 I have a paper for the volume that Deepak has edited and we've already had two rounds of discussion on that paper so I think for that reason Deepak was leaning on me gently to go more into methodological matters rather than giving again a summary of that paper. And that is exactly what I will do today, picking a little bit on the Indian context which my paper is all about and of course Mirdal's influence. I want to go into methodological questions and actually it fits in very well with what just was the two papers, two talks that you just heard. You know for my generation I went to a college in Delhi in 1969. Pressure to read Mirdal did not come from any intellectual desire but a sort of fashion statement. In 1969 Mirdal's book had just come out so to be seen crossing the college yard with one of the volumes in hand raised the status of our students who could do that and if you could talk a little bit about what was there in the volume that gave you an even bigger status in my college in Delhi which also happens to be Deepak's college. So that was the first encounter with Mirdal and for me the appreciation of what Mirdal does in his three volume work came much later when I had studied a lot of mainstream economics and I had become an economist. Looking back you realize that there were dimensions to Mirdal's writings which were actually in many ways in somewhat inchoate ways Mirdal was anticipating, anticipating not mainstream economics as much as the need for broadening of mainstream economics and you can think of many different ways in which Mirdal's ideas were sort of precursors of later work and there have been illusions to that already. For instance the fact that economics should be viewed as a discipline embedded in other social sciences which is such an important element in the work of Mark Granovater which is also reflecting earlier work like Polani's work but nevertheless it is in Granovater that you get this coming up very prominently and Mirdal was already beginning to stress that. Values, the important role of values and hidden assumptions, sort of values that are inside our head and we are very often not even aware of that and they play a role in the way we analyze the economy that was there. The new institutional economics, the political element is there in Mirdal in the Asian drama and also in his book on political element in the development of economic theory. So what I'm going to do is, just to give you a sort of two-liner gist of the way I write the paper but I don't want to go into that, is the paper is the political economy analysis of India's development over the last 50 years and India does stand out in a particular way India's development that India made very heavy investments in certain political institutions early and this largely we owe this to Nehru, the stress that democracy was important, freedom of press and media was important, tolerance and open society was important and that's why when in 1991 India's economy is opening up, it's a very strange situation where economy wise India is a pretty close economy. In 1991 the famous statement which stuck in my head is Manmohan Singh in his speech as finance minister says India is going to bring down the tariff ceiling to 150%. There are very few countries which can bring it down to 150, you can raise it to 150 but economy wise India was becoming very closed but actually in terms of ideas, books, films coming into India, India was extremely open. So a lot of political investment in the beginning which may have slowed down the economy somewhat and we want to capitalize on that early political investment now and that's the challenge with the Indian economy currently but that's my paper and the full paper actually a very long version of it is going to go up on the wider website very soon so you'll have access to that. Let me just jump to some methodological questions which arise to do with hidden assumptions. You know I wanted to stress I mean this sort of it's hinted in Mirdal that there's a lot of hidden assumptions in economics and I think that's extremely important. There are two kinds of assumptions when you do economic analysis some that we write down explicitly as axioms. You take general equilibrium theory, you write down formally as axiom one axiom two axiom three human preferences have these qualities etc and these axioms have come under a lot of analysis and criticism. What gets overlooked is a whole lot of other assumptions that are built into the woodwork of the discipline. They don't even get written as axioms, they are presumed. It's almost like the analogy is in geometry. Euclidean geometry write downs a whole lot of axioms much later people realized that there are other assumptions in Euclidean geometry which are not even written down as axioms which are presumed and other geometries came into existence and I want to give you one or two examples of that and one which is the interface of the state and the government the state and the market a very important topic in Mirdal's work. You know take the idea of the invisible hand which gets codified later through the work of Kenneth Arrow and Gerard Debrou in the what is called the fundamental theorem of welfare economics and a lot of conservative neoclassical economics treats that as the sort of central argument giving it force. What is the argument that individuals if you give them freedom to choose we have our budget sets given our income we want to spend it on different commodities if you give people the full freedom to choose from their budget set the equilibrium that will come about is Pareto optimal so individual optimization leads society to an optimal state which was Arrow and Debrou's work which was a formalization of the idea of the invisible hand individual selfishness is good enough to reach social optimality and as soon as you have individuals freedom to choose from the budget set being curbed by the government you can't do this you can't do that there is a tax on a commodity the optimality result goes and that's why came the orthodox view that the government should not interfere but just think for a moment when we start with the assumption that individuals having the freedom to choose from the budget set we've already made other assumptions in reality individuals can do many other things than choosing from budget sets they can beat up other people and steal their goods they can use speech to hurt other people variety of things individuals can do extort exploit all that is assumed away we just start by saying that individual freedom is the ability to choose from your budget set the other things are assumed away but in the woodwork they don't appear as axioms if you began with the wider set of possible things that human beings can do hurt other people hit other people exploit other people start from that then the same theorem would have to be stated very differently if you restrict individual choice so that they can't do any of those other things but only choose from their feasible set budget set then you will get optimality the theorem begins to look totally different that if you can somehow constrain individual choice from hurting other people exploiting other people and simply choosing elements from their budget set you get optimality that sounds like restriction of individual freedom leads to social optimality so a lot depends on where you start from and you reach conclusions out of that the way it is stated these assumptions in the woodwork are extremely important to bring out I'll give you one more example something that very early I wrote a paper in economic and political weekly it's one of my shortest papers with one of the longest titles it's a two-page paper the title of which and this is when I was introduced to mainstream economics in my early days the papers title is why we do not try to walk away without paying after a taxi ride now after a taxi ride it's actually very easy to run away without paying and I know this is possible because there are college friends who had tried that and got away very easily you jump out of a taxi and run away in the crowds of Delhi you're gone without paying but most of the time we don't do that and I remember in the early days when I would ask my economist friends mainstream economist friends why don't people really run away without paying their thinking is so deeply entrenched in mainstream economics the answer would be well it's plain simple cost benefit analysis if you try to run away without paying there's a very high probability that the taxi driver will come chasing after you catch you and thrash you and the pain of that beating is greater than the ten dollars you would have saved by run away by running away so the reason why we pay is that the taxi driver could hurt you and that's why you pay up if that is the only reason you pay the taxi drivers are strong and they can threaten to pay you another question of rationality opens up and that question is after you have paid why don't taxi drivers turn around and say I will thrash you the way I would have thrashed you if you hadn't paid unless you pay another ten dollars taxi drivers typically don't do that after you've paid they take it away and that's the end of the matter the point is deep down in us as passengers in the taxi driver there are certain values and prior norms of behavior one norm is that you don't typically try to run away after a taxi ride without paying and for taxi drivers after people have paid you don't threaten to thrash them there are exceptions I know but this value is very important the reason why people don't steal other people's wallets again again mainstream economics will tell you they don't steal other people's wallets simply because they could get caught and they could get punished if that was really the reason that every time you went into a bus or a metro you'd feel dreadful to know that everyone is calculating whether it's worthwhile stealing your wallet or not and doing cost-benefit analysis and for that reason not stealing your wallet the answer is no human life is full of values that we adhere to and life is made possible by that I feel this these references from Mirdal come out very very important but I have to say he doesn't take the analysis far enough but on the other hand this was writing in the 1960s a long time ago the fact that he's pushing you in these directions is extremely important and I wanted to jump to just one last comment two minutes it'll take me Deepak is one area of development economics which has become now very very prominent which is the use of a lot of data and randomized data and I do think it's playing a very important role but one thing again which gets missed out over there by people who do this analysis they think that if you adhere to the laws of statistics and do randomized control trials and get some hint of causality that's it that can take you to policy what I want to stress is that the importance of common sense and intuition without that actually you would not be able to go an inch forward from these statistical findings and I'll give you one example very very famous paper and a beautiful paper by Raghav Chattopadhyay and Esther Duflo this is about women leaders in local village councils if you have women heading these local village councils does it make a difference to the kinds of policies that are adopted and they got a beautiful randomized control trial because India in 1992 made some amendments to the constitution whereby one-third of the Indian local councils chosen randomly could only have women leaders so here the choice of a woman leader is not because the local council is progressive and they vote that way but through a lottery you've already chosen which ones will have women leaders and then you collect the data and see do you get policy differences and yes indeed it turns out that you get policy differences the policy differences provision of public goods improves is that enough now to get to policy conclusions and causality the answer is no the sample they are statistically very fussy so what is that argument that you've got a population from which you draw a random sample on which you do these tests and you get this finding but pause for a moment the random sample in that in this case the study is done for the states of Rajasthan and West Bengal will this work in Latin America or South Africa or East Asia the popular the sample that they drew was from not from those populations to extrapolate to other countries other regions and to the future you are taking a study done in a particular region in a particular period and you're going from there statistics will not help you go there their intuition and common sense play a critical role there are many ideas from randomized controlled trials which are intuition will say these will carry over elsewhere and our intuition will say these may not carry over elsewhere and I can give you examples but I'm short of time where the intuition goes one way or the other intuition what I call reasoned intuition and reasoned common sense play a critical role translating the statistical findings to policy and one mistake that economists make economists do who do hard good work is they don't recognize enough the power of common sense and intuition it is the combination of the statistical finding findings theoretical findings with what we probably for evolutionary reasons have very strong powers of human beings have intuition and common sense which have survived evolution so there is something to be said for the intuition that we have that has to be combined with these results before you can carry them into the world of policy thank you very much