How to Get Into Public Policy & Development Economics (Harvard Economist POV) - w/ Jishnu transcript

Top 5% of urban families, how much do you think they earn a month? The answer is 20,000. That's $200. The top 5%. And it was the first of like a dozen things he said that just wrecked me. 30% of public schools have fewer than 30 children. One out of every three adults are functionally illiterate in the US. Professor Jishnu Das is the economist who built the data behind health coverage of over 150 million people in India. India. There is nothing called technical data.

It's always human data and if you want to understand human data, you have to understand humans. Looking at these data and a huge proportion of women are coded as salons. Now that's going to be a problem because they are going to be ineligible. So I asked him the question that by his own admission basically nobody asks him and it cracked the whole thing wide open. Everybody knows the latest Tik Tok trend but we don't know basic things about ourselves and yeah nan it's on you guys.

Please join me in welcoming professor Jishnu Das who has a PhD in economics from Harvard and a prior degree from the University of Cambridge on the ready set do podcast. Without any further ado let's jump in. Professor Das, welcome to the show. Thanks, Naman. Glad to be here. On your long list of achievements and the diverse experiences you've had throughout your career, I found that a couple things stuck out to me the most.

Maybe you will disagree but from the outside working with developing a plan that applied to over 150 million Indians for the RBSY seems like just an unimaginably difficult task at least for somebody uninitiated with policy work and how these things occur. So do you mind, you know, kicking us off with kind of setting the stage with what exactly that project was and what the inside of working on something that has such a grand appeal and outreach in terms of lives impacted looks like from your POV?

Uh I you know from the outside it looks more special than it is. Uh so the bottom line is look this was about the time this was being led by a very dynamic I officer called Anil Swarup who has retired since then and I was at the World Bank in Delhi and my colleague Robert Palasios was working with him and he pulled me into it. So I just worked as part of a small team uh on you know setting up and thinking through what this project would look like like and then from the outside it always looks a lot more glamorous than it actually is. Um so a lot of the work on the political side was really Anil Swarut's doing in terms of thinking through uh and I think he made like 200 visits to different states over one year. So uh really an enormous amount of leg network and a lot of patience on on on um doing this. Uh

then our work my work at least was a lot of thinking through how are the accountability structures going to work? How is the data flow going to work? Um so you know the initial data flows the architecture for that was all my design. Uh and I worked with the groups to actually put it into place and stuff. uh but also a lot of things on learning how health insurance may or may not work right so I think we learned a lot from doing it uh but the dayto-day work is [snorts] a endless series of meetings a lot of data analysis a lot of data cleaning right uh so it's not you know there's nothing that says policy is glamorous and then you have something else and then you have data the bottom line is any policy is first and foremost for people like me is an enormous amount of research and data and trying to understand how these things are working

and then as the policy went into place you know it's like any policy uh people constantly try to take advantage of it and your role is to see what can be done to keep the policy on track. Mhm. I think the most notable thing that I'm taking away from this is how much of this is relying on data which is not something that at least I was expecting to you know for that to be the big takeaway here. So when you're working in these teams do you have dedicated like tech professionals or is this something that public policy experts such as you are kind of trained to do in your curriculums and such? Uh how does that part work? The hard part. So, so we always we can always hire people. We can always hire firms. So, that's never an issue. Uh, the hard part of all of this is not actually doing the coding and the tech.

It's designing architecture, right? So, so, so if you think about something like this, the fundamental architecture you have to think about is what units does data flow across, right? So for example, you know, and and the way I like to think about it is always think about the precise actions that happened at each data flow point, right? right? So here the hard part is thinking about okay, when a patient comes in, exactly what data is captured at the hospital, do we actually think it'll be captured?

Do we not think it'll be captured? What do we think will happen? Right? Mhm. And then that data needs to flow from the hospital to whom? Should it flow to a central insurance group that then you know to the organizing say the state authority who then passes it on to the insurance company or should it flow straight to the insurance company? What if there's a problem with the data? Do we then go back uh when does the reimbursement kick in? So all so it turns out that you might think about these as technical problems, but they're not actually technical problems. They're human behavior problems. And I think the big thing that we bring in as people working on policy at the at the intersection of evidence and policy is to understand that there is nothing called technical data. It's always human data. And if you want to understand

human data, you have to understand humans. humans. Interesting. So, I guess when you put it that way, that almost makes the problem sound like it's no longer an objective problem with a set set of constraints or guardrails that we're working against and it just starts to feel like an, you know, just the worst mess of tangled wires that is basically impossible to sort out, right? So, how how do you work through that? So, based on what you just said, right, you have your system set up. you know this is how the data should flow and then in one particular district we find that for whatever reason there is like just a gap like people aren't entering that data there. So what are you as the developer of the policy in this case supposed to do with that? Is this just like an iteration where you now fix that and keep fixing things as

they come or do you have tentacles on the ground as well to make sure that these things don't happen? Like what take us through that problem solving phase of this if you would please. So I I think this question naman that you're asking is a question that very few people ask uh and more uh worryingly it's a question that strikes very few people. So the fact that you're asking it itself puts you kind of you know quite exceptional in the set of you know people thinking about how to uh so I'll give you one example then I'll give you one let me give you one example perfect perfect so how a lot of our uh a lot of our social programs are based on this BPL list below poverty line list right yep mhm and it's a list of people who are below some proxy line uh in one state I won't name the state the the big thing is most of the BPL

relies on the household head up household your household is a BPL household you'll get a ration card rsby came in and said no I need names of people in the households right got it um and I need names of people in the household and I need and who is eligible it's the head it's the spouse and it's up to three children right at the time we were doing it so then who is not eligible. A sister-in-law or a Sally is not eligible. Yep. Yep. So, I'm working in this state and I'm looking at these data and a huge proportion of women are coded as sis.

Now, that's going to be a problem because they are going to be ineligible, right? So, uh a lot of them are coded as sis. So then first they try and convince me that this is culturally appropriate for this state because a lot of sister but I'm like because somebody got to be missing in some other data you know what I'm saying like I can't you know I just can't do it like right yeah and right so no no but forget forget that so the whole story so then I'm in a then I was like so then the guy calls me at 11:00 problem solve okay so I said problem so then I figured out all the sals they had turned into other right some other relation fine I was like look this is not solving it let's try and figure out what happened so what was the reason why we had this right the reason we had this was way back when when the BPL lists were first

made and a huge census survey was done the state was in a rush to provide the BPL list so that things could be put out Right? Right? The data entry was contracted out to many many firms. Right? And what was really important for the data was the head of household and the assets they owned which would be used to classifies them as BPL or not who cares about the others. So what did the data entry guys do? They said yeah let's do the let's do what did the state do? They said let's do the header household in the assets. If it is a BPL household, let's put in all the names.

But if it's not a BPL household from our algorithm, then let's just put in Sally for all the other relations, right? Including children and spouse, right? Let's not waste more time, right? Let's not even fill in their data, right? Yeah. Uh fine. But then the Supreme Court said, "No, no, no. You have too few. You have to expand this. You have to think about, you know, there was huge protests. There was this, there was that." So then they expanded. But when they're expanding in the data, they're changing you from a no BPL to a BPL, but they're not going back and looking at what data they had entered.

So suddenly all these sies that they had thought would never be an issue and would never be used have suddenly starting to get used. So you start to see so it's crazy, right? I mean you're like why the hell am I diving into these details? Yeah. Right. And the point that I wanted to make was exactly this nan which is 90% of our work is that it's understanding the fine details between human behavior administration and what happens so that no one is excluded. excluded. Right? Because in that fine detail we would have excluded 30% of people who were eligible. Yeah. Yep. Right. Yep. And then it's also not enough to say that under the garb of trying to stick to a plan, you're effectively eliminating the very people that we were wanting to serve sometimes. Right.

That's that's exactly right. Right. So our job comes in as when we are designing data, when we are designing data flows, how do we take the human into the loop? Right? And the human is because the data are for humans. Mhm. and they're for people. It has no purpose outside the people that it's serving. serving. Right? Uh so that's really what I try and try and you know what we did a lot of in that program and we tried to think very hard about what where are the problems going to be and so what processes need to be put in place. Mhm. Mhm. To solve or at least be cognizant of these problems as they go along.

Yeah, I bet that. Yeah, it really does sound like every single day on something like this would have thrown a new challenge, fighting a new fire somewhere and it it it does sound a lot of fun, but also in like a really stressful way. Does this type of thing stress you out normally? Or when you were doing this, was that pretty stressful, would you say? I mean by the time you're finished doing it your you know uh your body apparently releases a lot of hormones which make you forget all the all the trauma but uh not completely. So yeah any of these projects right the when you're in it the amount of work is enormous not only because of the fact that there are problems and firefighting to be done but the fires are deliberately lit. Ah right. So it's not that you're firefighting things that were coming along. It's that now hospitals

are trying to make money. Insurance companies are trying to they are deliberately going to look at the scheme in a way that they can light fires. Wow. Yeah. I didn't even think about that. that. So it's anticipating the fires then dealing with the new ones that come up. Mhm. and then trying to figure out whether you need policy changes to handles the really really really long big ones. Yep. So I think something you said really has stuck to me and I've been thinking about it since I heard it which was that a lot of this type of work involves not solving technical data or sifting through it or utilizing it but people data and you said that you have to understand humans right at the heart of all of this. So I am curious based on your experiences like I know you you were collecting Ivy League degrees like

they were infinity stones from Cambridge and Harvard and such but when you're in India right obviously because that's your like roots and such. So I maybe I'm tempted to presume that you had a better understanding of what that type of arena looks like. And then when you went to the west and obviously you've done so many studies and groundbreaking research there. You've won so many awards for your work in the west like can you talk us through how if at all you had to adapt to learning about people from scratch right or is that not something that you had to do because people are really the same everywhere.

So I that's also a very difficult question. I I I think the place so you're [clears throat] totally right. I mean look I worked in India. I worked in Kenya. Whenever I work, I try and spend a lot of time in the field because I don't understand things very well unless I do. Uh and there's a lot of India that people haven't spent a lot of time in, right? Uh I spent a lot of time in different places. Pakistan I worked a lot on education. So I've you know wow wow been all over the country. Uh Kenya I worked a lot uh uh you know starting to work on Nigeria. Then this question of the west is interesting because I I'll tell you what the problem is. The problem is slightly I would say you put your finger on something very important and then let me tell you one way in which I would specify the importantness of that

which is yeah Nan when we are writing an academic article you have to think about who you're writing it for in some way right right correct yeah of course now if I'm writing it for an Indian journal I know roughly what my Indian audience is thinking thinking about or what they're thinking about right um if I'm writing for a you know all the top journals in economics the west journals are western journals right then I have to think about the problem is not that I don't know what's happening in India or I can't find out or if I do enough work and I do it it's that I have to translate what's happening in India into a set of questions Mhm. that the people in the west are interested in. Interesting. Okay. So, but then I have to figure out what the people in the west care about. Mhm. Mhm. Right. And the further away you are from

Manhattan, the less they care about what is important and the more they care about what's interesting. Right. Interesting. so important versus interesting. Okay. I So what is really golden is what is so what is really golden is something that's important and interesting. Mhm. Policy always privileges what is important. Academia always privileges what is interesting really in a way. And then what's really golden is the combination of the two. Correct. Mhm. So so uh uh so it's important to know that that you know it's very important to know that uh uh even though RSBY allowed for three children on average in the first year only 1.4 1.5 members per household were being enrolled. Right. Wow. Wow. Or or whatever it was 1.7 it's in the public domain. All these things are in the public domain. But that's not interesting, right? What may be

interesting is to say something along the lines of whether this was a behavioral constraint on the part of households, right? Sorry, I'm not following. What would that int as in like do they just not want to accept help? Is that what that refers to? So what the what they might find very interesting is an abstract of a paper that has that as a first line. Okay, got it. But then says you know something like we are able to show that an intervention uh that nudged households by giving them reminding them of the salience of health increased enrollment by 0.2 children. Right. Uhhuh. Super interesting because this whole issue of salience is very important.

Right. And interesting for the audience here. here. Mhm. completely unimportant for a policy maker in India who will be like dude if they're not then I'm just going to cut the contract for the guys who are not you know for the smart card provider and tell them to go enroll other people right uh ah okay I see I see what you mean now got it so they don't they care more about how the problem was solved versus the actual results is that another way to put it or or or that the the the bigger problem that I mean the the The issue is you always have to translate what is interesting for a western audience. And what might be interesting at this point is something about how people's brains function function rather than the fact of this insurance scheme not reaching the people it is supposed to. Uh-huh. Uh-huh. Right. As a first starting point. So

that part is tricky. That part is very difficult. difficult. I see. So you're almost similar to what I have to do with like my shorts and such. have to hook the the viewer to actually care [snorts] about what it is that you're doing. Is that an accurate way of I I think that's exactly an If you were producing a YouTube short for the US, you would have something completely different from what you you would have to find something that the US audience finds interesting. That's right. That's right. Correct.

Right. Whereas if you had it for India, you would have to find something that the Indian audience finds interesting. Totally. It needs to be localized in terms of interest. I guess I have a dumb question. Why must we find interesting stuff for the west? Is it cuz they fund these research? Is that why? Yeah, I think that is the biggest uh I I I feel that that is the biggest problem and hardest problem to solve which is at some level you don't right uh but it's a it it's it's I think about it as the fundamental knowledge entitlement of the west at the moment right which is we have to cater to what they find interesting then you could say t I don't want to do that right mhm mhm the problem we are finding now one is really hard to set up a system where you say we don't care about what the top people in the field want because they

are in the US but we are able to maintain our high quality that becomes really hard you say I only care about what the Indian audience cares about but at the same time the quality of my work is super high so what we are finding again and again is when you give up on that and you say no no no we don't need to publish in the western journals, the top journals, suddenly you find huge number of publications in predatory journals and this and that which are not very high quality right so it's been so it's a tricky problem it's a really tricky problem I see it's a really really tricky problem so you can say hey who cares about the western audience my podcast is reaching you know uh 200,000 people within India totally fine But it'll also be the case if the best producers, if the best quality podcasts are coming out from the US, Mhm.

Mhm. you will also want to say, hey, if my I can reach that audience, I'm going to put it's a different level of quality that I need to cater to. That makes total sense. Yeah, I think you did such a great job of making sure that I stuck with you. And I just want to call out here that me along with a lot of my audience will be just complete newbies to this field. So I really appreciate you breaking this down for us in a way that's you know so accessible.

Um but just to take that uh podcast an analogy just one step further. Do you think that's changing at all? Cuz we have Raj Shamani now. So is that something that's happening uh in like the policy field as well where maybe we have some more higher quality players overall or organizations that are starting to break this wheel that we just talked about or not yet. Yeah. number one I I'll tell you what the issue is like if I think about the US right so the big issue is like in if you're working on the US you can publish important or interesting right so for example I'll give you one idea uh one one example huge finding big paper on the difference between blacks and whites in test scores in grade four right got it this is a top journal paper right It's not going to be a top journal paper to say there's a huge difference

in test scores across different casts or different religions in India. That's not going to be a top journal paper here. It'll be a top journal paper in India. That's right. Makes sense. Right. Yep. Right. Mhm. So then what happens is in the US you get a huge base of knowledge and then you get a pyramid moving to the top. Smaller and smaller papers. So say a thousand papers on what's happening on Tescos so papers on something that and then five top right right we are getting very good top quality papers in policy in other countries we are not getting the base so a lot of basic interesting facts are not making it out there you know I'll give you one uh average size of a primary school in India now uh a public primary school take a guess like in terms Student headcount we're talking. talking. Yeah. Yeah. I want to say

let's say a specific state say Telangana. Telangana. Talanga now. Um and this is like kindergarten through five. That's primary school. K through five. Yeah. Yeah. Um I want to say 500. I don't know. know. Right answer is 40. Wow. God. The biggest single problem in our schooling right now is 30% of public schools have fewer than 30 children. Wow. I that Wow. I'm Okay. So that's just a fact. Now important but totally uninteresting.

No, wait. How can you say that? That's so interesting. It's interesting for you as an Indian. You see what I'm saying? Right. Right. That totally makes sense. Yeah. They haven't even heard of Telangana. Why would they care? Right. I wrote Manhattan has this huge problem, which it does. It just came out in the New York Times yesterday. There was a New York Times article saying, you know, lots of school districts are facing small class sizes, right? Yeah. Nobody's writing a New York Times article on lots of school districts are facing small class sizes here, right?

Now, the issue is somebody should have written that. In fact, Gita Gandhi has a paper on looking across as the US at India. But it's a fact that everybody in India should know. Correct. I agree with I mean at least people doing policy on education and we don't don't right you add in you know the fertility declines so fra you know Telangana fertility rate is 1.5 mh mh right which is you know similar to lots of European countries mhm mhm right we are going to hit very very small classes at schools very soon this should be on top of everybody's mind but the fact is not there right is that a India specific problem I don't know here's another one fraction of American adults who are functionally illiterate functionally illiterate if they get a prescription that says take this medicine 2 days a two times a day for a

week they don't know how many times a day they need to take it if they get a notice saying get make sure your child is in school by 9:00 a.m. They don't know when to send their kid. You know that level. What fraction of adults do you think are functionally illiterate in the US? I'm going to say 28%. You're absolutely right. Right. So, one out of every three adults in the US. I I I saw this headline a few hours ago. That's the only reason why I guess that Yeah. Yeah. You know, but most people don't know that if you go to a restaurant, one out of four American adults can't read the menu. Right. Now the thing is like you know some of these facts will be hard to bring out in a society and make it acceptable acceptable right but a lot of our work should be just presenting basic descriptive facts right about you know a basic descriptive

facts about yeah what exactly are the basic details of our country correct totally makes sense uh uh Uh does that make uh uh does that uh uh Uh does that make sense? Yeah. Yeah. Totally. I guess so naturally and this is very closely related to the theme of my show. So if there is a young policy worker slash in your domain listening to this right now, maybe they're in India. What is something that they can do to fix exactly this very important problem that you just, you know, talked about just now? Is it just marketing? Is that what this is? poof. Oh, Oh, like I'll talk you through like a really quick overview of what I'm thinking. We have like this creator, we'll say, right? Their entire brand is centered around very short, punchy, informational, accurate YouTube videos about this type of thing. It we get it

really high production value. People love to watch it. It's presented in an entertaining way. In 6 months that it racks up like millions of views and we have at least in part solved some of this problem. Would that work? roast this idea for me. Yeah. I I feel that you know uh uh I feel feel I feel this is a place that we don't know right and I'll tell you why I think that and I try and do as much as I can. Uh I'll tell you three or four pieces of it and then you tell me what the solution is. Love it. Let's do it. So, so first of all, so I like reading math popular books and yo and whatnot, right? I like reading science books and stuff and one famous math and science, you know, people who do translation, it's called quant math and, you know, just look it up QA NA any article. It's great fun to read, right? I mean, sometimes I don't

understand them. They're too high-fi for me, you know? Sometimes I understand them, but they're fun to read, right? Interesting. Interesting. Then you look at who are the authors and they're all PhDs in math from Stanford, Chicago, Harvard, right? Who have then gone and done another degree in research translation, translation, right? Oh, okay. So, there's a degree called research translation for science and that's wild. I had no idea that crazy. We don't have it for for policy and economy, right? The reason we don't have it is because we don't produce enough, right? right? So at the end of the day, whatever like every day there are 500 new math and science papers coming out, there's enough to keep translators occupied, right? Correct. And they play a really important role, right? You know, in policy and econ,

it's slow moving, right? In education, everything we have learned, I could do in 20 podcasts with you. over everything we have learned over the last two day two 25 years I could do in 20 podcasts with you it's not enough to sustain these translators so then we get two problems a problem we get is so eighth problem we get is we don't get the professional translators the second issue we get is then people say say yeah you know when you do a podcast keep it simple yeah and I'm like no no no No, no, no. My idea is not to keep it simple. It's to keep it no more simple than it needs to be, right? I should not I should never oversimplify a problem.

And if I'm not being able to because none of this is rocket science. I mean, compared to the math these guys are presenting, none of this is rocket science, right? It means I haven't understood the problem enough or I haven't figured out a way to convey to you, you, right? That you can actually understand it. So I think this idea that we should oversimplify and dumb down things is just killing us. I think that's a bad idea.

Right. I think we should respect our audience. audience. I think we should expect our audience to do some of the work in following us along. We shouldn't dumb it down. There's no reason to think my audience is dumb. My audience is really smart on average. Right. So make the effort to make sure that we're doing that. Right. The third problem we have is increasingly donors have always done this but increasingly it's spreading. People want impact right. Y for an academic I think it's a disaster.

Uh for the following reason how so how do you want to think about impact right after all I now you know I was fortunate to go to a college that's fairly well recognized in India. I actually know lots of people who are you know at the top policy levels right impact I can go and talk to them I don't need to break my back doing research right right uh I can become a consultant right so if you're saying I'm going to measure you on impact right as they would say in Hindi you know I just need to go and have a drink with somebody have a dinner with somebody and you know people are asking they're saying what should we do about this I have this what should we do about this I have a rule that says I never talk to any policy maker unless I'm talking about something I put in the public space because I feel very strongly that

anything that researchers should be doing should be strengthening democracy and that when we talk to policy makers outside the space of published or publicly available material we're actually undermining democracy because after all I'm using my networks and contacts. Mhm. Right. To undermine a broader democratic conversation about something that's really important. Almost like insider trading in a way. Almost. Almost. Right. So I almost feel that idea that we should go after policy champions or we should be able to whisper in the policy champions ear is a completely bad idea. It's a horrible idea and I think it puts us in a very difficult position. Y uh so the way I like to think about it is of course we should go out and we should do podcasts and we should talk to policy makers and all of that but it's

an iceberg and what's below the surface is our research what's on top is my conversation with policy makers but without that you know iceberg model all that I have is free floating ice on top that's going left at some stage right at some stage north at some stage south at some stage whichever way the funding winds blow. Mhm. Mhm. Right. So I don't know. So so my view is very different. My view nan is we have to democratically strengthen policy.

We have to take people seriously. So I've been trying something very different which is I take my research findings and go back and present them in the slums, present them with the people we did the research with, present them in the villages and say look it's part of democracy. You should know what we found. Right. Right. Right. Uh And I don't you know you tell me right how will you make sure so I I and I think that idea key you know and I I take inspiration from people like Ambedkar people like you know Ambedkar had two PhDs right he goes back to India but he has no problem giving speech after speech to people from villages people from you know and I don't get that we gave up on that that right but you guys are the intermediaries and You have to think about how do I get this to the people right for whom it really matters.

Mhm. Without treating them like sheep, you know. Exactly. These are these are people who are very sophisticated. They are the people who are making the tough decisions respect that we give them that respect to say you are the person who's sending your kid to school. You're making a very difficult decision. Let me have you know let me tell you what I've learned. I might be wrong. I might not have understood things but let's have a conversation. conversation. Yeah, you know it's so interesting. I was recently talking to one of my filmmaker friends here in Chicago and I never thought that public policy would be connected to film making in such a direct way but he basically word for word in obviously a completely different context shared your sentiment around you know movies as they are made in in the west versus in India where in the west

they assume a lot of the times that we're going to go there we're going to show something really nuanced that's very easy to miss but that has a very important part in the story and they trust their audiences to make sure that they're paying attention they're smart enough to figure that out and ultimately that makes it such a more enjoyable movie experience to watch which not to say that all western movies are like this and all Indian movies are not like this but for the most part he had this sense that and he is a filmmaker so he understands movies way better than I do but he was like in India we almost always spoon feed our audiences that hey did you are you sure you get this one part? Okay, never mind. We'll we'll share it again in 3 minutes and then again and then suddenly it's just not even that enjoyable. So

obviously completely different context but I just felt like it would be it would make sense to share this in in light of what you just said. As far as the what do we do here, right? I feel like what I'm hearing is we just need to do way more research at the ground level. Would you mostly agree that that is the direction? Obviously, that comes with its own set of challenges, right? Where is the money coming from? Who is funding all of these studies? But directionally, would you say that is where we need to go? I I absolutely I mean, I think we need an enormous infrastructure and a real thinking about how is it that we want to make sure that basic facts, right? So I I I'll give you one more example and then you can you know how do we make sure so so there's a friend of mine who works on you know who works in Delhi okay

okay right so I was asking her what do you think the top 5% of people earn right in uh you know in say uh in in terms of per capita you know expenditures or or incomes close to that in urban India again take a Yes. Unfortunately, I did not see this headline a few hours ago. So, this will be wrong. I don't know. You said top 5%, right? Top 5%. 100 crores. I actually have no idea. Top 5% of urban urban families, how much do you think they earn a month?

Okay. I see. I see. Earn a month. I'm going to say 10 lakh rupees. The answer is 20,000. Top 5%. The answer is 20,000. And in rural areas, it's 10,000. Wait, I feel like I'm missing something here. When you say per capita, that means so there's a family of four people, husband, wife, two kids. Huh? You're saying of the entire Indian urban population, right? The average average of top 5% is Wow. Sorry, please continue. That's crazy, right? The bottom 5% is Dhazar Rupai.

Yeah, these are numbers we you know but what do we think? We think that because the person who is at our house and who we are paying 20,000 we would never think about them as in the top 5% of of incomes. incomes. Yeah. I'm sorry. I'm actually curious to ask you, sorry to interject, but when you ask people these questions, I'm sure it sounds like you do. Was I like way off or have I been kind of the template in terms of what people usually say? Cuz I'm very curious to know actually. Yeah, I this is the problem which is people are all over the place. If you ask somebody who's poor, they have a very good idea. That's so true. That is such a good point. Sorry, please continue. But yeah, I love that somebody who's rich, then so I you know, so it's not I don't know what's going on. Yeah. Yeah. Right. But it's the same like

interestingly it's the same problem in the US, right? I mean very few people know that 50% almost half the population cannot afford $400 for a medical emergency. emergency. I didn't know that. People don't know these things, right? People don't know 30% of American adults are functionally illiterate, right? So why is it that we are creating societies, right? Uh where everybody knows the latest Tik Tok trend, right? Uh but we don't know basic things about ourselves. And yeah, cheese, it's on you guys. No, no, I'm serious because you have to think very hard about how you're going to change this because yeah, in this day and age with chat GPT, right? Let's try, you know, just open up chat GPD whatever and say, you know, just ask a question, right? Let's just see how much just say what is the top 5% richest richest uh household

expenditures in India in urban could be let's do it real quick let's do it okay so question is what is so let let's do it sim simultaneously uh how much do how how how much do the top no what is the so Try this. What is the uh top 5% household expenditure? Mhm. Rural urban in rural urban India. NSS. NSS is our national sample survey. Okay. Let's see what it says. I may be totally wrong. In which case Yeah. So I guess and while it searches NSS is what exactly? Sorry I'm not fully NSS is a national sample survey. Uh okay which is the one that does all of this work and this was you know I think you it should hopefully report report the 23 24 numbers but they survey almost two lakh 60,000 households in urban and rural areas to get these numbers.

So rural 10k, urban 20k uh and then very roughly a four person household at that average would imply rural top 5% 40k and urban top 5% 80k. Yeah. Is this Yeah. So per person is 20,000, right? Yeah. Mhm. Right. So that's about $200, right? Per person. Yeah. True. uh uh which is which is what we are talking about right right right u you know try median and you'll find it's much lower median in India will be about 10,000 or so right for sure that tracks yeah uh so so this is not a hard thing to figure out right now my question is how do we make sure that this set of basic facts so the kind of policy I do is you know it's interesting it's you know some of it is basic facts. Some of it we've been trying to do a lot of that, right? Say what are the basic descriptive facts. Correct. Correct. And you'll find that it's very hard. Try

and do it. The wall of opposition you'll run into on many many things will be very interesting. Like in terms of people that are just hurt with this stuff. Is is that what you you I don't know. just try and put out something that says did you know right that even the richest 5% of urban India earns less than what the US did in 1850 or something you know I don't know what exactly that number will be yeah no that makes sense I see what you mean mean and you I I I suspect you will get an enormous amount of push back true no that totally tracks and your point is that that shouldn't be the case. We should be okay with looking into the mirror. My point is, no, my point is in every place there are different things that bring that consensus together. In the US, interestingly, it's John Stewart, it's Joe Rogan, right? It's Stephen Colberg, right?

Because if you actually watch the shows, the amount of data they put forth is just googling. Yeah, they're just googling throughout the show. phenomenal the amount of data that they put forward in a show. I mean, look at this guy Joe Rogan, right? I don't listen to his podcast. I don't know the guy very well. He's a top podcast and his things are, you know, you may disagree with him on lots of things, but he's not dumbing things down all the time.

I see what you mean. Yeah, that makes sense. Yeah. No, this has given me so much to think about and I'm sure even folks that are in the field currently doing this type of work, this is also I'm sure really helpful because yeah, I think people don't talk about this enough like we just see all of the highlight reels, right? But we don't see what's behind all of that and what it took to get there ultimately. And then what in addition to that what you just said um how do we make sustained change in terms of like this is how it is but it doesn't have to be this way. So how do we change that? And what I'm taking away from our chat here is it's through doing the work. It's through doing the research, presenting it for the people that are impacted by the research, not shoving it down in shapes that are easy to be accepted. It's okay if it has if

it's malformed, if it has weird shapes. Those nuances matter and we should not just hide them just because they're, you know, uncomfortable or because we think often incorrectly that people won't follow along. We need to trust people more in terms of accepting and engage in that conversation. So if somebody doesn't follow the right way to think about it is what did I do wrong that I couldn't explain it. Mhm. Totally. And that I think also brings down the barriers in terms of a me versus them, right? Then it becomes a we, right? Cuz we're all in this together. together. So wow, that's so cool. And yeah, I guess you know, final question here before I let you go. What is something that you're working on currently that really excites you? Do you have any you know really cool projects that you're cooking with right now?

I think the one that you can talk about at least. Yeah, of course. No, the one I'm most excited about you know in Pakistan we have put together the longest standing cohort study study I did see that it's like 10 years right or 20 years. 20 years 20 years for almost 20 years we've been following 5,000 children. And just to give you an idea, we started off in 112 villages and in our last round in 2018 2019, they had gone to 750 locations. So we tracked them down everywhere. everywhere. Some cases we waited 3 years until they came back from the Gulf or whatever, right? You know, for Eid and things. It's worked right. right. So what we are seeing now, what we're trying to do, Nan, is to try and see, we know everything about how they went to school, what happened during the school years. Mhm. And we're trying to see whether the

children who did better in school in primary school. And then there's this all this question about correlation versus causality which we are very careful about. That's the bulk of the work once the data has been collected. We are trying to see whether children whose test scores were higher in school are now earning more. Right. Right. Ah okay. Got it. And what we are starting to find is in fact interventions that improve test score in primary school and not even by that much maybe the best development interventions that there are uh and what's fascinating for me is India through its Nepun Bharat mission has actually doubled down on that okay okay I think it's doing that in absolutely the right direction we'll see how how it works out whereas in the west there's they're like giving up on it. So, you know,

know, and and it being helping primary students go to school more. Is is that helping improving the test scores of of children in primary schools? I see. Yeah. So, in the west, literacy is coming down. down. Correct. Yeah. People don't realize this, but maybe the literacy in in the west in 2014 was much higher than it is today. Right. Whereas India has really doubled down and said you know with this Nepun Bharat mission they have really doubled down and said we need to make sure that children are learning in school right right now I I think in some places it's happening some it's not it's a work in progress but it's the right direction and what we are now getting is the evidence to back that up right so we are finding that that call is absolutely the right call for this day and and that I'm really really excited

about. So, we're hoping to have a paper in the public domain over the summer. Cool. Awesome. That is so exciting. We'll see how that goes. Yeah, that I mean, I can't even wrap my head around sustaining one coherent project for, you know, 2 months, let alone 20 years. There is nothing coherent. This is all about, you know, serendipity, piece by piece, saying we'll never do this survey again. and then 5 years later saying, "Okay, maybe we should, you know, and then just, you know, people just egging each other on to do something." something." Wow. No, I mean, that is all I just feel like you've opened a door to a universe for me and I'm sure for a lot of our listeners that straight up did not even exist for us. And I actually had no idea this would be something I would feel so interested in. And I think I really want

to commend you for, you know, cuz some people when they're where you are in this field, they can be a little bit of gatekeepers, right? They'll be like, "It's not for you. You won't get it. Move along or whatever, you know, as you said, which is what you stand against dumbing it down is what some people might be tempted to do with a a situation like this. But I really appreciate and I want to thank you for not doing that, for taking us through the the bad and the ugly along with the good. So truly appreciate your time here today. This has been such a delight, professor. And maybe once your paper is out, maybe we can do a quick debrief on that. Uh if you'd like obviously to, you know, come back on the show, but No, no, I'd love to. Uh I'd love to. And so you know really my aim is to say there's nothing rocket sciency about this.

Yeah. Right. If you go to quantum magazine and you see the kind of math they are explaining it is rocket sciency and they can explain it to somebody like me who knows very little about math. If I can't explain policy and econ to you I'm doing something really badly. That's true. Right. So when it comes down to you know saying that it's really my fault that I haven't been able to do it well. it's it's it's not yours and then we have to work harder at it. Right. So I I have found that whenever we take the trouble, people always meet us halfway. They are willing to put [clears throat] in the work as long as we are pulling willing to put in the work. Totally. Yeah. It's like when somebody reaches out to you with a hand, you you you just have to grab it, right? How do you not? And it's just how humans work. So

yeah, except for us in academia, we are the ones in the well. So when we are reaching out with a hand we saying please pull us up right and we rely on people like you to pull us up I mean really because it's a pleasure always to speak with and understand whether our ideas are making it through right so that's really our litmus test and and you know I'm always delighted if we can explain something in a way that makes it to a broader audience