 Good afternoon everybody and thanks for attending my presentation. Thanks for having the conference and for having a chance to share what I have been doing in the last three years at the Center for Development Research in Bonn. I'm based in there as a PhD student and I've been working with colleagues there. Joachim von Braun was a big support and in both countries I had local support which I couldn't present the findings I'm presenting today. It's an attempt to do a synthesis of two chapters of my thesis and I haven't done that before, so bear with me if I... Yeah, maybe the consistency can be improved. But I will try to share and I will start with a brief outline of the trend of large-scale land acquisitions in the last six, seven years and I will then say how I conceptualized the phenomenon and how I went about the analysis and then I will come into the case studies I picked. It's a case study approach so there is limitations for generalization but I still think there's some findings which might hold true for other settings. We can discuss it afterwards and I'm happy to see that there's more research coming because there is need, obviously I leave a lot of blanks. It's difficult to get data on these land acquisitions. The attempt was done by the land matrix who use media reported deals and cross-checked them in country with partners. Those are the numbers they have produced. They are like a week old. They update and continuously you can go on the web page even if you know a new case you can tell them. I have doubts about the total reliability because there's media bias and everything but it's a nice approximation. These are the investing countries so you see Europe is there, Africa has quite a lot of deals but Asia is a big investor and South Africa also from Latin America we have a number of investments. Those are the recipient countries where the investments are targeted. We see that East Africa is very prominent, West Africa as well and Southeast Asia, Laos and Cambodia receive a lot of attention. This has partly to do with how vividly governments also promote these investments and what is the situation of land availability. What is the situation? There's a number of risks. Land grabbing is often labeled so if legitimate users lose their right to land be it the private farmland or the communal land around the village is a big threat. The unsustainable resource use is another threat not to be underestimated exploitation of labor but there's also opportunities. It might generate additional employment opportunities which then reduce poverty. Market access might improve as well as infrastructure. The question arises can they serve as an engine for growth and can this growth be inclusive? What do I mean by inclusive? We can discuss on that. I insert Eric Tobbex because he mentioned it this morning. I consider if it's pro-poor so the poorest part of the population is gaining at least proportionally to others and so inequality is not raising. I see those impacts not happening directly but through a number of channels. I identify five main channels for my research. There might be others in the two settings I was working those I think are the most important ones. Here we go. Three of them I consider as the factors of production so land maybe not surprising as the main transmitting channel labor and natural resources. The values of these factors might change land might become more valuable or less transferability of land might change and who gets access. Similarly for labor there might be more jobs but who has access to them and what is the wage level? Natural resources I think is important to underline that small holders and rural population they're not only farming and business people but a lot of them are also foresters in a sense so they use forest products to complement their income in the dry season. Natural resources surrounding these communities are important and I will show how that is the case for the Ethiopian case. Technology could be a big impact. Martin was telling yesterday at the opening that that is one of the ways to overcome poverty traps technology and knowledge about new technology this could spread through an investment potentially they could introduce new crops new production new ways of producing and also organization of production. So this is if we think an institutional economic power that is the kind of where the contracting and the organization of production might change and then there is the institutional level and the market level so property rights could be affected or violated and the property right regime might change so customary systems might be challenged and because the center might enforce the legal system in an area where before it wasn't enforced so there might be for the local setting kind of distinct change. How do I go about it? What I'm trying to do is impact evaluation and the biggest problem is attribution. I mean the world doesn't stand still while these investments happen there's other things happening and some of the change you will only see after a certain amount of time. Broadly there's two type of impact evaluations you could do one is the counterfactual based impact evaluation which is very rigorous you have a counterfactual and then you can see what is the treatment effect and you can measure effect. I'm not able to do that in my cases. There was no good data for the Ugandan case and in the Ethiopian case it's an early stage so it's too early to say anything about the impact. The data I collected could be used as a baseline in three, four, five years to do some measurements but so what did I do? I combined an ex-post and an ex-anti-analysis to share, to look at the both case studies. So one case the Ugandan case the Ethiopian case is an early stage investment Ethiopian was promoting these investment activities since 2008 and there was a big number of investments coming in and the case I looked at also started and I used the survey data I collected in 2010-2011 as a baseline to calibrate a model which then shows some potential future evolution of the project. The other case, Uganda is an old investment it has changed in the ownership but the routes go back to early colonial, post-colonial times in the late 60s and there I also was there in 2011 to get my data so I could use the observation from then and try to understand how we reached that situation so it's a more analytical narrative approach institutional economic like and today I will try to combine those two. Data sources so I used a lot of qualitative sources to get myself acquainted with the situation in both settings I spent significant time in both country contexts I mean that's what a young researcher is good for he doesn't have the big support but he has time so he can go there and sit and talk with the farmers and that helped a lot I understood much better what the situation was but I was also able to do a survey not very big 140 households in Ethiopia and 117 Uganda but still it tells something about the local population and yeah I will show some of that this is the Uganda case we're located in the eastern part of Uganda close to Kenya this is the main road going to the port and to the coast so quite a lot of traffic there's markets and shops around the road it's one of the poorer areas in Uganda and you see here those green areas are wetlands and the red area is the investment the investment has a history it started as a Chinese rice development scheme after 20 years the Chinese left and the government took over and was operating as a state farm and then it was privatized and now it's operated under an Indian UK private company the other case Uganda, Ethiopia is located in the very far western corner of Ethiopia Gambela region is the lowland part of Ethiopia bordering southern Sudan it's quite remote so in that sense it's quite different from the Uganda case and it's not very densely populated it's only a few people per square kilometer and there's two distinct ethnic groups in the Uganda case it's much more mixed there's a number of settlements those dots are the local villages and I did surveys in all these villages the investment is planned to be 10,000 hectares this is the size of the investment and there's a lake where they use irrigation both investments are producing rice so I have a similar crop to compare and in the simulation I will show the results of later I basically assumed an area of 100,000 hectares being the affected area so I used proximity to define the local context which I'm interested in I'm looking at the local implications and then I run some simulation where I take away 10,000 hectares out of these 100,000 and see how the situation changes but I come to that in a minute this is the situation today in eastern Uganda it's the rice producing part of Uganda for lowland rice and let's see how did we get there so I identified four drivers that led to the conversion of wetlands so today all the wetland is converted into rice fields situation you know from Asia for Africa it's not yet that popular a first driver was after the Chinese started and some people had worked there they acquired the skills and spent after work some time to plant on their plots a second driver was the restructuring of these commercial farms so when the Chinese left and the state government took over a large share of the worker were laid off two thirds of the worker so they lost their main source of employment and they started using their skills rice to produce a cash crop and the same thing happened again in the 90s so the organizational change was actually a big technology push which was not planned but it happened in addition there's two other drivers population increase and relative price change the source of knowledge so I was asking the today's rice farmers so where do you know where did you learn from how to cultivate rice and parents and neighbors are the most important source and if we look at the history even initially neighbors were important but the parents are becoming more and more important it's a small sample but it shows some of the representative some of the shares the other so the work at the investment side also matters and especially if I asked people so you learned it from your uncle where did your uncle get it from often the uncle used to be a foreman at the farm and teach them so the spread from the knowledge spread is something we can see land values I mentioned already there's a vivid rental market now in Uganda so all the wetland is converted but farmers who don't have land can rent land from neighbors and they change the plots very often on seasonal basis and I was interested to find information how did that change in the past and when was actually the point that all the land was converted and it was difficult to get good data on these issues but I did some recall questions and what is happening today is actually that they pay 120 USD per acre per season so that's quite an amount the local users pay on the rental market that tells something about the probability of growing rice in that area and the importance for the livelihood actually much more than the investor pays this is the situation in Ethiopia at the top you see the construction work going on on this large scale farm and here you see the test field which might be a situation in a couple of years if they manage to really operate 10,000 hectares the model I did and that model is always a kind of an approximation I basically considered those two groups which are quite distinct along a number of features and I have another paper on that I'm not going into detail each of the group has seven livelihood activities so mixed livelihood approach and for each of the activities they have each of the activities creates an output which are then priced in monetary terms and requires a set of inputs and then I can say as the investment evolves those might change the shadow prices of the inputs might change and there's two constraints in the area which I motivate from observation and from literature there's a market constraint the area is very isolated at the beginning but as the investment arrives more demand for local produced services will increase and there's a labor market constraint it's a very remote part in Ethiopia there's little labor rental between the farmers and there's no other jobs than the civil service with a few jobs and one hospital I do... well I did a bit more scenarios but what I present now is three scenarios the base run which shows the mix of these activities for each of the groups then I take away 10,000 hectares of forest savanna land and see what is the impact on the local populations and the local income and per capita terms then a second scenario is the evolution of the big investment to 10,000 hectares size that is the plan and the investor is very ambitious to following up and then putting a lot of money there so he might succeed it's a question when it will be profitable but they are trying to get it done an alternative scenario which is a utopian inclusive policy measure so if the Ethiopian government would decide to have only given away half the area to a foreign investor but invest also in smallholder productivity so increase some extension service a little bit more infrastructure so basically lower the isolation and increase the productivity of smallholders the moment is quite low compared to what can be reached in this area and those are the results of these simulations so you see the indigenous group on the left and the settler group on the other side and those groups, what do I distinguish here the indigenous is a nylotic tribe group that lived in the area for more than a century and the settler group is also a group that have been moved there by the former regime in the 80s but they are using different technologies for example the settler group is using ox-plowing agriculture while the indigenous group is mostly relying on manual agriculture so this dotted line is the per capita income for both groups and we see that the investment will lift people out of poverty on average for both groups however the gains of the settler group are even higher I will show that in the next table and we see a kind of structural change that employment, or farm employment is increasing not so surprising I'm running out of time so I have to be quick those are the relative changes of each of the activities and you see agriculture activity goes down so the gathering, so the relative importance and absolute importance, it remains an important contribution and employment is the biggest push for total income and incomes go up by 50% if the investment really evolves to this mega size with a lot of jobs created what happens to shadow prices, that's one of the nice features of doing the simulation that you can say what are the constraining inputs and what happens to their appreciation by the local population so one finding is that farmland per se is not scarce in that area, it's the labor invested in the land which is the limiting factor people could convert more of the savanna land into fields but they just don't have big enough families to do so and shadow price for labor goes up as other opportunities arrive forest land remains relatively high with the indigenous group and it is lower already in the beginning but it even falls by 50% for the settler group and the settler group has a more structural shift compared to the indigenous group okay, to sum up the value for land might increase and might not it did in the case of Uganda there is more population pressure and there's already high intensity of use in the Ethiopian case land per se is not scarce labor has the potential to increase income in both cases, the off farm employment was a big part of the simulation part in Ethiopia and also of the stories in Uganda natural resources remain important in the Ethiopian case if they lose 10% of the forest land they were using it causes 4-5% loss of income for the indigenous group so the impact is quite significant it's compensated by off farm employment but it's still important and that is not considering general biodiversity gains of forest land technology diffusion does happen but not automatically so in the Uganda case the early phase is to be said that the Chinese were also focusing on transmit technology, the big investor nowadays doesn't and it mainly triggered through shocks when people were laid off institutions in both cases the property right regimes were not violated when it comes to the private land so the farm land people had but the communal land was taken away and you could see a change in the power balance that was made as an important one what can I say regarding to the title of the paper it's a relative positive picture to growth, so the growth in both micro settings has been stimulated by the investment mainly through the employment channel but also through business activities of the shops and everything inclusive, mixed and here it depends whether we go for what definition we take in both settings it was the better off parts who could gain better, who were earlier like those who had access to wetland could gain into rice and everything but also the land less had a chance to get jobs the role of small holders in the Egypt in case I did the simulation it has similar poverty effects if you would focus on small holders besides in the Ugandan case and I probably didn't mention that the laying off also led to that people who had worked together in the brick factory or in other units they formed farmer organizations so there was actually transforming of the groups and some of these groups are very innovative in the rice growing in the area so the role of small holders remains very important thanks for your attention and thanks to those who funded my research