 I think I would start with probably what you have heard is the new normal for IMF, both in terms of I think the fund has learned particularly from the more recent set of crisis that actually have gone beyond simply the developing world but also some of the developed world that there are certain important issues that may need to be looked at beyond the usual analysis and the mandate that the fund actually has and I'll refer to some of the things in here and the main reason I started with that is really to applaud that ultimately the fund actually is aligning to a whole range of concerns that have been quite alive not only in the academic sphere but also on the policy side. If I was to roll back my 10 years as governor and I was to move forward with this kind of view that the fund now has I think my life would have been much easier than it has been in the cause of the actual part. So it is a positive applause I think ultimately that this important influential institution has come to terms with very important developments. Now my comments will be made from two perspectives. One is alignment of the fund research which we just heard with the major emerging development concern of inclusion and we know this comes on two dimensions as an end or objective on its own behind the term leaving no one behind and also as a means to development and I think the bulk of Jonathan's presentation has focused particularly on looking at inequality and redistribution in the context of how it aligns with developments in growth and also welfare more generally. A second perspective is alignment to the core mandate of the fund that of providing policy advice for macro stability and sustained growth. How does this perspective fit with this and I think the authors know that this is our second engagement we have had so I wouldn't want to belabor on the points that I made a lot earlier they made this presentation in Oxford I was also a discussant and I think there was a very high degree of appreciation for how the fund is actually coming to terms with fairly difficult questions in the context of the mandate. I did raise also a question in that connection of how ideas move from research which is sort of the thought frontier to operations at least in today's presentation we did hear a bit about how these ideas are being brought into country operational context but my big measure would be in the main policy documents of the fund which guides its operations and the relationships to the various countries. I think as you know the thinkers have always been ahead of the doers and bridging that gap between thinkers and doers within that institution it's not a small task and they know it and I think it is also the role not just of fund staff in that context also the board of governance which include precisely people who hold responsibilities in the member countries to really take these messages and get that reflected in the policy manuals operations that the fund would actually be making going forward. There are three conclusions that really grabbed me on the basis of the presentation. The first one is the fact that inclusion is taken to be a means to develop and this is really shown by the result that lower net inequality drives faster and more durable sustained growth or stated otherwise higher inequality is bad for growth. So and this I think is a very fundamental shift from the conversation that we had earlier. The second and I think probably the more powerful one is redistribution to realize inclusion as an end is generally benign in its impact on growth. I looked at that line which was fairly much flat and not showing any negative major impact but when you look at the other diagram which had both direct and indirect impacts when you combine the two actually you get that the combined direct and indirect effects of redistribution are pro-growth which is a very strong conclusion. This says we don't have really just to wait for the trickle down effects in the process of growth particularly where growth has tended to be moving with inequality but governments, countries can take proactive actions to actually engineer equality through redistribution. It was definitely a no-no in the past and this I think is a very very important turn of events. Now below that there are three others as a policymaker that really again grabbed me. One is austerity can be costly and we know in the past whenever you had a crisis the policies that typically all of us were required to pursue were pro-cyclical not countercyclical and when you are in trouble you are required to exercise austerity. I think part of the lesson must have come from QE experience that the world has gone through where in times of really stringent crisis in some of the key countries quantitative easing actually has been taken to be an instrument to get away from from that trouble and particularly for reversible shocks we know you can finance reversible shocks. It is only the permanent shocks that you need typically to adjust to and it is I think a very important if you want redirection of a policy stance from that kind of conclusion. I was also struck by the conclusion that paying down debt rapidly can be more costly than living with it. This is a huge statement and we know that's part of the new normal where you can live with debt to GDP ratio of over 100% in some cases provided you are still going through a process of growth and let growth actually get your ratio down rather than engineer if you want to pay down of debt in order to get ratios to the right context. And lastly it's the very strong again conclusion that expansionary monetary policy can reduce inequality in the medium term and that tight monetary policy can actually increase inequality. For some of us who have been practicing on that sphere it would be really really strange if I heard someone from the mission make this statement in those times but again as I said this is the reality and it is true that you have got just to consider the context in which these particular policy stances are actually taken. So let me conclude by just answering one question what I find significant about this research. One it is the fact that this research is being done in the fund because that was the last frontier against precisely the traditional view not against holding forth the traditional view and I must applaud the research team for being brave enough seriously. I worked in the World Bank myself in the past and I know how difficult these things can be when you try to make statements that typically are breaking new ground in an institution which is regimented by tradition. So this is one significance. Second significance is the fact that it goes beyond a good number of the academic research that was being done previously. We know Barrow in 2000 still upheld the Kuznetz U-curve, an inverted U-curve but observed the exception in cases where incomes are less than $2,000 per capita incomes are less than $2,000. That was the first sort of movement away from really sort of the traditional view. In between Nancy Burtzel and her team Ravi Kanbul, Alwaria all used selective evidence to support these same conclusions mainly using the Asian experience to show that you can have rapid growth also with decreasing inequality. And now we have here research which generalizes that particular result not just drawing on selective examples but using data sets that cover the whole world to be able to make the same conclusions. And I think for that I would end my comment in applauding the team for being brave enough and the challenge is to move the idea from ideas and good advice to practice. That still I think is work in progress. Thank you. In the interest of time let's zoom in straight to a few questions. We'll take just one round of questions and then we'll break for coffee. Please, the gentleman at the back. The set of questions and then I'll hand over to the panel. I'll be very quick. So Douglas Arendt I'm at the National Renewable Energy Lab. I've done a lot of work both with UNU Water but also with the broader IPCC and UNFCCC on energy climate inequality etc. So I have two questions very quickly for the panel. One did the theory that was presented at the beginning actually hold for China's growth because it's not clear that it does. And then secondly anywhere in your work or your colleagues work are you taking into account or plan to take into account climate both impacts as well as mitigation because it's already known that that's disproportionately affecting inequality. Thank you. Thank you very much for a brilliant panel. I just wanted to follow on Ben and Duluth's excellent points about how to take forward this brilliant research and perhaps just ask about three specific points. I think the evidence is very clear as presented by Jonathan Ostry and many others that income distribution redistribution favors growth. However there's still and there was in one of the presentations an emphasis on flexible labor markets. Now flexible labor markets have of course positive aspects but they also tend to weaken labor and if when labor is weakened through say weekly collective bargaining wages real wages tend to go down or go down as proportion of total income and that is a very strong element for deteriorating income distribution. So can we make sure that all the policy prescriptions are in line with the broad aim because if not with one hand we're saying let's improve income distribution but with the other hand the fund or others are recommending policies that may actually deteriorate income distribution. Secondly it was shown again in the presentation of Jonathan that repaying debt too quickly which means like having massive fiscal surpluses to do it has can lower growth and I think the evidence is very strong but for example in the Greek program they are still committed to have very high fiscal surpluses for a very long time over two and a half three percent which is you know very difficult to make consistent with growth and I think this is a very small example Barbados has just signed an IMF agreement which I think overall is very good but they talk about six percent primary fiscal surplus and I think if anything can kill an economy is a six percent primary fiscal surplus and finally Beno Dulu made a nice point that reversible shocks can be financed and I would like to ask whether there isn't a case for these excellent compensatory financing facilities that the IMF has to be increased particularly in the light of possible shocks which are beginning to merge in the world economy like tensions in the emerging markets likely increases in interest rates in the US and so on and whether it wouldn't be a good time to to revive these very good facilities which work well but which are relatively small scale. Thank you very much. Thank you very much Any other questions please Fin. Thank you very much for an excellent talent I mean I really do appreciate it and I certainly would like to sort of subscribe to what Benny said if Jerry Heliner had been able to be here today he would have said it's nice to see the IMF going wilder in in in the early years actually he and a number of his colleagues were basically characterized going wilder with wider but so this is just a stress that I do also highly appreciate this line of research and I really seriously do that I have one slight worry I saw that I mean in in two of the presentations there was a strong reference to to the use of an inequality database which is somewhat controversial and this is the salt width and there is actually advice out there not to use that database by leading people like Stephen Jenkins and others who have have looked quite carefully at it so I'm just sort of wondering whether I mean have you been digging more into this have you tested the I mean how conclusions might change if if you actually don't use his approach to filling the gaps it's just to sort of because I'm a little bit afraid that I mean might there be a backlash if it shows up that when a more appropriate database becomes available and then some of the conclusions might not stand up I mean I'm asking the question more is sort of a data a concerned person rather than in any way suggesting that I don't appreciate the policy conclusions because I do I mean and I think anybody looking at why does work would would would agree with that but I'm sort of slightly worried about that data point thank you thank you very much Finn the microphone here please um good now I'll send you any wider thank you for lots of very very interesting presentations so far the comments have been very positive from the flow in a slightly different track um and so in the first presentation Jonathan what I wasn't really clear about the causal mechanisms behind the inequality growth relationship is it true great constraints being uh being elevated collateral being moped for the poor that leads to higher growth is it true essentially human capital distributions changing so the equal opportunity changes that leads to higher growth is it true essentially market power being reduced or bigger firms that leads to higher growth because market park is just detrimental growth so all these causal mechanics are not clear and therefore one kind of worry is that what really capturing are simply correlations not causality because in a parallel regression the inequality growth issue goes both ways we know and it's very difficult to have instrument that actually meets exclusion restriction in a time varying sense so without getting into the causal mechanisms it's pretty difficult to understand whether we have a causal relationship here or pure correlation so I just worry about the fact there isn't much hasn't been as far as you could see more thought on the causal mechanisms similar question of second present uh second presentation expert diversification being good for growth but take take think about Bangladesh when balance out of the growth process pretty much went into ready made garments it still has about over 70% ready made garments nobody will ever say to Bangladesh that you know maybe in the future they should but right now they're doing very well to have the highest growth rates in South Asia and certainly in Asia itself um outside of East Asia so if you look at Bangladesh experience and diversification or concentration has been good for them but then you look at say an oil rich country so you concentrate on oil yes you can see the problems with oil concentration being essentially bad for growth or volatility of growth so then the question really is that what exactly are we diversifying away from and what are we concentrating into also are we diversifying in a way that's market friendly or market comparative advantage conforming or not so I think the story of diversification and growth has to be thought through a little bit more carefully because I'm not very sure you can actually go and stay two countries in many parts of Asia for example who are concentrating on particular kinds of commodities which have been good good for them that they should think of getting away from with those commodities and going in for diversification so it's a real bit more complex than what I think we heard so far thank you very much in the interest of time I think I'll invite the panelists and we can continue the discussions in the corridor so during the coffee break so thank you very much for the questions my colleagues will respond to some of them we're not going to respond to questions that where the basis for the question was not in anything that we said so we're not going to talk about compensatory financing facilities and and so forth I'm going to respond to a couple things your point about the SWID database I think you were slightly misleading in mentioning Jenkins criticism of the SWID database without mentioning Fred Solz rebuttal to Jenkins so we have to be balanced here people have lots of skin in these different games and we have to be scientific we have to put our data out we have to put our results out and we have to be open as you said to people running running new regressions using alternative data and overturning results we have to be humble we have to be scientific we have to be as objective as we can the only thing I would say also is that if you can suggest an alternative database that has the coverage in terms of the cross-section and the time series that Fred has we would be open to using that but as far as I know Solz is the only game in town for a study of of the one that we were interested in running covering a variety of countries at different stages of development and I think Fred the SWID database he is very upfront about the many many shortcomings of his database he has nothing to hide we decided to make use of this I think to ask this question but we're certainly open and humble and and and so forth channels the tremendous difficulty of causal interpretations of macro data again let's not be holier than thou here I think there is a long discussion on causal mechanism in the JEG paper there are probably 10 possible or more causal interpretations there are fertility channels there's Roderick's story about resilience there's access to education to health to the political process to nutrition all of these things this is we I think we are upfront and humble about what we can say on the basis of macro data you can say well if you can't say anything that's truly causally convincing then let's just go home we don't that's not our philosophy we we we did the best we can with macro data we're open for others to try and and do better I think Chris will talk about climate and David you may talk about the labor market I do want to say on China you know yes and no people say you know they'll they'll choose different bits of China's history and say what was growing well and highly unequal it was taking off from a position of tremendous equality and doing well yeah so I think our point is to try and draw inference from the wealth of all of the data and we have a discussion of China in the paper we have a discussion of Brazil you can pick different countries and and and use our results however you please minute each and then very quickly to respond to a couple of issues that came up one on climate change and the other on diversification on climate change we share with you the concern especially when it relates to macro issues you know you know issues of climate change that affect the macro picture of the economy this cannot be more true than when you look at small states where shocks are have an immense impact and this is where we started to do some more serious work if you think about the climate shock is more or less like an economic shock differs a little bit in the nature but it hits the country like an economic shock so we have a couple of excuse me couple of good work to show for what we have done so far the first is a wheel chapter in 2017 april 2017 I believe was more focused on developing economies as opposed to emerging and advanced economies and then we have a very nice paper just is forthcoming J and journal of development economics on focusing on the impact of natural disaster shocks in growth and inequality in small states on diversification I couldn't agree with you more I mean we are not advocated that diversification is good for growth for all countries actually the opposite and that's why we created this new index of quality just to make this point for commodity exporters in small states we recognize also that these guys have the comparative advantage on you know focusing on only a couple of products the question is how do they go beyond that and the answer could be on improving on the quality dimension thank you just a week on the labor markets I think like indeed we have recent research showing that you know policies to make the labor market more flexible along the line that you mentioned indeed a shown evidence to reduce the bargaining power of workers and to reduce the labor share of income so we find that for several advanced economies actually these policies have had a relatively sizable effect however I think like this is discussion falls a bit in what Jonathan was saying is that many of these reform like for example labor market are typically intended to boost efficiency to boost the supply but at the same time I think more work has to be done to realize that many of these policies actually have distributional consequences so I think like for you know we are doing a significant progress in showing evidence that this is the case and again I guess the next step would be to to tailor these policies and in the designs in a way that can have less unfiltered of I just want to to say that as Jonathan said that we are not going to enter into country-specific issues but there is an ongoing review of our concessional facility that should be completed this year just for your information