 1st. So my paper has two folds. There is the first part where I try to contrast or at least review briefly what happened in Mauritius and Madagascar to IMD economies in Africa. And I'm interested in these two countries because for Madagascar, Madagascar tried to replicate what Mauritius did and yet when you look at the situation in Madagascar today, Madagascar by most standards is a poor country compared to Mauritius. And I'm going to go over mostly it's a story using graphs. And the second thing I do is after I look at that then I say well what is really very different and if I don't want to limit myself to the institutional and policy explanations which I do in another paper, I did in another paper then I want to look at other things, you know things that we can change. And it turns out that infrastructure is part of the story and human capital also is a big part of the story. And so building on that I go to the second part where I do an empirical analysis of a group of African countries using variables that could affect some measures or proxies for or indicators for industrial development. So this is a graph that shows the GDP per capita in purchasing power parity values and the growth of GDP for the two countries. The red colored ones are for Mauritius and then the others that curve is supposed to be dark blue. So the blue ones are for Madagascar. And as you can see the per capita income for Mauritius, it grows very significantly over the period of 1982, 1981 to 2011 so the last three decades while on the other hand if you look at the per capita income of Madagascar it's actually declined slightly over that period which is a scary thing. If you look at the GDP growth rates Mauritius what's really interesting here is that Mauritius has low volatility in growth rates and Madagascar has very high volatility although sometimes they have high growth rates but then when it is negative growth it's also if you look at this year here 2002 it dips a lot. So there is volatility, there is on average lower growth compared to Mauritius and definitely much lower per capita income. So what went wrong? And then I started looking at a few other things. So because both of them focused on exporting manufacturers, textile mostly in the beginning. I compare here the value added of industry as a share of output and manufacturing as a share of output. And if you look at the two countries throughout this period, most of it at least, the difference is huge between Mauritius and Madagascar again the two red lines curves are for Mauritius. But what is interesting is that if you look at the period in the 70s, the late 70s period or second half of the 70s you see that Madagascar actually had a higher industrial value added than Mauritius's manufacturing value added. So it was high but it went down. So I'm missing data. I cannot talk a lot about the 70s because there is no data on manufacturing value added for Mauritius. But the point here is that there is huge differences for the absolute majority of those years. Then I look at the exports of manufacturers as a share of merchandise exports in the two countries and again Mauritius outperforms Madagascar. But if you look at the 70s, 1975 for example, 76, they were very close. The gap has increased over time. And then I use a few other measures for although it's not really a perfect measure of diversification or diversity of products, but at least if you know that a country is exporting lots of products, chances are that some of them are really linked to the primary sector. So again Mauritius outperforms Madagascar and then I use this measure of which I call sophistication of exports. The UnicTide has some data that show export dissimilarity with developed economies that... So this is like Mauritius exporting only to developed economies and Madagascar exporting only to developed economies. So I take those exports and I take the measure of dissimilarity. So the higher the measure, the more dissimilarity, the lower the measure, the less dissimilarity. And so if a country is exporting developed economies goods, then that country, his chances are they have more sophistication in their exports. And what's interesting is that Mauritius actually did not have a lot worse in the early days, in this period here in the 90s than Madagascar, but dissimilarity with developed economies went down if you go to the 2000s. So the red line actually is below the blue line there. So that doesn't really give me something that's consistent throughout the period to say that Mauritius definitely has more sophistication. And then you look at high technology exports, and again you see that it's in 2000s Mauritius exports a lot more high technology products within their manufacturing exports than Madagascar did. The problem with these two last measures is that they don't give a consistent story throughout the period. And that's why I don't use measures of sophistication per se, except that I actually do still look at high technology exports, but I'm thinking of it as just a part of diversification. The global competitiveness of the two countries, again, is also huge, the gap is huge there. If you look at both the rank and the scores of Madagascar and compare them to Mauritius, you see that there is a huge difference. Just like, for example, the value chain breadth, Mauritius is ranked 28th in the country among 144 countries that were surveyed in 2012-2013 reports. You look at things, the control of international distribution, 23rd in the world, but you look at Madagascar, it's 112 and 134, so it's much closer to this. This is a measure that the global competitiveness report uses to see if these countries add lots of value to the things that they make within the country before they export them internally. And I put the reference so that if you want to see how the calculation is made. But Madagascar doesn't add a lot. And then I look at some selected indicators of infrastructure. And the problem is that for some of the indicators, it's very hard to find data over a long period of time, so you just have points. I took, like in the first one, you have the paved roads as a percentage of the roads in the country. So the first quadrant here or half over here has Mauritius. And if you look at the height of those areas, there are definitely, there is no comparison between Mauritius and Madagascar. Madagascar, everything is very low. And Mauritius, everything is high. For example, the density of paved roads is 98%, close to 100%. While if you look at, if I compare just what I have there, which is 1990 and 2001, you look at Mauritius, it's actually went down, right? So the blue area there is higher than the red one, so it went down. And then mobile cellular subscriptions, although they are very imperfect really measure of infrastructure, if you ask me, they went down in 2009, but still no comparison, 31% versus 84% in Mauritius. And the last one, which is access to electricity, again, you are comparing 19% to about 100%. Huge difference there. So definitely there is a very significant difference in infrastructure. And in this graph here, I look at the urban population. Now the urban population was interesting about Mauritius is that it stayed roughly stable, okay? Over that period of time. And it is quite high, it's definitely higher than the average for Africa. In Madagascar, it went up significantly, but for most of that period, there is a huge gap between the two countries. Even towards the end, it's still, the gap is still important, more than 10 points. I looked at human capital, just using secondary and tertiary school enrollments rates, and what is really interesting is that in 1985 and 1990, these are points for those years, tertiary education in Madagascar was higher, the rates were higher than in Mauritius, okay? So this is even going to 1990. But look what's happened since then, 2000, 2000, 2005, higher education or tertiary education just went up a lot in Mauritius while in Madagascar it stagnated. The increase in secondary enrollments also is very, not very strong in Madagascar while in Mauritius it went up a lot. And so there is a huge difference there too. Okay. So building on that, I do an empirical analysis where I take data on the variables, some of the variables that I was comparing the two countries on and then look at their effects on six indicators of industrial development. And again, I mean, if you're going to tell me that these indicators are imperfect, I completely agree with you. There is not one single measure that we can use and say this is industrial development in the country. So the best thing we can do is really look at different indicators and then see if they are consistent with each other or the results are consistent. So the first thing I do is that I look at the share of manufacturers in total merchandise exports, how it behaved over time. Then I look at the share of manufacturing in output, so that's a different measure because we're not looking specifically at exports but we're looking at whether the country is diversifying both the production and the exports, okay? So a country could have diversification manufacturing and if it's a big market, they might not need to export a lot. On the other hand, if it's a small country, they might need to export more. They will have to export a lot. The third one is the share of industry in output and this is very... It's a very controversial measure if you are looking at natural resource rich countries. Why? Because lots of the industry can be linked to natural resources in which case they might not be really diversifying or doing any manufacturing. It's just natural resources and then they build an industrial sector, connect it to that and then they take it from there. But nonetheless I look at it and then the fourth one is the share of high technology products in total manufacturing exports and I'm using it here not as really sophistication but just if you are doing some of that then at least you are doing some of little bit of manufacturing or diversification. The fifth is a normalized Hirschman index of export products concentration. So I look only at the... Onycta has these neat data that look at these things but they're detailed. So I look only at the export side and then see if those exports are concentrated in a limited number of products or more. So the higher it is that index, the more concentration the country has in a few products. And then the fourth one is the number of products exported, which I used earlier and again it's not a very perfect measure of diversification but the chances are if you are exporting lots of different products that you are doing little bit things that are more than just your primary sector, links to your primary sector. And then on the right hand variables, I use urbanization and I talked about agglomeration and the big cities and so on. So I am using that thinking, there is no telling whether urbanization as a measure of agglomeration will really have a positive or negative effects on manufacturing for the reason that Anna explained and I'm going to go over a couple of them. Then I use as a measure of infrastructure the electric power consumption per capita. If you were in yesterday's, the first plenary we had there was lots of talk about electric power and as a matter of fact, Professor Nduli was saying power, power, power, right? Okay, so I have another reason why I'm using that and I'm going to tell you in a second. Besides the fact that also roads, paved roads, the data are so terrible that there is no way you can do anything with panel data using that. And then I use for a measure of human capital for all the stuff we have in the literature about the role of human capital and the skills in developing a diversified economy. I use a measure of human capital that is secondary education or enrollments for most of the measures and then when I look at the exports of high technology products, we want to look at tertiary education or enrollments rather. I also look at natural resource rent for a simple reason. Natural resource rent is an important thing to look at when you are doing African countries especially because of the role natural resources play. You heard a lot and you read a lot about the possibility of a natural resource curse in these countries, in most countries. But I use it also for another reason and that's the one also that Anna touched upon is that natural resource rent might actually lead to development of big cities or high urbanization without any diversification or structural change taking place. And I control for FDI because manufacturing in some countries was based on export oriented FDI. So foreign direct investment goes to those countries from multinationals, foreign multinationals to build a sector from which they export. I mean China for example did go through that at least in the early days. This is going fast. And then I include also trade with developed countries. I have a paper that's published in the review of African development. I can't remember I think 2011 or so that's I title development or growth by destination where I show that if you export to certain countries you might gain from that versus exporting to other countries. But here I look at both import and export to developed countries. Okay here is the reason why I used electric power because for this conference to show really the industrial development we're looking at electricity and electric power. Another reason, another thing is for urbanization. This is the map of urbanization in Africa and the two lighter colors are for the countries where there is very low urbanization. And you can see most of Africa is really very low. But you have the high urbanization happening and then when I looked at this map before I came to the session it just gave me another idea but I forgot to include this in my paper but I'm going to do it in the empirical section is control for the location on the Mediterranean and the Atlantic Ocean. If you see that most of those with darker colors are really around the Mediterranean area. North Africa most essentially and then West Africa and South Africa down there. So the ocean effect might be strong. So I'm not going to go over every single count so don't worry. But just to give you an idea of how many equations they did. So for every single indicator I did, I tried all these things and I looked at the interactions and so on. One important interaction that I looked at is the interaction between infrastructure and urbanization. If you go to some of the African cities today, they're big, right? You have these nice buildings. I was in, I go to Accra a lot. And I see all these nice buildings and nice things and everything and they're trying their best for infrastructure. Actually the roads now are slightly better. But there is a lot missing in terms of infrastructure. So you can have urbanization without infrastructure going hand in hand with it and that can cause problems. So I'm looking at the interplay of infrastructure and electricity and urbanization. And the second one going to the work of Golan and co-authors is this effect from natural or actually the other paper. From natural resources, rent on and urbanization. So I'm looking at the interplay or the interaction between urbanization and natural resource rent to see if it's going to have a negative or positive effect. So anyway, so from this I'm going to just talk about the results because I'm running out of time. What I find is this, overall for the most of these indicators urbanization has a definite independent positive effect or a positive independent effects when it's just alone. Positive effects, so that's agglomeration is good. I also find that infrastructure has a positive effect. The problem is that in some of these estimations the interaction between infrastructure and urbanization has a negative impact. And I explained that by the fact that infrastructure has not been going at the pace of urbanization. And so you get lots of problems. The other thing is, again, I find in some of the estimations that there is a negative effect from the interaction of resource rent and urbanization, which we were talking about earlier, which is also a bad thing for Africa. And then the third thing equally important is I don't find that human capital makes a contribution, a positive contribution or a negative in three indicators. It only has a contribution in to high technology exports which our share education has a positive, significant effect. And then for concentration measures, so when I use the last two concentration measures, concentration and number of products, there is a positive effect that's good. But not for manufacturing share in output, industry share in output or manufacturing exports. Now, this does not mean that human capital is not important actually. Human capital does not appear to make a significant direct contribution to these three variables that I, indicators that I listed. But that might mean that the skills that are needed that go with education are not there. Because if you look at secondary environments, they're going up for the absolute majority of countries in Africa and significantly for that matter. But it's not working. What it means is that this might reflect the low quality of skills that are supplied in the labor market and that education is not generating the appropriate type of input that we need for manufacturing. And I use that as one of the, or I qualify that as one of the reasons why Africa on average appears to be taking one step forward and two steps back. And the second reason for that title is this negative effect from the interplay or the interaction between electricity and or infrastructure and urbanization. Just one last thing that I'm going to do, I'm going to stop, I promise. I give you two examples here of Ethiopia and Gabon. Ethiopia has one of the lowest rates of urbanization. The country with the lowest rate is Burundi, but I don't have data for Burundi. So, Ethiopia has the second lowest. And if I use the interaction to model it to find the critical point because I find there is a threshold for electricity to work. I find that current rates now, because urbanization is 16.8% then the level, the threshold level from which electricity you should have would start having or infrastructure have positive effects means that Ethiopia should move from 54.3 kilowatt per hour to 80 kilowatt per hour. That's not a problem. Given the urbanization, if it doesn't change urbanization. If you look at Gabon, which were the country that has the highest urbanization rate in Africa and in 2010 it was 85.8, now it's over 86. Gabon now needs 4,070 kilowatt per hour in electricity, but it only has 1,004, which means that Gabon has actually a lot more job to do even though Gabon we know that has more organization and everything. So that interaction is very important and I'm going to build another paper around that. Thank you. Sorry, I went a little bit over it.