 I entitled my presentation Thinking Wider on Land Institutions in Africa and I think the challenge that we are facing and Alain already raised that very eloquently is that in 1990 to 2010 we saw a lot of African countries passing new laws and policies but the institutions to implement them have been lagging behind. I think the main legal innovation was the recognition of customary tenure but there was little change in terms of practice due to lack of clarity who is actually subject of these rights. Is it males, is it only females, is it the chiefs, the headmen or whoever and ways to exercise these rights and to transfer them. Also how they can be recorded and of course without having boundaries land rights are a little bit difficult to define. You have a lot of disputes possibly and the responsibility for maintenance and also access for arrangements for accessing land data by the private sector. This means that these innovations together have not really been enough for structural transformation and that's what has been talked about here quite a bit and that requires both long-term leasing and participation in value chains, valuation of land and so that it's possible to use it for collateral land acquisition for large infrastructure projects. I think in the World Bank we are having this regularly being a key bottleneck for projects and of course property taxation and levying of local revenue to provide services. Therefore we need to expand the horizon a little bit to focus both on formal and informal institutions both in urban and rural areas, use the new opportunities that are provided by IT remote sensing and connectivity and build the domestic capacity for policy dialogue and change. To put Africa in global comparison we put here an index of just four elements whether the disaster is digital because I think it's very difficult to plan if you have only paper maps and these paper maps and of course paper titles also can very easily disappear whether registry and disaster is integrated so that you can actually have the spatial information and the textual information on the same property in one go and whether most of the land and don't ask me about the definition of that in the main city is either registered or mapped. So these are four points countries can have between one and four points and you see here that Africa with the exception of Rwanda which is a green spot that is because of the size of the country hardly recognizable is quite red and doesn't meet any of them together with South Asia. So clearly there's a lot of things to do and what I want to do in my presentation is to draw out some implications for how to deal with customary tenure, how to formulate policies for large farms and I think there was a little bit of discussion on spill overs already yesterday whether it is possible and how it's possible to reap sustained benefits from land titling and what this implies for urban land tenure and competitiveness as well as policy and research. I think on customary land issues it is very clear that the colonial dualism that was introduced decades or even centuries ago has not been overcome by the legal innovations that have been introduced. Originally it was a myth of retarded small holders versus commercial farms that need to be promoted. Nowadays it is high entry barriers to statutory title both in terms of survey cost and in terms of inefficient institutions and that is often reinforced even in countries whereby non-transferable documents or documents that are not being maintained. This has two types of disadvantages. First of all it encourages speculative land acquisition by the wealthy and I think Tom Chain has been doing a lot in terms of documenting the increasing concentration of farms of land sizes in Sub-Saharan Africa and of course since all of the cost is upfront there's no incentive to use the title land effectively so much of it is speculative land acquisition. I'll show you a little bit of that later and of course the second element is that the benefits of customary tenure are largely foregone and the majority of the population who live under customer tenure especially women are left with very very little security. Sambia is one example, title means or acquiring title means that land is irreversibly transferred out of customary to state land which means that essentially the chiefs sell their consent of course they need to give their consent on paper. The cost is very high even if it's officially without paying off the chiefs and of course what you see then is that informal documentation emerges a substitute. What does this mean in practice? We did a small survey there of emergent farmers and of course not surprisingly and I think political economy is one of the currents that really goes through the session. It's the guys who are politically connected and often the chiefs who acquire land outside their kingdom in other chiefdom areas who are the ones who get formal title because they're politically well connected and informal certificates which they call chief certificates are not very equitable either but if we look at the productivity impacts we find the title even though there is huge costs associated with it has no productivity effects at all which is consistent with speculative land acquisition informal certificates are associated with significant increases in yields as well as profits. Of course that means that one desirable step really would be to see whether you can actually do away with this artificial dichotomy of customary earths versus statutory and document customary rights and I think Mexico shows a very interesting example for that. Elisabeth will talk in much more but what would be required to do that would be to define the status of the chiefs area demarcate and publish them to support basic resource inventory and planning put in place the regulations and publicize the internal and outside external transaction and then of course provide the basis for maintenance and the link to a national registry. Technologically I think all of these things can be solved they're not very difficult but of course the political economy means that there's a lot of people who gain rents from the existing system who are opposed to that. The second item that I want to touch briefly is on improving data on large farms and policy because I think as I already mentioned this is very important because there's a significant shift to expansion of large farms. Malawi provides an interesting example because 75 percent of the area is cultivated by small holders 25 percent by estates they have average 35 hectares were given in the late 80s but there's virtually no estates. What we did is we started by computerizing the registry then had to do a listing because the data was such crap that we had to actually see where things were located document the boundaries update these and add key characteristics and then use the sample of 2250 and divide it into as is seen here on the slide to complement the LSMS. What we find is that clearly sample composition matters quite a bit if you take only the ones so I think this one is the small holders this one is if you're at the estate so you clearly see that there is some shift in the so that the inverse relationship actually changes a little bit but even the estates are not homogeneous you have a significant difference between corporate estates and household based estates and I think these are tenants that are associated with estates and if you do this with profits I think you also see significant differences so clearly I think there is this raises a number of policy questions first of all the figures I showed you there are based on cultivated land the figures change significantly if you own land so I think clearly we need to ask why do estates not cultivate all of their land or why don't they do mechanize and does it the insecurity of tenure preclude them from leasing out the rest of their land to small holders second is that government loses significant revenue I think we did some simulations that could actually add up to five percent of public spending because 90 percent of the leases expired lease payments have never been adjusted and are rarely collected and interestingly enough as part of our policy intervention on the development policy loan we asked them to issue 30 000 bills and lease renewal notices we thought that would never happen but it actually happened and what they found is to their great surprise is that there's significant interest to buy these lease holders to renew that and of course that now creates an opportunity for doing very interesting policy experiments and seeing how this can actually play out and what are the conditions and the willingness of these leaseholders to pay and of course then the third question is what to do with defund estates land should that revert back to customary or should it be distributed to progressive investors and can it be used to generate spill overs and I think that will become relevant especially after the next elections then I think I think I have to skip over that because Alan will kill me otherwise but just to illustrate I think the Malawi case was of course all unmechanized and I was very surprised myself not to see any mechanization you see in mechanized cases which is Ethiopia here is much more significant differences between small holders which are illustrated here and estates which are large farms which are also significantly larger this one is for mace but I think you can do this for almost every crop and again I think it illustrates the importance of getting large farm data to inform policy and of course also it raises the question in the case of Ethiopia why we don't see so why we do not see any spill overs and of course that is a very important both either in terms of employment or in terms of technology transfer and that clearly is something where I think policy and some coordination and planning could be quite useful which can be combined and we hope to be able to combine that in the case of Malawi with recognition of customer tenure which I think Mexico provides a good example then of course let's go for the next item is to look a little bit at the weather tenure regularization can actually have any impact and the bright spot that I showed you on the map earlier provides a good example Rwanda implemented as many of you may know the largest program in Africa that regularized about 11.5 million parcels 64 percent of them agricultural and interestingly enough 86 percent have women as a sole or co-owner so I think that is unprecedented the neighbor Uganda has 23 percent I think this was interestingly enough that was preceded by extensive pilots and it's very low cost it's five to six dollars per parcel so I think nobody can tell me that that is not affordable we looked at the pilot impacts from the pilot I think we see significant investment effects but also some issues on female access that need to be addressed in the rollout for the national rollout where we did a randomized where we were able to actually argue with the government to randomize the rollout we see that perceived rights are much more secure we see higher levels of market participation lower labor for land preparation but not harvest which is stronger if the initial rights were weaker increase in wage employment and food security and I think all of these of course are consistent with structural transformation and of course if we complement the household data with administrative data which is on mortgages both the number and the amounts you see that there was actually significant expansion and that is not only limited to urban areas so how can of course the question that comes up everywhere is how can this formality be sustained and I think because informality I mean it's no use if you just regularize and then it goes back to informality I don't think this is an issue for urban areas we see in Kigali six percent of parcels are or sales for six percent of parcels of the national total are registered every year but of course we see a big return to informality in rural areas what can be done about that or the government of course was concerned as well so one step was to provide information and access to service points which they call sector land managers using administrative data we can show that placing a sector and the rollout of overtime of the sector land managers we can show that placing a sector land manager doubles registration rates for sales but has no effect on non-monetary transfers of course for more non-monetary transfers which is inheritance and gifts and others and it's very easy to link the personal and the land registration since ronda has now achieved a full a full online connectivity to all the sectors and the fully computerized personal registration system this is something that experimenting with and that would routinely then reflect changes in individuals personal status in the land registry but of course the big elephant in the room relates to fees I think currently ronda has a flat fee of 40 per parcel which was incidentally recommended to them by the world bank and that of course makes sense in urban areas but not in rural ones where the value of the land may be actually less if you have a very small parcel and so I think that is something that the government is changing actually we are doing an experiment right now where we have a survey in the field to document the level of informality and then also see what is the elasticity to fee changes of course one other important element is that rather than look to look at the benefits from from registration and registering subsequent transaction of course that means that you have to enhance interoperability establish online links to courts and banks regulate and publish valuations to also set prices for expropriation a link to building permits make the lands more say a land registry more secure and all of the wonderful things that are associated with the multi-purpose disaster just to give you an example here I think just the incremental revenue in kigali alone so only the capital in terms of taxation the government could create is about 5.9 million dollars per year which would in 10 years more than pay for the cost of the program then I think the last thing that I want to touch on is the conundrum of urban land tenure I think clearly urban land rights are essential for african's development urban I think what we see in much of africa is urbanization without growth which has been driven by resource rents and imports without actually providing a huge land appreciation has gone into private pockets instead of service delivery and public goods there is without land rights you cannot plan you cannot value the land of course banks cannot use it as collateral and of course that will increase the cost or will will drive up wages make and make cities uncompetitive what is surprising is that the demand for title even among low income households is actually very high we did an experiment in daris alarm where we actually did some surveys we reduced the cost from the official fee we did a lottery with vouchers and we found that an average willingness to pay off about 35 dollars which if you compare it with the cost of the program in randa means that you could easily do that interestingly enough this could also have a very strong gender effect we did we tested that by making some vouchers conditional on putting on you get the money only if you put a woman on the title and we worked with the woman's NGO who were very active in pushing female rights interestingly what we find is an impact is that or that compared to 25 or 4 percent who wanted to put a woman on the title in the baseline 89 percent actually did so when they applied for the title and actually made down payments for that but interestingly enough the monetary incentive had no impact it was all the awareness raising but of course there is very strong political invested interests opposing these titles titling I think we never were able to issue any title because it turned up because of all kinds of political reasons that were brought up so we actually had to return the down payments that people made and they were not very happy about that I can tell you so what mean what does that mean in terms of taking moving forward on policy clearly I think we need to go beyond the ministers of lands to deal with ministers of finance and local government I think in many cases and some be a good example title is needed to collect property tax the only they collect only 10 percent even on the title land but of course the local governments has very very little coverage so I think they have right now about 63,000 of 400,000 properties for Lusaka on title and they collect about 50 percent so I think they are our best allies in terms of actually moving that and interestingly enough by a quirk of the legislation in some via local governments are actually allowed to issue titles so we are working now with local governments to issue titles in different cities and create a lot of demand which I think is the only way of getting them people in the ministry of lands moving and actually getting them to wake up of course a second element is to involve the private sector I think India has been experimenting quite a bit with PPP's build operate transfers which I think are very well suited for land administration because of course the this rapid expansion and project mode that is possible in Rwanda where you have a president who will monitor every day what his people are doing but in many in very few other African countries so clearly that is something that the private sector is much better and local government should in principle be able to issue bonds to finance that so what does that mean in terms of policy recommendations I think there is a clear sequence and where of course new technology can help us a lot which would involve first of all computerizing existing data to be able to diagnose the data quality see what's actually going on some of the things that are illustrated in terms of large farm performance highlight the option the gaps and the options to close them and also I think that means you shouldn't underestimate the scope the corruption in land ministries and of course if you computerize things at least you have the possibility to eliminate all of these because you have an auditable trail of that means at least you can implement the security measures much much better than a paper based system second is to adjust regulations and institutional arrangements I think there is I mentioned a couple of examples but clearly means that needs to be very country specific and of course and then expand coverage and interoperability to actually harness the full benefits from a functional functioning land institution I think the global global community has a very important role to play in three aspects one is of course I mentioned already policy conditions I wasn't a great believer in such policy conditionality until I found out that in a couple of cases actually it worked quite well and of course I think it works because it can actually help some progressive parts in government to push for changes that are needed in any case I don't think we should overestimate our power there second is to have simple indicators and of course SDG indicator 1.4.2 and 5.8.1 and linking to some of these elements that I showed you earlier from the doing business is very important to raise the bar globally and create competition and of course there is need to build capacity for domestic policy analysis which is why we are working with AARC on PhD and master's course in terms of the research agenda of course I think all of what I told you is at three thirty thousand feet so you need to have local evidence and research to put this down to earth and deal with the enormous complexity and variability that is invariably associated with if you deal with land and I think that is very important because you don't have a second chance on land if you get things wrong and screw it up it's screwed up for a long time and you will never be able to fix that and of course people's lives are at stake there so clearly adopting that and then checking the impact is very important I think there is a lot of policy and institutional changes that provide very interesting experiments and of course with increased access to administrative data and remote sensing there is really immense opportunities that haven't been available 20 years or 10 years ago when Alan and Betty did their book so I would hope that combining spatial data satellite imagery and farm household surveys as well as the data on urban tax collection would provide enough material for another attempt by Wider to actually tackle these important issues and provide some policy advice that is urgently needed to improve tenure security for the poor and better functioning of the economies there thank you