an innovative economist a hyperactive one as well and mrs dr defloor is a professor at the mit she has been lecturing there for seven years and she studied at the economy parish she's director and founder of general poverty action and for one year she was at the college de france which is the one of the most prestigious institutions in france where in 2009 she inaugurated a she's the author of various books in particular two books that were published in italy as well and she just finished to publish or to write a book in english which has been published now in the u.s and which will be published next year in italy as well she received an award as the best youngest economist and in 2010 the prestigious medal john bates clerk well at the beginning esther duflo did not deal only with the economics she was more inclined to historical studies and working on soviet union she went to moscow during the perestroika period where she went to study the five-year plans and where she met daniel cohen and it was at that time that she decided to devote her activity to economics although in a wide sense in the wild sense of the mean of the world mr jiflo was included among the 100 most influential people of 2011 by time magazine but what is really important is this new original approach used by esther duflo in the area of the fight against poverty in all its forms health and microcredit education there is more than one billion people who still live in poverty in the world she has a very pragmatic approach which starts from tangible and simple facts without any ideological considerations with a view to looking for pragmatic solutions after experiments on the field esther duflo challenges preconceived ideas and answers with the pragmatism after listening to all the people involved so she uses applied economics she goes to the various countries on the field in order to investigate problems and to see whether solutions work and this after a scientific systematic assessment so this is the big contribution made by esther duflo that is the assessment of poverty plans based on experimental science and she will tell us about this and finally before giving her the flaw i would like to say that one of the characteristics is her dissemination inclination i spoke about her books earlier but she's also the founder of the magazine american journals of economics and among the various activities she for 10 years wrote on in liberacion a monthly article where she dealt pragmatically with the various fields the fight against aids in africa inclusion in the school of migrant children or there was also an article about the introduction of mobiles in kerala and how that had helped local fishermen redis reduce waste and increase their income so technology and its impact on the fight against poverty so the floor to you as the floor that's it for italian that was just a test um thank you very much for this uh kind introduction at least i think he was kind i couldn't really understand it the debate about about poverty in the world uh very quickly becomes a debate about aid quickly enough it becomes a debate about whether foreign aid can really eradicate poverty or not and that debate becomes itself a very polarized a very caricatural debate very quickly on the one hand you have jeffrey sachs saying that aid foreign aid on its own can eradicate poverty in fact when you wrote this book the end of poverty in 2005 he said that poverty could have been it could be eliminated by 2015 if rich countries agreed to spend enough on aid and in his view we already knew what to do we just had to spend the money it's a question of political will on the other hand we have other people like bill easterly who should have been here unfortunately couldn't be here and at this festival and don't be so moyo more recently claiming that in fact aid cannot eradicate poverty on the contrary it might be part of the problem rather than being the solution it creates rents over which people can fight it also prevents countries to develop their own solution that may be more adapted to their own problems in their view the only thing that can help countries and people get out of poverty is freedom by which they mean both political freedom but also economic freedom so this debate becomes a somewhat ideological debate which is related to the team of this festival which is how far can economic freedom go in eradicate poverty and how far could the government go in eradicate poverty unfortunately this debate is not very easy to resolve we don't have all that many countries if you look at africa africa got a lot of money over the years in foreign aid and it didn't become any richer in terms of per capita gdp so you might think well that shows that africa that aid in africa was a disaster but of course you have no idea what might have happened if africa didn't get aid they might have become richer or they might have become poor we just don't know so it's not a debate that we can really easily solve another problem with that debate is that it tends to stifle energy because it focuses it focuses attention on eradicating poverty like we have to do the whole thing and preferably very quickly and you know there is about a million a billion poor people in the world and they are that's a lot and so when people start thinking about billion people poor people in the world they're thinking this is an immense problem and what can i possibly do about it and that makes people scared and i'll give you a the results from an experiment that was conducted by paul slovik and his collaborators so they went to students in who were having their lunch at the cafeteria of a business school wharton and they showed them a picture of a girl a little bit like that and they said this is rokia she's seven-year-old she's at risk of starvation and dropping out of school and your donation to save the children that's an organization in the u.s can help her and then they went to another group of people who and showed them no picture but a list of numbers about million people starving in angola and zambia and i can ask will you give some money to save the children to help address this problem and the question was who is the better fundraiser it is little rock here here or is it disturbing million so we are going to take a vote so uh who everybody has to vote i'm not going to have no abstention who is the better fundraiser is it rokia is it the starving million there is too many abstentions you have to vote so anywhere okiya wins and you're absolutely right rokia raised 2.83 from the students and the starving minion raised a dollar sixteen so then they did another experiment they wanted to know whether if you made people aware of this bias that we have to give money to an identifiable victim as opposed to giving money to like the entire problem uh do you erase the bias so they again went to two groups of students and first of all they said you know what in our previous research we have noticed that people tend to give more money to an identifiable victim then when confronted to the whole problem and to another group of students they went and then they showed them either the rokia picture or the starving million statistics so now my question to you is does that erase the bias to give this little speech first so erase the bias let's take a vote not erase the bias too many people abstain i need more hands which way or the other so not there is the bias right so unfortunately you're wrong because he did erase the bus but in a somewhat counterproductive way which his people didn't start giving money to the starving million they just stopped giving money to okay so what is what i think this experiment illustrates is a fundamental characteristic of human beings which is our first instinct is to be generous but then we think again and we think well if i give a little bit of money it's a drop of in the bucket and the bucket probably leaks anywhere so there is really no point trying and so we just withdraw and start to stop thinking about it and that's what happened in this experiment when you make people aware that rookia is just one out of millions people feel well there's nothing you can do so what a lot of our work the group of the the group of people i'm working with uh is to try to go back to thinking about poverty not as one big problem made of starving million or a billion people living under a dollar a day but a series of identifiable concrete issues that we can start thinking about and resolve in other words not try to solve the problem of poverty one rookie at a time but try to solve the problem of poverty one problem at a time and if we start thinking about it that way then we have time is in our hand it doesn't really matter so what i want to do today is to give you a little bit of a flavor of how that works and how we approach the problem and what we learn along the way maybe things that are not particularly obvious to you that we found in in our in our field work the first thing that might not be completely obvious uh to you is that sometimes uh you can save money by paying people so a lot of organizations who help the the poor feel that it's very important to respect market mechanism in helping the poor so what they want is that the programs that they are doing can be financially sustainable that people are willing to pay for them so in fact when you're an economist particularly in economics trained in a little bit of public finance it's a strange idea because there are things that society wants more than the individuals and in that case we should be willing to subsidize the individual to undertake that particular behavior immunization is is a clear example of that if i get immunized that helps me but that helps you as well because you're also less likely to get the disease so not surprisingly the international community has been spending a lot of money subsidizing immunization so particularly the gavi alliance and the gates foundation and despite that something like 25 million children are not immunized every year that's what you call a last mile problem the money is there the infrastructure is there the vaccines are there and yet it doesn't happen so why doesn't it happen uh to find that out we started working in an area in rajasthan in northern india where about when we started one percent of children were fully immunized that is they had received the five shots that are essential for immunization and we started wondering why is that the case so one thing that is clear is that it's not because they don't care about their children all that they don't care about health because the same families who don't immunize their children spend about seven percent of their budget on health care which means that the kids are not immunized against measles but if they do get measles the family will spend an enormous amount of money trying to solve the problem so what you have is these empty village subcenters but crowded hospitals so a first problem might be a problem of information people are just not aware of the problem so this is not a problem that is only in in developing countries this is a clipping from the new york times 300 people diagnosed with mumps in suburban new york people refusing to take vaccinations either for religion reason but also some for example because they think that the measles vaccine is related with autism so information is always an issue but even once you're convinced you still need to go from intention to action and going from intention to action inundation straight that means going to this place that's a sub-center and what's important about this sub-center is that it is closed this particular one has been closed every single week we visited it for for a year on average the sub-centers are closed about half the time due to absenteeism of nurses so that's the first problem and you would think well you know people would easily get discouraged if every time they go to try and immunize their children the sub-center is closed a second problem is that even if the sub-center were open you could say well suppose i don't immunize my child today and i wait till tomorrow what is the chance that they get measles in the meantime it's very low so the cost of postponing for just a day is not that large and today we are very busy there is a lot of things we need to do and it's not very nice to bring a crying child home so you might think well i'm just going to wait till tomorrow and then tomorrow comes and you might think well i'm going to wait till day after tomorrow and never end up doing it so if this is a problem then you might think well one way to try and fix the problem is to give people a reason to act today and the way we did that is we worked in a in we set up an experiment very much like an experiment to test the impact of a drug but to test the impact of these policies so this is udaipur strike viewed from the sky we worked in about 120 villages and we picked randomly three groups of villages the first group of villages the white one we did not do anything left it as usual in the blue one an organization we worked with set up monthly immunization camps to make it easy for people and in the red villages on top of the monthly immunization camps people got a kilo of lentil for each immunization for the child so a kilo of lentil even in india it's not a lot of money so it is not something that is going to convince people if they feel that immunization is dangerous but if the problem is procrastination then it will help them it will because it will give them a reason to act today and not to wait uh till later and because the experiment was randomized at the beginning everything is pretty much the same after 18 months the immunization rose in the control villages to about six percent in the villages where the camp were put in place it went to 17 percent that's already an improvement but adding the lentil increased immunization by 38 percent so that's a huge improvement so the organization we work with asked us that's nice it's a huge improvement but can we really afford those lentils it makes the program not sustainable to give people this incentive to do something so we calculated the cost of immunization of immunizing a child in the camp villages and in the campus incentive and this is what we found it actually turns out to be cheaper to give lentils than not to give lentils so why is that the case well because once you set up a camp you're paying someone to immunize to immunize the children if she works harder because many more children are coming then her time is used more effectively so actually you end up saving money even though here i count the lentils as a cost whereas in fact you know as an economist i should count it as a transfer and therefore not even consider that it's the cost but even counting it as a cost it ends up being cheaper so this is important because it tells you a mistake that a lot of organizations and governments do which is to do the cost benefit analysis wrong sometimes it's actually the most efficient thing to help people about to do a behavior that you want them to do anywhere here is another one television is more important than food so where is that coming from well when i was uh growing up about when i was about 13 was the time of the live aid concert so i was thinking we are the world and my view of a poor person was someone who was an ethiopian child with a distended belly and in that i was very representative of a lot of people who think essentially someone who is poor is someone who doesn't have enough to eat and there is this very strong idea that there might be a poverty trap based on nutrition so the story the logic of the story is simple you you don't eat enough so you're not very productive so you don't make make much money so you don't eat enough so you're not very productive so you stay poor so this is the logic of this is super superficially very sensible but the only problem is that that doesn't appear to be the case because if it were true what we should see is that when poor people get any little extra money they should immediately try and buy more calories more of the same food in order to become more productive and try to escape poverty this way and what we find is in some sense exactly the opposite here is india india is becoming richer and richer over time people clearly like food there and as india is becoming richer there are fewer and fewer poor people fewer and fewer people live under a dollar a day so you would expect that there are fewer and fewer people who are undernourished and what you find this is researched by angus did and is exactly the opposite which is more and more people eat less than 2 400 calories a day in 1983 66 percent of people eat less than 2 400 calories a day which is a threshold for decent nutrition and by 2004 it was almost 80 percent yet people are not exactly complaining they are not reporting like a food they are less likely to report lack of food so the problem of nutrition in india exists people don't eat enough calories they also don't get enough micronutrients like iron but it's not a hunger problem it's a nutrition problem and it's not a problem of constraint it's a problem of choice and here is another experiment that makes this even clearer that the way we are thinking about food policy might be completely wrong this is an experiment from that rob jensen and nolan miller carried out in china and what they did is a program which is very much like the type of programs that we see to help the poor worldwide which is trying to make food cheaper so they went around and run an experiment where they distributed people vouchers for discount on the price of rice in a region where people mostly eat rice so when the rice when the price of rice went down by 10 percent people started eating less rice the rice consumption dropped by 2.3 percent so far the economy is here it means rice is a gift and good something which until this experiment most of us thought was probably an urban legend something that did not really exist but aside from this what is very important is that this is the this type of policies making food cheaper is the way that we're trying to get the proof to eat more in fact what is happening when you make the basic food cheaper is because rice is such an important part of the budget it makes people feel a bit richer then what do they do with the money well they're spending on things that they like which turns out not to be rice obviously which turns out to be shrimps to some extent and other things like cell phone minutes to some extent so that doesn't mean we shouldn't be worried unnecessarily about the nutrition issue for example in india but half the children are stunted that's very bad for their long-run productivity but it means we need to worry about it in a way that respect the fact that the poor are not machine they are not maximizing physical fitness or they are maximizing utility and we need to take into account those preferences when constituting policies which is very rarely the case and it's certainly not the case as far as food is concerned where what food policy means in most countries is dumping grains on people which seems to be counterproductive if this example tells us anything here is another one that i think is very relevant to the theme of this festival is this uh idea that there are very many many entrepreneurs in the developing world and the great virtue of the micro credit movement and to some extent of the social business movement which has been uh kind of supplanted it right now is the idea that oh the poor they are so entrepreneurial they can lift themselves after priority we don't need to give them anything we just need to give them access to credit and then they'll take themselves out of poverty so this is based on something which is quite real which is the poor are very often entrepreneur in the very specific sense that they do on their own business across 18 countries where we've collected data 50 percent of the urban poor operate a business and something like 20 of the rural poor also operate a non-farm business on top of that a large number of them have a farm if you compare this to the oecd number the number in the oecd is closer to 12 percent so is this really is this israeli telling you indeed the pros are or entrepreneur the one wrinkle is that even though they have run small businesses sometimes more than one that's not something that they seem to be particularly happy to do and one way you can see that is by asking them what they would like their kids to do and we've been asking this question in several countries in the world this is a question this is the answer for udaipur the same place where we did the vaccine experiment 18 of them would like their sons to be an employee in a private firm a little over 40 percent would like them to have a non-teaching government job and 34 of them would like to them to be a government teacher none of them said that they would like their sons to be an entrepreneur what these numbers tell you is what people aim for what people would really like and dream for for their children is stability it's the exact opposite of the life of non-stability that they have so then you might be wondering well if everybody wants to be a an employee why aren't there some firms that actually employ these people why are people why do people have to run their own businesses and the reason of course is that there are no jobs to be had so what people do with without the micro credit loan is to start a very small farm because they are in some sense buying themselves a job so how come you have all these tiny tiny firms and they are not managing to grow into larger forms that start hiring more people so what we think is happening is something like that which is this is a production function so the first the beginning of it is quite well known as a well-known shape it is diminishing return to capital which may be very high return at the beginning but it flattens out so imagine you start a shop you you clean up a room you buy some shelves and then you have nothing so if i start giving you a little bit of money to start to to get goods then you're going to have a high productivity of these goods eventually your shelves are going to be full however and then that's it if you wanted to become bigger you would need to start hiring people that would need to be complicated there is a next jump so if you don't have this next jump things flatten out of course if things flattened out so quickly after relatively low investment then you would have no large firms so we know there are large firms even in countries like india so what is happening is that there is another production function out there there is this qr production function which requires a much larger investment up front so for example instead of starting one shop in your in your room you kind of open a little retail store and that larger investment if you make it then the product there is more scope for improving the for for increasing the size of your business at a high level of productivity the problem is that if you're very poor no one is going to lend you enough money to make it worthwhile to choose this new better production technology so if you're very poor you're going to be stuck in the kind of businesses that really never really grow past some level and where we have a big thing we have a big missing middle over there where there is kind of lots and lots of very very small firms and very few of this larger firm that manage to actually hire people and and um get a more stable kind of employment for people this also tells us why the debate on micro credit is so fluid so the whole hope of micro credit was give a little bit of money to the poor and they'll progressively pull themselves out of poverty by their own bootstrap and we see how it wouldn't happen here they would get stuck on um even if you the microcredit was actually very nice for them very helpful it helps them get to m which is already something so it is not that micro credit is either useless or fantastic microcredit is very helpful but it's very helpful to go from o to m and it's not very helpful for most people to escape poverty and that debate is just never going on the debate is between a stuck between micro credit is easier the best things in sliced bread or a complete disaster which is the more the situation that more than the general position now unfortunately and what you can see from this graph how an invention like microcredit could have its use without being the way that you would have a lot of people escaping poverty to have a lot of people escaping poverty you need to fix the next problem which is how do we get sufficient amount of money into the hands of people who are capable to start a firm and that's a much harder problem and not one that has been solved so far here is another one that sort of the new fun the new fashionable thing now that micro credit is so 20th century is insurance and the idea of insurance seems pretty obvious on the one hand the poor suffer from tremendous amount of risk because they run their own business because they have a farm so imagine for example that you there is a drought then you're kind of you know this is a disaster or you become sick or you you have an accident then if your entire world chatters and the risk is not only costly once it hits it's also costly in anticipation these people over here they are farmer in kenya and they grow maize and what's remarkable about the maize they grow is that it's traditional maize not hybrid maize and the reason why they grow traditional reason not hybrid is made one of the reason is that hybrid maze is more risky and they feel like they can't afford this kind of risk so given that you would think well then the insurance companies should jump on this the poor need the poor need the the risk someone like could go and find them and sell them insurance and they would be happy and make money and they also would be better off so that's a place for the market to come in so it's not really happening so you might wonder why well the reason is that the demand for the products that the insurance company can offer is actually quite low and here is an example from an experiment that dean cannon chris udery conducted in ghana what they tried to do is to sell a very conventional weather insurance product basically if the rainfall at the nearby weather station falls below some threshold the policy pays you so the reason why it's an attractive product is because there is no reason to go to your farm and check anything so this is robust to mall hazard this is robust to adverse elections this is not very expensive to administer so that should be a product that is really a win-win well when it's offered for free so what they did is one of this experiment again when they go to see various people and offer it at various prices so when it is offered for free 100 of people want it but reassuring it would be a bit worrying why some of people didn't want it when it's free but as soon as you start charging people are less enthusiastic in particular the actually fair price is nine point five galaxy per acre that's what you need to charge to in expectation not lose money and at nine point five gallon cd per acre 40 percent of farmers by the policy in addition you have some administrative costs and you also need to uh make a little bit of money so the insurance company would like to start 14 would like to charge 14 ganacity per acre and if they charge 14 galaxy per acre eight percent of people want to buy the policy that is too few farmers to even bother so why is it that no one wants to buy a product that seems to make so much sense and the reason we think is precisely because the product is so simple the only product that the insurance companies are able to offer because of the problem of adverse election moral hazard and administrative and risk affords a very simple basic product so for whether it is this product parametric weather insurance for health it is catastrophic health insurance like hospitalization policy not outpatient care so from the point of view of the poor this is not a very attractive product because it is full of holes so imagine for example this farmer his field is completely dry he's still not going to get any payment if it turns out that in the weather station the the the well the rainfall was actually just above the threshold and since there are not that many weather station there is a lot of variation within what's happening in a weather station so from his point of view what we are trying to sell him is an additional lottery it's something which may which he has to pay for and may or may not pay him in the state where he want where he needs it of course if he was fully informed and rational he would see that it's a good lottery because it is a lottery that is more likely to pay when you need it but the way he sees it is look i already have this incredibly risky life and you're asking me to buy another lottery ticket you must be fallen on your head in addition there are problems of trust and when he's not paid he's always wondering whether it's because something fishy that did or something like that so as a result the demand for the product that the market is in a position to offer is actually quite limited so what do we do well one possibility is to say fine at the market price the demand is very limited the government could subsidize they don't need to go out and offer the insurance themselves but they could say we are going to subsidize the premium to some level at least for a while as people get used to this thing work here's the last thing i want to say which might be uh which is important to conclude which is the voice of the poor is often ignored but not for very strong reasons the reason why i want to conclude on this example is because when i present the type of work that i just presented today in a lot of cases people say yes that's fine that's interesting you're like an engineer more than an economist you take a problem and fix it that's nice but really you're missing the big picture the big picture is the politics if the politics is good policies will follow if the politics is bad policies will never follow in other words you can think about immunizing children but right now in cote d'ivoire the reason why children are not immunized has nothing to do with in procrastination or information it is because the whole country is in the middle of a civil war so at some level of course i don't disagree that politics is important but there are two things i want to say the first one is that politics is also politics doesn't happen in a vacuum politics is not the magic way in which the the will of the people transform or doesn't transform into a set of policy recommendations politics is also a set of rules and institutions which are often very detailed and the details of this institution institution how they operate on the ground makes a huge amount of difference i'll show you one example of that it's an example for brazil brazil is a pro-poor big pro poor democracy so we know that the big politics is fixed there so here is what you have to do if you want to vote in brazil first you have to vote because voting is mandatory so you go to the floating pool and this is your your ballot so for those who were in america some years ago this is a butterfly ballot uh there are two sides to it so to vote for your propheto it's not that hard you need to tick the right place in the right column to vote for your very adult you need to find a long list of names find out who you want to vote for and write down their names or their number without mistake so thomas fujiwara a graduate student at the university of british columbia studied what happened when this system was tr was replaced by an electronic voting system so this is what they did in order to make it easier to count the vote to introduce these little voting machines so now if you want to vote for someone you punch their numbers and a picture of the person you want to vote for appears now if you didn't want to vote for them you can change or if it's no one then it will come as a mistake you can also change as a result of this policy eleven percent more votes when it was introduced eleven percent more votes were valid it was first introduced in larger cities so you can compare the large city to the less large city 11 percent more voters more votes were valid who are these votes these voters whose vote was invalid before mainly the poor of course because they are the one who couldn't vote the result is that they elected different representatives representative with lower education level were elected this representative took different policy decisions this in particular they spend more on health care which is pro-poor in brazil because all the rich have their own private health insurance and that changes people's life in a very important way the birth weight improved among the poor so all of that to say that this is in a sense this is again another technical fix you would think now something maybe not very important compared to large things like democracy or the will of the people but this is how democracy works in practice democracy is not all that different from anything else the other thing i want to say and that's the other importance of politics is that it is true that politics matters but within a set of institutions the details of how people's general preferences are translated into laws and regulation makes a huge amount of difference a lot of the policies in the world that are meant to help the poor people fail not because of some big conspiration against them but because of what with my colleague abedit banerjee we call three eye problems ideology ignorance and inertia you have policies that are designed in ideology starting with some cartoon character view of what the poor are and what their problem is like for example they are starving or they are entrepreneurial or what have you which are sustaining ignorance of their conditions on the ground and persist out of sheer inertia so i'll give you just one example to clues of that this is india another big purple democracy we are we're interested in looking at the effect of the village education committees the village education committees have been introduced all over india under the inspired by the ideology that the poor need to be and want to and able to participate in the control of the education of the of of of schools in their in their villages that's fine that seems like a nice idea so every single village in india has a village education committee we went and went to those uh to ask to this um minister to the ministry to the to the bureaucrats actually how they decided who was going to be on the village education committee and they told us you have the parents of the best performing child in the school the parents of the lowest performing child in the school and the parent of the handicapped child and then we ask how do you know who is the best performing child in the school since you've just abolished exams and then big silence and which quickly shifted the conversation to something else so we went to the villages and asked people what they knew about those committees eight percent of parents knew that the community existed two percent of parents knew what they were supposed to do and the kicker perhaps is that 25 percent of the committee member had no idea that they were committee member so needless to say this is a policy that is that has no effect whatsoever and it is not because there is a conspiracy from the elite in the village to keep the poor out it's not none of that it is because the policy was never set alive was never never got a life on the ground so at some level this is a bit depressing to think that oh so many so much waste around at some level it's actually quite encouraging because if the problem is mainly that in a lot of cases we've not taken the trouble to figure out how things were in practice to try things out to evaluate them rigorously and to scrap them if they don't work then we can just start doing them and this is a message that in my view is a message of hope because it means we can do much much better in other words my very strong belief is that there are piles and piles of dollar bills on the ground which are coming from generic incompetence at all levels and that we can just try and do things a little bit better with not even much effort it's also a message of responsibility because it also means there is no hiding behind the problem is too difficult the problem is too big it's not a problem for me everyone can play their own little bit if people have money to give to an organization they can choose an organization whose programs have been evaluated and work if someone is a teacher in a school in india they can try to teach to the kids that they have in the school and not to some curriculum if someone is a bureaucrat they can also probably improve things so everyone at their level can sort of make a little bit of difference now and once you realize that then in some sense the whole debate whether we can eradicate poverty or not does not usually matter because the whole point is to get started the whole point is to get started to fix one problem at a time trying to identify first what the problem is not try to find a problem that fits to our solution but try to find a solution that fits to the problem try it out and if it doesn't try something else to move on thank you i thank very much esther dufu for this lesson this conference uh we have quite 15 min 20 minutes uh 30 minutes for the for the questions please i give you a video well i have a question for many years for many years the fight against poverty has been connected with the demographics the birth rate what is the importance of this parameter today that's a very very interesting uh very interesting very important question i think there are two aspects of this question is one to what extent is uh democracy and high natality rate a problem both for poverty but also for the environment and the other is suppose it is how do you why is it the case that people choose to have such large families or do they choose it or do they have the large families because they don't know better and what are their incentives and does it hurt them at an individual level i don't really know the answer to the first question i think in doubt it is probably worth thinking that it would probably be better if the population stop climbing so fast but i don't think we have huge amount of evidence but there are clearly pressures on on resources on natural resources such as water and climate on the second question what is striking is that at an individual level it is not that clear that having large families make people poor at an individual level it is also not clear that people have large families because they don't know better when you look at the details people seem to be quite smart about the number of children that they choose to have and in particular they have they have a lot of children in part because they need the labor in part because there is this risk that they don't know how many of them will survive and they don't know how many of them will work out and they are worried about what's going to happen in me to me and when i become old who is going to take care of me and i think the what this leads us to is to think well if high population rate is a problem the way to think about it is maybe the right answer is less a population policy a traditional population policy involving dumping a lot of contraceptives on people but more like an economic policy in sync in trying to change the incentives for people to align the private incentive a little bit better with the social incentive right now the private incentive is still to have many children and it could be different if there was other ways to ensure your your survival in old age can you hear me so apparently you are confirming what was said recently in a meeting in at the university of trent or some time ago well we too up to some time ago thought that it was enough to have a lot of aids and concentrate aids in a place in order to resolve the problem so don't you think that it should be important to have connections with the humanitarian organizations voluntary organization church organizations in order to bring cultural aid economic aid in order to guide these people and involve these people so i think that this is what i learned from you and do you think that we should have these uh networks of different organizations well i think you uh that's you're exactly right that the what the question that matters is not so much whether aid is good or bad but what works what program is effective and what program is not effective irrespective of who is paying for it if the aid is paying for it or an ngo is paying for it or the government is paying for it with their own budget what matters is that the policy is a good policy that's what is him that is it is what is important and you write that so in i i run a lab called the jamil poverty action lab where we work with a number of ngos to try and help them evaluate the impact of their programs and we are trying to you know extend the number of people we work with and i think that has proved quite effective and important well good evening and congratulations i work in a voluntary association i did the war the second world war so i'm pretty old well i say that in order to fight poverty we should involve governments de gasbori went to america to ask for help in order to resolve the poverty of italy and then what about health health is life life means work and work means income so health is a fundamental principle so even in africa aid represents only 5.7 percent of the budget of the governments so even and if you so even in africa it is a pretty small part of the entire story so you're absolutely right that governments need to be involved that the the fact of the matter is we have to design uh collectively the world has to design and find out good policies and this is how we're going to make progress and aid can help and i'm all in favor of aid and i think we should give more money once we know how to give it i think it's both a moral obligation from the rich countries and something which makes a lot of sense because that can maintain some equilibrium in the world so i think we could spend more money once we know what to spend it on and we should on the second part of your question which was health i couldn't agree more which is why i spoke about immunization health is in a sense a field of great promises and great frustration now the promise is that there are a number of things that we know work that doctors have shown us work like immunizing children sleeping in a bad net drinking clean water have tremendous effect on income of people in the future that's the promise part the frustration part is that why aren't those technologies taken advantage of more often so we are busy developing new vaccines and in the meantime in rajasthan one percent of kids get immunized so the the question becomes how do you change the the system and the health policies such as such that the majority of people can benefit from the the existing health technology and of course all the future one we're going to invent well i had the opportunity of visiting ethiopia and i met masses of young people many young people so my question is these tens of millions of young people so do we have hope that they have some life prospect and i'm referring to a life which is worth living i'm asking that because we in the west go for excessive consumerism made of soups and mercedes and rolls royces and we consume so many resources that we steal from them so the question is do you think that these millions of young people really can expect a decent life given the situation that we have the short answer is yes the longer answer is per se having a lot of young people can be an asset i mean you can call it a democratic you know they call it sometimes the demographic dividend if you compare india and china now china has grown more historically but now china is aging because they were so effective in reducing fertility and india is full of people who are in the age of 14 20 but also 25 and 30 who are the one who are going to be productive in the future so the big question becomes can you find a way to for these people to do something helpful and i think a key key part of that is going to be education and one of the problems of developing countries today is that although a lot of children go to school and the school systems don't deliver very much to them the quality of education is low about half of the kids finish primary school and cannot recognize a simple paragraph and that of course limits a lot their chance to to make a living and so i think a key issue that is in front of poor countries now is what to do in particular with respect as far as the youth population is concerned is how to improve quite quickly the quality of education and i think the good news is that we have a little bit of a sense of how to do that particularly at the primary level the less good news is that that's not really implemented in any large scale and therefore there's some reforms that needs to be made in a lot of countries and also that what's going to happen to secondary school is a little bit more up in the air which is as these kids become adolescents and will go to secondary school it's a little bit harder to have good quality secondary education you need more you need more educated teachers and these teachers presumably have other things to do with their time and so i think a big question mark and a very interesting problem uh to to think about is how can you do secondary education for a very large number of people maintaining the quality while keeping the budget of secondary school in czech that's for a country like ethiopia but also like india is a i think one of the challenges ahead prague your research showed that uh the entrepreneurship is not a solution itself for the poverty uh actually i'm surprised that 50 from the urban poor's are entrepreneurs but i suppose that are uh scientific data uh what do you think about the social entrepreneurship regarding the reduction of the poverty i'm asking you because this region trentino is known that in history in the end of 19th century resolved somehow this problem solved somehow this problem of poverty by appealing at this solution of social economy and it's well known as like one of the largest uh developed cooperative sector i suppose in the world maybe i i suppose i i would i would need to understand i haven't fully understood what social entrepreneurship means because it means different things to different people so i'm going to answer the question i hope it's your question but i'm not entirely sure the the idea of social entrepreneurship is uh is the idea that you can have so to speak a double bottom line so you can make money or at least not lose money and at the same time do something that is actually helpful that has a that that that is helpful for for poor people so for example you could start a little factory that produces nutritional supplements and then sell the nutritional supplement to the people so kids would be better nourished and at the same time would make some money so i think that's a that's a solution which is extremely attractive and like everything that is extremely attractive it has a danger of being almost like too attractive which is i think there are cases where in it is true that the market is missing and someone could come up and try and and and in and work in this market and fix the missing you know plug the missing market by uh figuring out some new system and i think micro credit is a beautiful example of that where most micro credit organizations make money or at least don't lose money and at the same time they deliver something which i'm still convinced has value on the other hand the danger with the social entrepreneurship movement the way that you hear about it today is sometimes you have a sense that they think that they can solve every single problem in this way and they are forgetting the problem of demand that if you're trying to provide something for which the demand is not there you're not going to make money and there are things that society wants to do should be done from for so from a social welfare perspective but that indeed the individual demand is lower so for example no one i can predict that no one would make money trying to immunize people and yet we need to immunize people so social entrepreneurship is not about to fix the problem of immunization things that are maybe less obvious is clean water you would think well could i make money trying to sell water filter to people well at the moment you would not make money trying to sell water filter to people you would lose money because it it cost some amount to produce the water filter and the demand for quality of water for some reason is very low people are not willing to pay a lot for having clean water so you see that so you think why aren't they not willing to pay a lot that's a complicated story that i'm not going to get into right now but that's the fact of the matter so does it mean that we should not worry about providing clean water of course it's not it doesn't mean that but it means that it will need to be subsidized then of course it can still be uh provided by a social entrepreneur but they will need to be in society a contribution of society to make that happen so i think one need to be creative which is one need to say well if there is a missing market and there is a market a solution to that problem by all means we should put it in place but we shouldn't leave under the illusion that every single problem is uh it can be solved in this way thank you sir it was really great lesson today i understand that you are proposing a more scientific approach to policy you know trial renderers and thinking about which policies work and which don't work and scrap those that don't work which seems an important you know message to give on the other hand i'm a little bit worried about whether there are general lessons that we can sort of learn in the sense that there is a telogenate in the plants there are there is a telogenate in the policies but there are there is heterogeneity also in the population in the places social networks are different and different fragmentation languages costumes beliefs and so are this will make trying everything on everybody you know in order of magnitude more difficult so from your experience is it enough to see that i don't know that is particular well today we talked about failures for the most but suppose there was some successful story somewhere can we learn that that is going to work also in some number of other countries among other realities or we need to go one by one that's my question so at some level it is it can't be the case that it that every single person is so specific that they need to have their own little set of policies on their own or every single village otherwise the whole concept of policy would go away so as long as you acknowledge that policy is a sensible world then you acknowledge that the heterogeneity is manageable that i can learn something about you that is relevant to your neighbor now of course there are limits to that uh and for example it seems intuitive that if we give school uniforms in kenya that will and and they do help kids stay in school longer that might that policy might not work in norway so we know that there are limits to what generalizes what doesn't in program evaluation as in policy so there are two ways to to go about this problem or this this this issue one is um replicating experience so in a lot of cases uh once we start having confidence that something works it's because something has been tried in in some context it's sufficiently different that it makes sense to try the the thing then the second part of the answer is that the experiment is never sufficient on its own the experiment is only always a data point in some kind of a theory that you have which can be explicit or which can be implicit uh often in academic journal is explicit and policy discourse it seems it's implicit and in a sense that's why we we for example with a visit uh um uh in the book poor economics it just came out we start to we try to go uh to go one step further from just evaluating from putting things next to each other or something in in this in my previous book and to say well think of each experiment as one data point or one touch of paint and not only experiments actually also like simple you know descriptive statistics uh like means and stuff like that they progressively um amount to some kind of a painting which gives us an idea of what is the structure of the problem and things like that and it is that painting that gives you a sense of what is that gives you a sense that makes that makes you able to predict what is likely to work as a function of the features of the context so the experiment on its own is just a data point it's not because the sun rises every morning to the east that you're sure that it's going to work again next tomorrow once you know that for a couple of times you form a theory for what from what's happening with the movement of the cell and earth that's the same thing in social policy but it helps and also once you form predictions then you can test them by running new experiments you get uh some feedback on what your theory is so you have this kind of constant process of induction in the deduction sometimes the data is a bit ahead of the theory and sometimes the theory is a bit ahead of the data but they keep kind of talking to each other a last brief question i'm sorry but it is so late i have two very brief questions the first one is related on insurance when you talked about risks correct me if i'm wrong but you suggested that governments should subsidize in order to deal with the premiums now my question is uh we've all we all know that subsidies often are are criticized because they create some sort of dependency so how do you evaluate this project as becoming sustainable if they continue receiving subsidies from governments and the other issue is regarding insurance i don't know if in your experiments are able to find anything on informal insurance mechanisms in these developing countries thank you so far the first question two answers one is that i think as i said at the beginning i think there are things that we should be willing to subsidies forever and ever and ever and they are things that we are subsidizing forever and ever and ever in immunization for example is free and compulsory in this country we are very rich already people could afford the cost of immunization we consider that immunization should be subsidized because the social benefits is larger than the private benefits and i think it is very dangerous this idea that everything needs to pay for itself we we are not look this is if we're thinking of policy we're seeing from the perspective of the government we are not like a person trying to manage their budget we are thinking about the whole social welfare and things don't have to pay for for themselves one by one by one what needs to be sustainable is the whole society so that's one so if it turns out that health that insurance like health insurance for example needs to be subsidized until kingdom comes then let's let it be because hence is something which is whether it's contagion something where you know if someone is very very sick and goes to a hospital that cost us a lot of money so you know what's the problem with subsidizing so that's answer number one and so number two for the particular case of insurance i do think that there is a process that needs to happen which is i don't know it for a fact but i i think that people could learn a little bit better how insurance works what it does what it doesn't do it's not an intuitive concept at all and therefore if you started by subsidizing it it is possible that people valuation of the insurance product would increase and therefore you would be able to reduce the subsidy you know people would learn from their neighbors people would learn from their own news etc and you could experiment with it very simply by subsidizing insurance very heavily in some places not some other and then see what happens over time and an experiment very much like that was done for subsidizing bad nets um we showed those those effects where that over time you can reduce the subsidy more if you started with a very heavy subsidy because people have learned uh the use of bed nets so i could envision something where it's not that people will get used to handouts is that they will get used to insurance which would be good um social about informal insurance yes there is a tremendous amount of work on social and informal insurance very interesting very important um it's it's necessarily limited because it's uh geographically limited in general so it's to start with can't deal with aggregate sharks and it's also limited in its scope because it's usually made of one of bilateral relationship one-to-one relationship and so people are not able to handle a huge shock so what we see for example is people are much better insured by their friends for small things and not very well for like a health catastrophe so the informal insurance is better at dealing with the kind of day-to-day problems and not very well at dealing at catastrophic situations where you would want to have some formal possibly subsidized insurance well thank you very much i would like really to thank esther duflo for coming to train to today it was a pleasure having you here and thank you for taking so many questions