 Hi, my name is Rachel Gazelkwist and I'm a research fellow here at UNU Wider. I was wondering if you could tell us a bit more about some of the policy lessons that are emerging from the analysis of the global MPI. In particular, for instance, if you think about changes, the countries that have changed most rapidly in the global MPI measure over time, is there anything that jumps out in terms of the sort of policy steps they've taken? Or comparing, for instance, the global MPI with income-based poverty measures where there's the greatest gap, is there anything that comes out in terms of policy there? Thank you. Okay. Yes, here? My name is Pascal Doe. I am the author of a PhD dissertation, The Role of Higher Education in Poverty Reduction at the University of Tampere. And within this research, I use two contradictory titles. One was Why Poor People Remain Poor by Ng, and another one was Why the Rich Countries Became Rich by Reinhardt, published in 2007. I was going to find out if the MPIs are cotominous to mitigation of poverty. And if it is, I would like to know how you situate the role of higher education because when I wrote the thesis, not only education and higher education but science, when I wrote this thesis, I argued that science, higher education was very central, even in addressing the 8M digits. And if you side with me, then I would like to ask another question. Would I have the impression that when I see the example of Africa, that the MPIs undermine the power side of the game, political poverty? What do you think? Thank you. Okay, we'll go here in the front. Hello, I'm Carlos Grading. I'm a research fellow here at Unowider. And I would like you to stand a little bit on the added value of using an aggregate multi-dimensional index, complementing the well-established consumption poverty measure used by the World Bank. I have no doubt that we are adding something with this new index, but I still wonder what is exactly what we are adding. Because if these dimensions were highly correlated with consumption, we would get basically the same measure, maybe with a different level of the poverty line, higher or lower, but the added value would be small. We always should be looking at different dimensions separately, but the fact of aggregating all dimensions in one index. So it is the lack of correlation between consumption and this aggregate indicator where we find the added value. And I wonder what do you think is the main source. So I could think in terms of measurement error, maybe consumption is not a good measure of well-being, because it's more difficult to report than years of schooling or having a dead child, etc., which I think they are easier to report in serving. Maybe it's because in some areas income doesn't buy better education or better health because you need some infrastructure, you need schools, you need healthcare centers, or maybe it's because some households could achieve their consumption above the poverty line, making sacrifices in their health or in their education, because, for example, using child labor or working longer hours. So what type of households are we adding to the global poverty picture that we would miss using the consumption poverty measure only? Thank you. So three quite challenging issues, policy relevance, education, Africa, and the value added of an aggregate. The mismatch between income and multi-dimensional poverty at the household level is quite large. For example, in Bhutan where 12% of people were monetary poor, consumption poor, 12.6 were multi-dimensionally poor, 3.2% were both. So overall 21% of people were poor, but only 3.2 had both types of poverty. And if you look in China, if you look in Chile, if you look in different countries where the poverty measures come from the same survey, you can look at that mismatch and it's consistently high. Why? We don't know. That requires qualitative work. As you said, some of it will be the volatility of the consumption poverty measure, some of it may be household spending patterns or burdens of disability and dependency. But we also need to know more because these are households maybe where the service provision is similar. Another cause of mismatch would be service provision being quite different in one area than another. And so you do see places where there's plenty of service provision but perhaps not the livelihoods. And so the MPI levels would be low. Or vice versa in the region Gaza since we're on Bhutan or Papua New Guinea in Indonesia, it's the same. I'm sorry, yeah, the monetary poverty is very low but multi-dimensional poverty is very high because there are no services but there is for various reasons in the two regions income. So I think they're capturing different kinds of deprivations. Now what does an aggregate add to a dashboard? And I didn't show the slides on that but if you were to look across the 10 indicators of the global MPI then you would find that 75% of the population were deprived in one of them, 3.9 billion. And so it's quite capacious. It would identify a lot of people if you said if you had any of these deprivations you were poor. So a feature of a multi-dimensional poverty aggregate is that in a sense it focuses in on people who have multiple deprivations simultaneously and so your resources, your fiscal resources which are limited are going to be contained a little bit to the people deprived in a third or half or whatever the poverty cutoff is at the same time. And so there's plenty of empirical work showing what the difference between just having a dashboard and where any deprivation reflects is it equal in importance in a sense and having an aggregate multi-dimensional measure. I don't know if that helps. And to Dr. Pasquale in Higher Education. So you asked if reducing MPI was co-tournamentist with ending poverty and clearly it's only partial. So like the target said poverty in all its forms and dimensions and all its dimensions would not be possible to measure. So an MPI only tracks the indicators that it contains. So for example if it doesn't contain livelihoods or if it doesn't contain violence and those are important deprivations then zero poverty could be coincident with high deprivations in those areas. I think your question about higher education was a different one and it was really saying in an environment in which multi-dimensional poverty is going down perhaps for people without higher education what is the role of higher education and an educated population in driving that change and that's a much more interesting question but a complex one so it's not one that we've looked at in terms of looking at the average level of higher education prevalence and seeing if that coincided with a faster reduction in MPI for the other portions of the population but I think it would be interesting to explore. And to Rachel clearly that's a very important question and a very ambitious question. We don't attempt to have cracked it. In a few regression analysis we've looked at a few determinants and what's very clear is that however you measure growth elasticity the elasticity of growth and multi-dimensional poverty are very low so we don't have the same relationship you get with income poverty. It's not a surprise back from Bourguignon and others at the World Bank showing the correlations are different and so the drivers would tend to be more associated with proactive social policies and public expenditures as well as activities by other stakeholders. It's also interestingly not necessarily good governance at least not in low-income countries so we did multi-level analysis my colleagues did it didn't find it and also if you think of Nepal which actually reduced its MPI from 26 to 2011 the fastest it didn't have a government for part of that time but somehow they were able to sustain the social expenditures and also they had high remittances and high backflow of educated labor into the country. So it's a very complex story I think it's country by country we haven't done enough to be able to generalize yet but again our data are online and so I'm hoping that others might take them and do the findings we haven't been able to. Okay. Thanks for this we'll take the next round. Yes here in this middle. A little bit in the front. Hi this is Jonas Uotinen from the University of Turku. I'm a PhD student at the economics department. I wonder is there a way for the MPIs or a modification of it with a possible addition to it to also inform the rich countries on their aims as it seems there are possibly a bit lost if we kind of try to move beyond the gross domestic product then it's like okay what shall we aim then for? And this relates to also the what is still missing from the capability theory. There was Barry Schwartz for example in 2000 in his article who pointed out that it seems that deducing from the like or he stated that the increased freedoms in United States as proxied by the increase of purchasing power over the 20th century has come has led to greater anxieties and other mental disorders instead of it being the contrary. So it could there be some kind of addition that could lead us towards understanding of these issues. Some of the things that come to mind is perhaps some Aristotelian ideas of realisation of the human potentials and what those be as well as the developmental trajectory of the human according to the western developmental psychologists and thinkers. So these are some of the things that come to my mind that possibly could be added or something like this. I would like to hear your opinions on this. Thank you. There's a hand in the back. Thank you. I'm Jim Sinyan, researcher, University of Helsinki. You admit that poverty is a very fluid concept and my question is simple. What's your definition of poverty and who is poor? Okay. Then we take one in the far back again. Hi, I'm Pia. I'm a research fellow at Wider. I have a two part question and I would appreciate your thoughts first in your researcher boots and then in the policymaker boots if you can imagine to Trump. The first one is regarding the SDGs and implementation of the monitoring. An OECD report last year tried to put, to work as many of these indicators as possible and even the most developed and let's say data rich OECD countries would not have most or at least I think more than 80% of these indicators ready lying around. My question to you is for the developing countries where these gaps are even larger, what would be the indicators you would focus on in data collection? Where do you see the biggest gaps? The second part of my question relates to limited resources in the sense of where do we get more data from? There's also a big push, also championed by the World Bank for identification for development. Now that has great potential for example linking and composite indices of course the different data sources but it's also very big exercise. Usually it is argued from the side of we can more easily distribute benefits and we can more easily monitor. Now for your exercise where do you see the value added or would you say actually you would prefer that more modules would be hooked on the already existing surveys that you work with? Okay, another three. Relevant to rich countries, definition of poverty, major data gaps and where do you hear, who gone? Yes, very good. So Johannes, in terms of the rich countries certainly there's nothing that stops them also from having a multi-dimensional poverty measure. We use silk as comparable across countries and we've been involved in that Silk II and are involved now in that Silk III exploring those options. But your question was wider than that and it might be relevant to mention the case of the Gross National Happiness Index in Bhutan which is a nine-dimensional index which has health education living standards like a poverty measure, governance, time use in the environment and then community culture and psychological well-being. And in terms of developmental states it was interesting when there was an international community interacting with Bhutan's Gross National Happiness Index. One of the lessons that they got quite interested in was that they realized that happiness is a skill and you can learn to do it. And so in a sense also coding where people are on that skill of happiness which maybe people think is Buddhist but you can also from mindfulness or from psychology you can enter into that path. An interesting thing is when Bhutan updated its Gross National Happiness Index in 2015 from 2010 there are statistically significant increases in material and income in jobs in healthy days in the services of water and sanitation and roads and electricity in education. So all of those went up significantly and overall there was growth in Gross National Happiness Index. But in the psychological well-being category every indicator significantly declined so that was satisfaction with quality of life, positive affect, negative affect and spirituality, sense of belonging declined, a feeling of etiquette and courtesy declined. And so what was interesting is because you had that panoply of indicators you could see the simultaneous movement in different directions and I think that's quite useful in a sense catalyzing a conversation about where societies are going and whether that matters, whether they like that change or might want to redress it. And to Jimson, we work with, most of my work is not on the national MPI, on the global MPI most is with national governments who are doing their national MPIs when I'm not doing research. And clearly the priority is that the voice, the protagonists of poverty, the poor people would have a very strong input. So for example in El Salvador there was a two-year participatory exercise with different communities after the government already had a draft measure and the draft measure had health education, living standards and employment in it. And after the participatory exercise they added lived environment and violence because those came out so strongly from the communities and it seems very important that a measure of poverty should reflect the experiences articulated by those in poverty. At the same time you need very much this not to be only a statistical exercise because if the minister of planning, the minister of finance, if they are not on board and don't see its relevance they won't use it. And so there's also been an interesting engagement in countries setting up different kinds of committees that in a sense have the user of the statistic so that by the time it's launched President Cordrea launched it the head of statistics explained it the minister of planning Nancy and the minister who is doing the targeted social programming both spoke of how they were going to use it. And that means that a statistic has an audience that's anticipating its reduction and thinking towards action. So it seems that both involving the poor communities and in a sense the users of the statistics is key alongside of course the statistical community. And Pia, it's interesting in terms of data gaps what is, what the, we are secretariat of a network of 53 participating countries who are designing or using multi-dimensional poverty measures so I learned by listening and what a number of them are saying now is that really it's up to them to prioritize and they may not prioritize the missing data they may prioritize actually making some steps based on the information of data that they already have and I think localizing or whatever term is going to be used but setting priorities is very much you know the topic of the day because it's impossible to do them all and for me there's a big danger in multi-dimensional poverty measures if there is a explosion of them of simply statistical overload there's too many numbers and what we observe is that when there is a multi-dimensional poverty measure then in a sense it brings together 10 Costa Rica has 20 indicators which is too many but it's a lot but it brings them under one umbrella and so in a sense they get a kind of serious attention and the danger with 232 is just dissipation as we all know and so trying to find ways of making it easy for again statistical users to engage with the data that do exist in an action way I think is more of a priority and so that's I'm just thinking of the countries they are all doing new surveys they're all adding little modules but it's the ones that really seem to be more important in their contexts and regarding the identification there's also a downside for example in the Adhar program in India there's people who will be left out people for whom it doesn't work the fingerprint or there's a risk of it being used in a way that it may not be completely constructive and might be exclusionary or harmful rather than just providing benefits and so I think there are a lot of ethical questions it would be ideal if we could match different data sources but the distance from here to there is quite big and so what we observe is that in many of the conversations on big data or the geospatial or the merging of data sources it's very exciting but if we want poverty data now at the moment that definitely includes poor people it really has to come from household surveys and I see that one of the dangers is that there are champions of big data there are champions of geospatial lists or administrative vital statistics that but there's really no champion of basic household survey data that's extended in the way Sir Tony Atkinson recommended to look at missing populations and you know some of the well-known problems of household survey data but I do see that as an area that desperately needs ongoing investment and is being overlooked in the conversations Thank you Yes, here in the front and then we will let's take the front first Hello, thank you I'm the chairperson of the Finnish Deaf Association here in Finland my name is Marku Jokinen and I'm also the president of the European Union of the Deaf my question is more of a comment to the MPIs about the dimensions of poverty one of those is education and I'm thinking or my question is what would be the other other aspect would be also language and the use of language and also the barriers in using language because education you can get to education but if the education is given in another language than what is your natural language then it's going to be a bigger barrier to you than getting into the education or getting the information if there's information and you don't get it in your own language it's not usable to you so that's also a barrier in your life if you don't get the information that you need and you cannot do the conversation with different people and you also face this even greater barrier in your education and in acquiring knowledge what is your thinking about the MPIs and the poverty connecting to this access to information and to language so can you measure this and also can you eradicate poverty by this and what is the relationship with language and I'm not a statistical person myself but I'm thinking about what is your opinion of this I'm more of a linguist and an educationist so this is where my question is coming from, thank you Sure there was a hand here in the middle Hello, good afternoon my question is here before I ask the question I would like to say something here I think being a poor child today is most likely to be a poor parent in the future being said that my question is for you can you think an individual can be trapped in a poverty if it is how does the MPI will elaborate to explain it will help to reduce the poverty to cast SDG goal and my name is Rajiv Casey from Nepal, thank you Thank you and then here in the centre just next to the one who is taking all the video from today Hi I'm Surabhi from India so I just wanted to ask how do you find solutions or how would you implement these goals in places that are so diverse even if you consider nations each a country like India is so diverse the struggles of people would differ geographically I'm a little nervous but if you see poverty it's the struggle to I think it is when people are struggling to get access to health and education and these are constructs and the I see poverty there when people are trying to get access to this because it's imposed on them in some way and if and as you say the surveys are done at the national level and even there the data would be skewed because they wouldn't probably be accessing every region I'm interested to know how the data set is collected related to a question that was asked there but again it's how do you connect these goals to places that are culturally very diverse Sure so how does this relate to the use of language access to information I guess the second question was about intergenerational transmission of poverty and how that relates to the body dimension index and then how do you really do this implementation across such diversity Very good in terms of language the example I can give is from Colombia and it is when they realized that their national measure was not relevant for the indigenous community and so they created an indigenous MPI with the community and their number one request was for education in their language and so it was a very articulate demand for changing the specification of education to include the language of instruction and so far that's the only country that has put language into their national MPI but because that has visibility in other countries that have an indigenous population at least that has been discussed so that it's incomplete but it's what we have to Razif's question in terms of intergenerational so the global MPI uses repeated cross section data but there are studies with panel data and with panel data you can see in a sense the chronicity of multidimensional poverty and also see what deprivation changes so people drop into poverty and for if you isolate the people who are chronically poor across time periods you can see what is the caustic combination what are the deprivations they always have now clearly that'll be partly a function of how you've designed the measure but it's interesting also to see what combination tends to be associated with chronic poverty so at this stage we're simply doing a kind of a descriptive analysis but a colleague is thinking then of extending the poverty trap literature and doing more analysis on that but right now we're just going through different data sets with panel data to try to uncover those combinations that tend to regularly be associated with chronic multidimensional poverty there's also different papers for example Luis Felipe López Calva of the World Bank found in one paper that multidimensional poverty was associated with chronic income poverty another paper found the opposite so it's still a literature in a time of flux but interesting questions and to Sabida I think I'm not in terms of your question it's a very good question and I must say I'll give a little advertisement for sense and solidarity I don't know if you've read it there's a lot of economics for everyone by Gendres which came out last month it's a fantastic book and it's very much a book which has relevance to these conversations and to your question because he articulates how research on poverty cannot go without a very real interface with poor communities and their activities and that certainly Gendres embodies that in his own work and observations in India the MPI work has gone ahead at the state level not nationally and so Andhra Pradesh for example has a state level MPI Assam had one and it's state level human development report so those have been the areas of interest but I think the I gave earlier the example of El Salvador but some of the interesting work is when NGOs understand an MPI it's very simple it's like a counting based measure and so the NGO sector can do it easily in their own communities you don't need fancy software so if you think of in Kerala the program of targeting the destitute families or whatever it's a counting based measure if you have three of these nine deprivations you're identified as eligible so it's an intuitive work approach that can be used locally and not just nationally and I think that makes it more relevant as you said to the different deprivations of different cultural groups so that those are some suggestions but in India that's yeah we're not doing anything nationally your question about data so we are waiting for the NFHS for data which should come out in December January in India and it'll be representative of the district level nationally so there's a huge survey but it will have some representativeness but again it may not pick up on specific context of different regions okay we will take the next round so we will up here almost in the front good evening my name is Obain Godfred I'm really interested in social and public policy and I know what you are coming up with is going to be some of the basis for most policies, formulations within national and international levels you talk about education in relation to poverty eradication and I'm really interested when you talk about having access at least to the grade 8 and I really want to know how getting up to grade 8 for example can really effect poverty reduction or eradication, eradicate poverty looking at most especially if you look at developing countries curriculum structure and mode of transmitting knowledge do you think or is there any kind of empirical evidence which shows that once somebody reaches grade 8 it can really effect promoting or eradicating poverty that's my question, thanks okay and then there was one here, yes next to you Charlie good afternoon my name is Jovin enlightening presentation thank you very much if I made two questions in two weeks the United Nations conference of parties is happening and that could be one question you would like these decision makers to have in mind in order to move the conversation forward the conference of what? the COP23 is happening in Germany and second question is as you churn all this data with MPIs and so on I'm curious to know whether you see companies, startups, SMEs trying to plug into technologies etc. in order to move things forward in terms of applications in terms of making use of these great sets of information that you produce thank you okay here in the front hello I'm Jukka Birtela from University of Tampere and also affiliated with UVU Wider I'd like to hear your views on moving beyond poverty to inequality to what extent would it make sense to use the same underlying indicators to measure multi-dimensional inequality because now it seems that the poverty is measured both in monetary terms and then in a multi-dimensional way whereas inequality continues to be measured only in terms of either income or consumption okay on the eradication of poverty on the use application of the generation of data that has been taken from this and then on can this be replicated for inequality very good in terms of the eight years of education that's basic education and so we just used it because it's one of the levels if you look at the national MPIs that are being developed most countries employ the number of years of compulsory schooling which is in effect at the time and they vary a great deal what we do in the global MPIs we had actually a range of different cut-offs for that variable and we consulted we implemented six years, eight years, ten years eight years is also not compulsory in all countries but it seemed because it's the basic education standard and it's beyond primary school it seemed one that we went with but I don't you can cite certain papers but when it's a global measure it's going to be messy and so the eight years is going to be relevant in some places and not relevant in others but similarly a measure of acute poverty will not be relevant in all places but it will capture in a sense a consistent population of deprivations during in terms of the there were two different questions in terms of the business MPI that is taken off in Costa Rica so Costa Rica has a national MPI and actually its development was co-funded by the business community and then the largest bank was curious whether any of its employees lived in an MPI poor households so they got a PR firm to do a survey using just the questions they needed to construct the national MPI and they did that and to the interest but also consternation of the leaders they found that there were there was a non-insignificant number of employees who were living in MPI poor families and furthermore they weren't only in the lowest paid jobs there were some in the middle paid jobs and these tended to be because of high dependency ratios disability at home high unemployment among youth and so they looked at the again the composition the profiles of deprivations and the implemented some vocational training programs in order to recruit some of employed people and give them proper skills and some other programs that's then taken off and now I think 80 institutions 80 in Costa Rica are trying to replicate it so that's a point to watch and some of them are international firms so it may move outside Costa Rica soon so I hope that's enough in terms of inequality theoretically it's completely possible so what you have a matrix of people and dimensions and if you take the 01 whether or not they're deprived in something you could do that or you could take their overall achievement but if you take whether or not they're deprived in each indicator you could take the reverse of that which is whether or not they've attained a certain cutoff in that indicator and aggregate and so you have a score of weighted attainments which is currently meaningful for the population and then you can take that vector and you can make any inequality measure you want tile one two Atkinson Genie whatever 90 to 10 ratio I think empirically the problem is on the data side not on the measurement side so for example I mentioned that if you take a union measure of the global MPI then 75% of the world are poor and that is largely driven by cooking fuel so many people cook with wood but they may have good chimneys so it's actually not a health risk but the survey doesn't have the right question so it's what I would call a spurious deprivation it looks like a deprivation but it's probably not for them in the case of the global MPI we censor it we clean it if the people are not deprived in one third of deprivations 60% of people in Bosnia cook with wood but they're not all poor then we censor it from the data set so what you would need to do for an inequality measure is limit it to those variables where the deprivations are accurate in a sense each attainment or each deprivation is accurate and you don't have the data errors of some indicators so some of the global MPI indicators are very rarely censored they're quite accurate and some particularly cooking fuel is quite inaccurate so that would be the difficulty James and I are doing a paper now both looking at inequality among the poor and then also looking at this attainment matrix and I can bring on inequality measures and what you can do multidimensionally with very similar techniques as a county based poverty measure okay further questions okay yes here there are so many that second rounds will be difficult hello I'm Soumya I'm also a scholar at wider I was wondering if multidimensional poverty would be much lesser in those countries where gender relations are better because you're looking at these basic services water, sanitation and those and particularly at children so children are much better off women are also much better off in the region so what are your views on that thank you okay one in the background I'm just curious on the question of aid allocation coming from an aid agency how do you see the MPI and the impacts if you compare with income poverty and the 190 any if you try to think about allocation by countries or by sectors or by themes and the second one is the use of MPI in terms of guidance or policy reform you talked a little bit about it this obviously it's a complex question but I would be interested if you have any examples or if there is dialogue with governments and if you see anything in terms of expenditure patterns and thinking of expenditure patterns okay thanks and then here in the front and that I think Tony Addison from WIDA the very first WIDA annual lecture was given 20 years ago in March 1997 by Nobel laureate Douglas North and the world has made quite a lot of progress in 20 years on poverty partly due to the MDGs and the SDGs so if we could look 20 years into the future what do you think the world of poverty will look like how much progress will we have made and how will we have made it and which countries might be ahead and which countries behind so if we go forward 20 years what will our world look like okay three small questions one on the balance the reference to gender the whole question of aid allocation and policy reform and how the index might help inform that and then what's going to be here in 20 years very good so on gender I must say one of our disappointments is that the data that are available do not permit us to make a gendered MPI but we can't have one where we can disaggregate meaningfully by gender we've disaggregated the global MPI but we haven't published the results because it's not correct it just reflects the demographic structures of the societies so I think really we have called for in fact our network I mentioned we're the secretariat of our network the network designed a gendered survey and proposed it for the SDGs so that we could have within the household multiple respondents we looked into a feasible and inexpensive way of sample design for doing so and we designed the survey it didn't go anywhere so I think really there I would join my voice with many others and call for better gendered data going forward because we can't really use this data and answer your question because we simply have no information on intra household sharing that being said we have done individual child poverty measures and we're doing more of them we're doing a lot of them where we use the global or the national MPI for a number of indicators and then for example schooling or cognitive development and nutrition or health we look across the age cohorts of children and have individualized indicators and those we do show up gender differences so that's I think a step forward and there are data for doing child poverty measures in the mixed surveys but it is a it's a disappointment and it's baffling that we don't have better data in terms of aid allocation it's not an area in which I'm an expert but we had a little bit of a look this year because of the first of all the disparity between the low income country category and where the MPI poor people live with 72% of them living in middle income countries which is similar to the percentage and the summoner found with monetary poverty when we looked at the aid distribution and we took use different definitions and they're published in our policy briefing of this year we found that the distribution across low and lower middle and middle income countries was actually surprisingly more balanced than we would have anticipated but what we found was that the allocation per poor person according to the MPI was very, very noisy with India and China having less than a dollar per poor person per year and other countries having a huge amount now clearly that's influenced by PPP clearly it's influenced by national public expenditure patterns that aid flows tend to complement but there's I think a lot to be unpacked and so all we can do is we can say per MPI poor person what are the aid flows to these countries in nominal terms and then the conversation must go beyond that with others and in terms of policy change I could give the example which President Juan Manuel Santos of Colombia would give in that case which is he launched the MPI in 2011 and he set a target for its reduction and then he worked with Mackenzie and he worked with public expenditure people to try to figure out the best patterns of fiscal expenditure to be used he had a committee that met twice a year the ministers could not send deputies they had to meet with him annually they updated their national MPI and in between they updated it using administrative data to try to look at the trends and then when an indicator was not moving to plan they had responsive policies and so they were able to greatly accelerate the reduction of multi-dimensional poverty they had analyzed it back since 1996 using the same survey so they could accelerate reduction but it was using the same fiscal envelope and so the same budgetary envelope just spending it better so that's the kind of story as I mentioned Costa Rica is another one that's invested a lot in allocation because they found there was zero allocation to some of their MPI indicators and that the allocation to the poorest regions was not comparable to the extent of poverty in those regions so the rebalancing of existing resources seems to be a very common first step that does have an immediate effect Looking into the future I'm not very good at that clearly my hope would be that the acute poverty could be eradicated when one doesn't know going ahead and so I'm not sure how sensibly I can address that I think what's clear is that as countries come to very low levels of multi-dimensional poverty like Mexico then they redefine their ambitions and they use a similar structure but now Mexico is 1.2% according to the global MPI of poverty but 43% according to their national MPI their aspirations and frames of reference have changed and so in terms of poverty and development hopefully there would be a re-articulation of the appropriate sets of goals clearly if climate change permitting and all of that we are able to to reduce the kinds of acute poverty that we now face and I think also that the psychological domains and that the other domains of well-being may come in more strongly as we are able both to measure them and to think about sensible and pluralistic policy responses to them there seems to be an appetite so in a number of countries not just Bhutan we are seen not only the poverty measure but now them wanting to do a linked well-being measure that includes the dimensions of poverty but then goes beyond it to some of these soft things as an experimental measure not yet to have the kind of seriousness of a poverty measure but to try to keep these things within the field of vision and that I think will be in a measurement terms quite difficult but interesting Sabine can I sort of just add one observation and we've been spoken about data and so on and I just hope you will bear with me just making one reference to wider work on Vietnam we have actually put out a whole series of YouTube videos on what the data revolution actually means in practice so I mean I sort of want to try to get a sense of what some of these things actually mean in concrete practice out there and what you can do once you start over a period of 10-15 years having built up a panel data set sort of what does it take to actually get to the point that some of these questions can be addressed in greater depth and so on I mean I'd like to sort of just make that point because I genuinely agree with you on this issue of that we do need to make sure that household surveys do not get disappears in this sort of big data is now all over there and it can answer all our questions big data cannot answer all our questions we do need to get down to the household an individual level and collect that data in order to really come to grips with it and so I'd sort of like to make that point I may just as one final observation or question I participate in the formulation of the SGGs and I mean I was in numerous meetings and of course there's the number of indicators and measures and so on grew bigger and bigger being after all trained as an economist I started getting a bit worried because I was sort of thinking okay how do we operationalize all this I mean I certainly agreed with the political intentions behind and so on but I did get a bit worried about it and then every time I would sort of ask about do we actually need this measure or this indicator and so on I was constantly confronted with the following statement rights are indivisible you cannot sort of start trading one dimension off against another one you have to have them both I'm kind of pondering what you think about that I hope my question is clear because it's something where I believe that at least the economics profession but also other professions have a big issue that we are not always kind of managing to get to click among us so I was wondering whether you had some reflections on that well I think two or three things one is that not all SGGs might be framed as rights so that's why yes there is the language of integrated and visible there because they're interlinked but perhaps there's also a language of prioritization and then if you think of the incremental realization of social and cultural rights there's also a recognition that with limited resources it's essential you know it's impossible to advance on all of the fronts together and so prioritization and setting medium term goals is essential because it is impossible to do otherwise and I think that actually where countries are engaging the SGGs and not all are that's the exciting point is when they feel empowered then to look at that basket of goods and select the first step of actions to be undertaken and that we'll actually put them towards the goal of realizing the SGGs but do so in a way that also coheres with agenda 2016 or with their national development plans or with some other priorities which are also salient because it's not only the SGGs that have a voice in so many contexts sure time has run out I hope that first of all you will join me in saying thank you very much to Professor Akira this was great as an afterthought there is a reception there is a glass of wine and there is a bit of food also outside can I invite everybody to join us in the reception and let the conversation continue and thank you very much