Event: Lessons in State Capacity from Delhi’s Schools

Salimah Samji Welcome to the Building State Capability talk series. I am Salimah Samji, and I’m the executive director of Building State Capability. I am really excited to welcome Yamini Aiyar today. Yamini is a senior fellow at the Saxena Center for Contemporary South Asia and the Watson Institute at Brown University. She is also formerly the president of the Center for Policy Research in Delhi. She comes to us today to talk about her really great Oxford University Press book on Delhi schools. Welcome, Yamini.

Yamini Aiyar Thank you, Salimah. It’s such a pleasure to be here and especially to be in conversation with you. The bit that Salimah didn’t tell you is that we are, I first, I think my second job when Salimah had just come out from the Kennedy School in Delhi, we were office mates at the World Bank and have continued to be friends and collaborators since. So I can’t be more excited than to have a chance to talk about my work with you because it all started with you too.

Salimah Samji We definitely have a history. Wonderful. So what we’re gonna do in today’s talk is I’m gonna have a conversation with Yamini and then we’ll open up to Q&A. There are two mics on the side. When you ask your question, please make sure it ends with a question mark and keep it short, but also introduce yourself, tell us what program, what school you’re at, and that would be really helpful. So without further ado, this book is on education, right? So, in 2005, Pratham’s status of education report came out and showed that 50% and has happened for a decade after that and then every two years after that. And the consistent result has been that children can’t read. Like basic foundational learning is just not there. Fifth graders cannot read second grade sentences. And then, you know, the NEP 2020 report also showed, this is like many years later, that 50 million children in the country are lacking foundational literacy and numeracy, right? So it’s in this backdrop of education, your work before Delhi is happening in Bihar. So I was wondering if you can take us to Bihar in 2010. And you know, this is. In 2005, Bihar had a new chief minister, Nitish Kumar, who came with a big promise on the policy of education, really wanted to strengthen education in his state. It’s been five years, lots of experiments have happened, lots of learning happening in education in Bihar, and you are in Bihar. Why don’t you start us off there?

Yamini Aiyar Thank you. So I think what was just an important kind of addition to what you’re saying, from the late 1990s onwards, the Indian government, particularly at the national level, began for the first time really to embark on a policy agenda for mass education. So the real puzzle in India has always been that. For a relatively, for a new and very, very poor nation, the choice at the national level was to invest much more in building higher education institutions, particularly the Indian Institutes of Technology and Indian Institutions of Management as two good examples, but didn’t embark on a national effort for school education, mass education, and even basic literacy. Of course, India is many countries in one, different states have done different things. So there were parts of India, education is a state subject, school education was a state subject and then kind of became a shared concurrent subject between a state and center. And some states made a lot of rapid progress, but at a national scale, we hadn’t yet moved in this direction. And that big step takes place in the 1990s. But it kind of followed the global narrative, which was access. So the main emphasis was on building out the infrastructure of school education system. So literally the school starting from scratch, school buildings to hiring teachers and bringing schools to children to ensure that every Indian had access to a government public schooling system. This was, of course, checkered in many parts of the country, not least of all the state of Bihar, which is one of the poorest states in the country. And it’s funny we’re talking about Bihar today because it goes to the polls tomorrow for the first round of its elections. Nitish Kumar, the Chief Minister from 2005, stays on as Chief Minister in 2025, so it’s been 20 years in the process. He has allied with every political party that there possibly can be. And the shine that he had back in 2005 has diminished significantly even though he still remains chief minister of the state. So it’s also a story of kind of the shifting nature of Indian politics. Coming out of in the 1990s as the government of India began to pump in more resources into building the education infrastructure, Bihar was broadly considered to be a significantly laggard state. It didn’t do much. In fact, the big interesting puzzle there was why is it that despite the government of India offering money to the state and other parts of the country, particularly poorer states that were dependent on the government of India for resources were busy building schools and hiring teachers, Bihar was doing nothing. Just an interesting nugget. One of the reasons for this has to do with the peculiarities of the nature of social mobilization in Bihar in the 1990s. Bihara was one of the states that was really the home of lower caste political mobilization. And when the chief minister at the time and the political party came to power, the big challenge they had to face was in trying to disrupt the upper caste dominance that was the bureaucracy. The education system was putting in money and had certain rules on teacher qualifications. Given the nature of the caste system it was largely more likely that upper caste teachers would be qualified to come into the education system and absent a kind of sincere political vision the easy choice was just not to hire. So in some senses, one of the most powerful elements of Indian democracy, its ability to really give access to the most marginalized and discriminated sections of society through democracy, access to state power, also created a kind of complex notion, impact on the process of development. Teachers weren’t hired. Schools weren’t built. Do nothing was the approach more than anything else. So when Nitish Kumar comes in, the nature of his social coalition, he represents one other element as part of the caste structure, but also came in partnership with the BJP government that is now at the head in Delhi, and they represent the upper caste. So that social coalition enabled Nitish Kumar to do a lot more. And he made education one of his big planks. And the first thing he had to do was to build the infrastructure for education. That’s why I went into this long story, not just because it’s interesting, but also because it gave the context of where we were. In 2010 in Bihar, a lot of very serious, so there was all this money that was sitting in the center that the previous regime had not used, they were then able to access this money and start building out, so they built the schools, they hired the teachers, they had one of the most interesting and rapid successes in reducing the percentage of out-of-school girl children through a lot innovation that was big, innovative mobilization efforts, including a very important scheme called the Mukhyamantri Cycle Yojna, which was essentially handing out bicycles to young girls who finished Middle school so they could go to secondary school so that they would have access to schools and in fact this has kind of become in all parts of North India a big scheme that got replicated from those from the Bihar experiment. So the first foundational step had been built. And in many other sectors too, you know, from no governance to little governance movement had happened. And the challenge that confronted Nitish Kumar when he comes back in his second term, now with a landslide majority and a lot more expectation, was to take the system forward from there because we had learned again and again, it was visible in Bihar too, that building the infrastructure for education is only the first step. Ensuring that students come is an important second step, but the second and a half part of that is to ensure that there’s teaching and learning, right? Like it’s great to have the building, it’s good to have a teacher, hopefully the teacher is showing up occasionally, but then what happens? And we learned again, and the other report that you mentioned is again really important because it played a really significant role and it took a long time, the better part of 10, 15 years of serious advocacy and coalition building. To shift the policy conversation away from the challenge just of infrastructure to recognizing that infrastructure while necessary is not sufficient. And we need to think about what is happening inside the classroom. So in 2010, a lot of opportunity opens up for experimentation. And like many chief ministers, and I’m sure we’ll talk about it at some point in the course of our conversation, one of the, you know, it’s both good and bad, but, doers, when new political parties come in and you have these doer chief ministers who want to reform, they open up for a lot of nonprofit organizations and academics, a whole range, the whole self-development world to show up. I used to joke with colleagues in 2010. That if you wanted a job in international development, all you had to do was to just stand in Patna airport, you would get picked up. Somebody or the other was from J-PAL to what was then DFID, to the world. Everyone was there, as were we. It was an exciting place to be because there was so much happening. And Pratham, the NGO that is the mothership of the ASAR report, was working with the state government, doing a lot of experimentation around this question of learning. And their basic framework was this. The structure of the classroom is around age and grade. So if you are seven, you are in class two. If you are eight, you move on from there. But the nature of how children access the education system and the way in which teaching and learning takes place or doesn’t take place means that by and large, and many of you have probably read and studied a lot of the RCTs that have brought this out. In a big way, the classroom itself has a very multiple and dynamic structure of learning. You will have the first two rows that are close to curriculum level and the last two rows that can barely recognize letters. And so they were trying what they now call teaching at the right level, forms of remedial training, which was to group students by their learning level and it helped them try and catch up to where the curriculum expects you to be. Lant’s paper, which always has a nice title so you don’t need to read the paper, but you must since you’re students, but he kind of summarized it really well in the title of a paper that said the negative consequences of an over-ambitious curriculum. The curriculum is here, the classroom is here in lots of complicated multiple ways, and the first two rows that the teachers teaching kind of are nodding their heads and taking their notes and copying from the blackboard and everybody else is lost because the system is designed to teach to the first row is not to the back of the classroom. And this was the kind of theory of change that can we bring everybody closer to where curriculum level is expected? And in fact, the challenge is not even the curriculum. It’s basic things, students sitting in class five can barely recognize letters or recognize numbers, forget doing basic levels of arithmetic. So an RCTU accompanied one of the experiments, which was around a summer camp. And the summer camp took students, grouped them by level, trained government school teachers with all with a kind of more interactive type of pedagogy not the boring stuff there many Indians here they’ve all done probably CBSE ICSE so we all in the same game we know how to memorize but our critical and conceptual learning happens outside of the structure of the school if it happens at all I’m sure it’s happened for all of you because you’re here. I can’t say I learned anything, but I can memorize really well. And so it goes. So the teachers were given these alternative pedagogical materials. And in the summer camp, the RCTs found that there was a significant improvement in foundational literacy among students that attended the camp. Now the students go back into school because ultimately the summer holidays have to end and the teachers go back to school and the second part of the experiment was to see how much of this gets incorporated in the structure of the classroom and how much of this actually rolls through not just the students that attended the summer camps but more broadly for the classroom. And they found that there was absolutely no progress, right? This is an RCT that I’m sure many of you would have, it’s probably part of your classroom or work here as well. And I was, and like all good academic publications, the paper gets published 10 years later. But Pratham was smart enough to say, this is really useful for us to have a serious evidence-based conversation with a group of policymakers. So, and we don’t care when AER actually goes through its review and publishes it, let’s take this. And Michael Walton, who used to teach here, was then with me actually at the Center for Policy Research and he was involved in this particular RCT. So he, me, Rukmini Banerjee, Pratham, we all landed up in Patna to share the findings. I was just a fly on the wall, observing what was going on. And we walked into this room. The education minister, newly elected, is sitting there. And the chief secretary, one of the most important, the big boss, the most-important IS officer was sitting in the room. The education secretary, and sundry others. And it’s very rare to walk into a government office and have such a serious senior group of people sitting in the room. And they looked at Michael and said, Doctor, because, you know, give us the diagnosis and give us the medicines also. This is Bihar. Everybody has exciting ways of telling a story. So Michael then kind of sits up and then and puts on his PowerPoint presentation another 10 minutes go, because no one could figure out how to put the huge TV that had just newly arrived in the Bihari government, just to also give you a sense of how it had kind of been upgrading itself to more 21st century. Finally, somebody showed up, the TV comes on and Michael starts making his presentation. You know, we go through the numbers and they were not surprised by anything that we saw and the conversation really starts, really begins. And this, you know, if we started at five by now, it’s like eight o’clock at night. And we had all taken a flight from Delhi early in the morning and had been planning a really good dinner. But dinner time was passing by. And finally, we left there at ten only because we were exhausted, not because they were tired. They wanted to continue the conversation, but at the heart of it was the critical question that motivated the work I did since, but I think also is the biggest challenge that confronts our debates on state capacity. We know how to do the mission. So we know how do the summer camps. This is the chief secretary at some point just had heard everything and he was just like, okay, look, we know to do this in a mission mode in a summer camp, we can align everything, we can mobilize. It is when it goes back into the classroom, when it becomes routine. That the whole thing breaks down. And really the challenge that they asked the doctor, whose medicines were really about, you know, take this pill for seven days and all will be well. But they were asking for a long term, you know, good health, go from the allopathic medicine to Ayurveda and stop eating the chocolates and be healthy. And that’s really where the whole thing sort of hit. And there were two important things to come out of this. It’s very easy and there is also evidence and truth to it, so it makes it even easier to legitimize, the view that we can’t get things done because the teachers are incapable or they are not accountable, they don’t show up. Lant has this wonderful paper on the flailing state right like the head knows what to do but the arms and legs not able to pass it down all the way to the sinews and in our imagination in fact in our lived experience and how we talk about the state, you know, the devil is always the guy on the front lines the teacher who’s not showing up absenteeism rates indeed are high the doctor who shows up only rarely and when he does there’s a 50 percent chance that you’ll get diagnosed for diarrhea or you’ll get over-medicated. The block development officer who wants bribes before they put you into the list for benefits, and so on and so forth. And I thought in this conversation, something else was going on too. All of this is true. And we often think of our solutions to state capacity from that point of view. But here’s the same teacher who doesn’t show up and is lazy and is apathetic, but is also being able to do something different given an opportunity. And then she goes back to the system and somehow things don’t happen. So is this the flailing state story alone or is the flailing state story just about the frontline that is corrupt and doesn’t show up and doesn’t want to do things or is there something else about the structure of the system that is making the same actor behave very differently in two different settings. And that really for me became the real puzzle. So when we talk about state capacity in India Is it just about three kind of things you need to do to monitor better, to shift the incentives for the frontline by using carrot and sticks kind of methods that we all know and are familiar with or is there something more systemic over here that we need to get to the heart of? So Bihar kind of by the way now is in a somewhat depressing situation in the sense that all the the excitement of those days no longer resonates in today’s Bihar. And I think part of the problem was precisely this, that you can build, you can go from zero to one, but to go from one to 10, you have to really get to the heart of the systemic things. And there are many parts of that systemic thing, including social inequality, caste, all of these are crucial elements that shape how citizen state relations play out. But there’s one bit that we don’t talk about at all, which is the black box of the system of the bureaucracy itself. It is spoken about largely in the context of the ways of the IAS, Vishnu is here. He’ll tell you much more about that. But very rarely, he’s one of our shining IAS office stars of the Indian Administrative Service, but very little is spoken about all the others who really are what we as citizens encounter on a daily basis, the teachers, the doctors, the block development officers. What is the context in which they operate and why is it that they only function well when you exit them out from the system and fail when you put them back in?

Salimah Samji Fascinating thing that you realize that the same person behaves very differently in two different settings and there’s something about the system that’s not allowing these gains that an RCT, gold standard, tells you that yes, this is an effective strategy, but when you try to do it within your larger system, it doesn’t happen. So fast forward to Delhi 2016. You now have the opportunity, new government again, this time it’s Delhi, and key focus on foundational learning. And this is when you embark, which is kind of the bulk of the book of this approach. You used ethnographic study as a lens to look at this. Can you share that study design? Why choose that as a way to study and better understand this thing that you had already seen in Bihar?

Yamini Aiyar So, you know, at the heart of this question that the Bihar education bureaucracy and policymakers were asking was a question that emerges quite routinely even in our conversations about RCTs which is how do you embed change in a long-term sustainable way? And the answer often is, you can’t embed change because the incentives are not aligned, the political will disappears, the bureaucrat shifts and it all falls back to square one. And in some ways, you could argue, I mean, of course there’s a lot of debate about what’s the right technocratic set of policy, instruments you need to address this problem of learning, that debate continues, it’s a live one, it is a dynamic one, and it’s an important one. But over the years, the RCTs had actually built up a fairly robust evidence base to tell us that when it comes to the challenge of foundational literacy numeracy, moving a child from not being able to recognize letters, to recognizing letters, to being able form a full sentence, to being able read a paragraph, to having basic comprehension in a short period of time. Teaching at the right level works. You can see by the shifts. So in a way, teaching at right level gives you at least the beginnings of some type of evidence base. And in every single experiment and over the 10 years, the Pratham J-PAL partnership is really exciting because of the way in which it uses, Prathom uses the evidence. To innovate each cycle round and in each point there’s a new RCT that tells you what’s happening, right. So one of the big learnings from Bihar was you can get things done outside of school what do you do inside the school so then there was another experiment in Uttarakhand that was just with training in Haryana there were there was some efforts for training that included the mid-level of the bureaucracy and so on and so forth and every level you get this kind of range of evidence. But at the end of each of these stories was the same question, which is, as one IAS officer put it, India is like this burial ground of pilots. At one education meeting, I was sitting. He made this kind of grand statement, and as IAS officers do, left the room. But I thought, ooh, nice little quote. I’m going to use it at some point and made a note of it. So there is great value to these grand statements, too. And so that was kind of on my mind. We went back into Bihar in 2012. When one district in Bihar called Jehanabad had begun to try to entrench this teaching at the right level through more active training of the mid-level bureaucrat, grassroots level bureaucrat or mid-management bureaucracy of Bihar. There was a position called CRCC, Cluster Resource Center Coordinator so they were doing a lot of active training on the CRCC with the district magistrate, the IAS officer, playing a very important role. And then they scaled it up in the state. But by the time we got to see what was going on, the program had disappeared. And the standard story was political will, Nitish Kumar, who had kind of been spearheading it through the chief minister’s office. Politics had come in the way. In fact, he had resigned from his chief minister post. And so, you know, while hearing these stories, I was kind of thinking to myself, The evidence base and the rigorous quantitative evaluation techniques are not going to be able to get me to understand the heart of what’s actually happening inside of the bureaucracy. When we went and spent some time in Jehanabad, conversations with the CRCCs and the BDOs were beginning to open up something new for me. In two ways, I’ll just give you two quick anecdotes. One was: Just in talking to the frontline management. And then we went around to different parts of the country and had similar conversations. There was, you know, as the conversation would move forward from the problem to the solution, I often would ask, so what do you think, you know you can do about all? Because of course everyone has an opinion. I mean, more than an opinion, everybody has an opinion and is also able to identify where the source of the problem is and the source the problem never you, it’s always somebody else. And in the course of this kind of response, they said, and I open the book with this story actually, which gets repeated everywhere. If the government wants, they can do lots of things. Now, you know, it’s a nice statement to make. If anybody above me wants to do lots of things, everything can change, those with power. But you’re looking at this frontline officer, and they are in the state partly because of the power of the state. And they excude power in every way, right? Like they’ll show up like three hours late, the Ray-Bans will come off, three minions are carrying their bags, someone with their lunchbox, someone with the briefcase, someone will their mobile phone, they’ll sit down, they’ll press the bell, three more people will come, some orders will be given, I mean it’s all and you know the citizenry who’s hanging around are told to please leave the room because you know, the boss is here and then the boss has to meet the Delhi people who have come and the Delhi madam sits down and ask the main questions. But you’ve shown me all this power! And yet you kind of when I’m asking you what is it that you think you can do, you’re actually reflecting power somewhere else. And it’s very easy to dismiss this as yet another way of legitimizing your ability to do absolutely nothing. But I thought there was something a little bit more to it because I don’t know if any of you were to come to me and I’m in a powerful position and talk to me that I would be willing to admit to you that I’m actually totally powerless. Every boss is completely powerless, but no one likes to admit it. Why is it so legitimate for these officers to consistently speak of themselves as powerless? And I only got that because I was talking to them. So, method, right? Like, obviously, there are some things that quantitative methodologies are really useful for, and rigorous RCTs give you an ability to understand at a very granular level how change can happen in terms of the technocratic choices of things that you can input into a system, but it cannot tell you about the system. And the second thing that it doesn’t tell you is how the system is reacting and responding. So when we went back to Jehanabad a year after this experiment happens, and we went to these CRCCs who were mobilized to be part of this process, they’re all sitting, they’re very excited to chat. And they’ll open their Godrej cupboards in which their diaries were, they all had to maintain a diary as part of the training and they’d open out their training diaries. They’re very excited and proud to show all the work that they had done in their diaries, but it stops the day that the bureaucrat leaves and the program kind of loses its energy. So then there’s empty pages. And you ask, you know, why is it that nothing else happened? And, you know, you start getting a new story of how performances are internalized and understood, which is very much about responding to what is asked for rather than…And again, here, this is not a story of the lazy person or the apathetic and incompetent person. It is a story or somebody who has internalized that performance is about something that is told to me to do when I’m given the order, I’ll do it and I’ll it really well. But there’s a passivity that’s baked into it. Again, which I only got by sitting, observing, listening and trying to take it a little bit more seriously than my instinct was giving me to take, which is to say, as we all think, oh, the government is bad. Private sector is good, it’s kind of we operate in a certain type of binary, at least the Indian elites do. So for me, the question I was trying to answer, how do these ideas get embedded in systems? What is it about the system that is resisting and distorting, but also thinking of itself as a third party in this whole process of implementation that is causing this to then you know, create a vicious cycle. And I recognize that none of this can be answered by coming in and even doing a one-shot qualitative survey, semi-structured interviews is what we are always told. You can always put them as a side to a quantitative questionnaire and maybe we will get some insight, my economist colleagues would always say. And I thought, no, there’s something else that I want to do, but I was waiting for a sign. I live in Delhi. And in 2016 a new political party comes and rather intriguingly different to even Bihar, it makes education one of its big political platforms and you’re driving in the city, you’re stuck in traffic jams for hours and hours and half your life goes away in that, so you listen to the radio a lot and the deputy chief minister who was also the education minister at the time was on radio talking about things like foundational literacy numeracy, which is rare to hear on radio of a politician. So obviously something more was going on over here and we got lucky because I was part of a think tank. It’s based in Delhi. The Delhi government in its early days was really interested in feedback. Eventually, you know, politics takes over, but in the first few rounds there was some energy and interest. And so they were looking for somebody who would sit and document to do feedback. So they came to us. And we, as a policy think tank, this couldn’t be better because you want to partner with governments. So here was an opportunity. And they were really, they had other people doing RCTs. There was some CCTV camera thing that an RCT was to be done on. And I was not interested in any of that. But because they were interested in feedback, it meant that we had access into the schools to sit and listen. So it opened up the opportunity and it was very clear. You could do what, you know, apart from giving them feedback, we could do whatever research we wanted to do. So we then started looking to see what are the most sensible ways of designing method that will help us get to these embedding questions that we wanted to ask and do it by listening. And so we adopted ethnographic methods. Because we live in Delhi, I mean, research location helps. We were here, we had the infrastructure of one organization. We were able to build up teams. People came and went, but you still were able to keep a steady, solid team over a period of three to four years. It’s easy access, you could be home in the morning and in a school 45 minutes later and back in office by lunchtime. So it allowed us to innovate as well over time, even in response to government questions. So they wanted to do a teacher survey on time use because a lot of teachers said, we are not able to do things because there’s too much administrative burden. And the politicians and the bureaucrats said, but actually, if you think about it, it’s not that much, but let’s get some evidence. So we had experimented in Bihar with time use surveys of frontline workers. So using our ethnographic lens to say, actually, let’s do this quantitatively. How can you actually see what is it that a bureaucrat, frontline bureaucrat does over the course of a day? So by following them, but in a structured and rigorous way. So we brought the time use in both to respond to a government demand, but also it contributed a lot to our survey. So we mixed things up quite a bit. And because we were doing feedback with government, it meant that we were in constant conversation with them, which helped both our understanding of how reformers are thinking. But also then we could apply some of that to changing things on the ground. So a unique position of being in your living working and breathing in your research site which is very rare also helped us develop this much more kind of qualitative and ethnographic method of studying the state and I think there’s something to be said here to study the state quantitative tools will only get you so far and really being embedded gives you deeper insight.

Salimah Samji And I think what makes this particularly unique is that it wasn’t an evaluation. You are not bound by how many times, I mean, we’ve seen this plenty. Oh, we know this isn’t working, but we can’t change it because then we’re gonna like tarnish this evaluation. So you’re actually functioning for your paper as opposed to what’s really happening on the ground. And when you take this position of we’re going to over a period of three years and I think it’s 337 classroom observations, eight schools, 105 days. Of school observation, 200 teacher surveys and 2,000 circulars that you evaluated, it allows you to really put that learning lens. We’re here to listen, to learn, to understand. I’m not in here to write a paper that’s gonna be published. I’m here to just understand why is it… That when a teacher does this outside of the school system, it works. Same teacher, but when they come into the system, something is broken. What is going on here, right? And I think that’s really, really powerful. And what I also really think is, and people don’t do research like this before. How many of you have heard of Judith Tendler? Oh my God, zero. Okay. People don’t do research like this anymore. Judith Stendler was at MIT, development in the tropics is what she wrote. Good governance in the top. Thank you. Good governance and the tropics. And then Albert Hirschman, how many of you have heard of him? Great, thank you. These are like development projects observed. These are people who used the power of observation to really understand how bureaucracies worked. And so this I think really is in line with that kind of learning. So you’ve observed all of this. What are some of the things that you learned from the teachers? They’re the ones in the system. What did you learn from the head of the school? What did your learn from students? It’s all of these stakeholders that you’re spending time with and that you are talking to. What is coming out in these three years?

Yamini Aiyar Before I answer your question, just one thing to add to what you said about method and studying the state and literature gaps that we have. As I went about trying to find a better way of framing the puzzles that I was confronting, I started looking around at the literature on the Indian state. And actually it’s really surprising how, so A, we don’t have any serious public administration as an academic discipline. So there is very little that we have, that actually studies bureaucracy through a rigorous academic lens in terms of the, so there’s a lot of documentation amongst the historians about the state. The political scientists are really interested in an interest. So there’s lot of stuff on the grassroots state from the point of view of kind of seeing the state and seeing like a state. The economists are really interesting in understanding how development projects are implemented. So it applies that lens. But there really isn’t any rigorous work except for bureaucrats’ memoirs and some fiction writing that gives you a granular understanding of the nature of the state itself, who these people are, what they are. The context in which they operate. So then I looked at the literature outside of India, and actually the public administration literature outside of the India is quite robust on thinking about this differently. So Herbert Kaufman’s forest rangers? And I was trying to understand the puzzle in an era before mobile phones, GPS, tracking systems, Foucaultian panopticons that are what administration looks like today. The forest rangers showed up in the middle of nowhere and did their job. And like the big puzzle was why and how did that happen? Right. There’s not there’s no thinking like that, at least when it comes to thinking about the Indian bureaucracy. Good governance in the tropics takes, you know, one municipal government in Brazil and tries to understand how worker motivation, worker incentives were changed, not through kind of private sector tracking and performance incentives and withholding salaries, et cetera, But really through a completely different set of things to kind of push intrinsic motivation. You know, street level bureaucrats, Lipski’s work, there is, you know, which has spawned a huge literature globally in political sociology about the frontline bureaucracy. Nothing rigorous like that in India. So in a sense, I found that there’s this huge gap and what I was also trying to do while understanding this challenge of education, also trying build this body of, you now, building, sort of seeking to do some field building by trying to bring this into our conversations on state capacity. So what did we find? The bureaucracy sits in a particular context and you’re never going to reform outside of that context. When we talk about policy and the urgency of reform, particularly when it has to do with transaction intensive frontline tasks, where it’s not just the fiat of our government circular that will change a rule and suddenly things get implemented. So that works if you want to, If your goal is to ensure that money goes into a bank account, you build up the… Kind of public digital infrastructure. You ensure everybody opens a bank account and you transfer money. Boom, you’re done. You can argue that’s state capacity. I don’t think so but people do. When it comes to a classroom, I mean I can transfer all the money you want, but in fact the structure of the classroom and what actually happens is very much about the relationship between us right now and our relationship in this minute will be very different to our relationship when we go out and it’ll be very different today to very different tomorrow. These are transaction intensive tasks that are very local, very context specific, but they also shape by a broader system. And so when you’re trying to change things, ultimately the change agent is not the minister or the secretary of education, it is in fact the teacher and the headmaster who are the people in this context who are supposed to actually implement these ideas. So how did each person respond and in what way? Well, the first thing’s first, in a government particularly, and we kind of looked at the colonial history of paper in the Indian bureaucracy, nothing moves without a circular, nothing moves with out a file, we could not enter the system, we could enter the school without a file. Once your name is on a file you’re a really empowered individual. The whole movement of the right to information was premised on this notion. If this file can be made transparent to citizens, citizens can mobilize to ask questions and make claims on the state. And I caricature a little bit, but the average day of the school would operate with the headmaster coming in, putting on the computer, clicking onto the MIS system to see what the circular of the day is telling him or her to do. And in our time use, you can see how this shapes patterns of behavior. And I’ll tell one story because it’s funny more than anything else. But we were in the state of Madhya Pradesh, not Delhi, but same pattern gets played out there, waiting to meet the block education officer who showed up eventually looking very Harris because he had received an order. The order now comes via WhatsApp from his boss, the district education officer who no doubt received this order from his both and his boss to ensure that all schools had undertaken as mandatory surya namaskar exercise for yoga in the yoga class of the attendance period, which is a zero period at the start of the school day. This was very important to the political project of the government at the time. And so instead of doing anything else he had spent the whole morning even while we were sitting there making phone calls to find out did it happen yes no put the phone down and call up his boss 10 schools done 20 more to be done so I mean the order kind of shapes your day. We maintained a database of about 8 000 orders that were received over the and this is not by no chance no means exhaustive of the orders that was issued to the schools, we analyzed about 2000 of them, the order creates order within the bureaucracy, it structures the bureaucracy in a particular way, it gives everybody a sense of what they need to do and so they respond to what is said in it. We did a textual analysis of the orders and shouldn’t be it’s a hierarchical structure. You’re ordered to do things, you’re not gently asked to do anything. And for a system that doesn’t have any enforcement practice, the order itself becomes a very funny kind of legitimization of setting limits and constraints on what you can and cannot do. So your day is structured by this order. And when the reformers were trying to introduce this idea of flexibility in the classroom. So one of the big pieces of reform was to say, for a year, we’ll take a grade and reorganize grade by level of learning so that students that were closer to curriculum could be taught in a particular way. Students that were further from the curriculum could be thought in a particularly way. And this had its own sets of problems which I’ll come to shortly, but the reformers felt that we should, part of kind of motivating and energizing the teacher was part and parcel of this, so they wanted to write these orders in less coercive and more kind of encouraging ways. But the system didn’t know what to do with it. So when these orders would come which said and or you can do x, you can do y, the teachers were scratching their heads going but what are we supposed to do? As one teacher said if you’ve told us that we have to drive with the GPS and then you suddenly shut the GPS off, how are you supposed to allow us to function? Now this is actually a really important insight, because in a lot of where…So, if the reform, so one of the things that you find a lot is every time the government has tried to decentralize by empowering the frontline, you get huge unspent balances. Or as Manish Shishodiya, the education minister, used to say, when he first walked into schools, he would ask the head of schools, what is it that you want me to do? I can give you an untied fund of X amount of money to do improve infrastructure, do whatever you want for your school. And the headmasters wouldn’t say anything. So now you could interpret this as just apathy and laziness, but it could also be a system that is so designed to following orders that when you free it up, that becomes more tyrannical than the order itself. So, challenge number one, right, a system that is deeply hierarchical, when it has to organize reforms around things that require more, you know, a local knowledge feeding back into how, or local dynamics feeding back and shaping action, it finds itself somewhat stuck. And cultures of bureaucracy designed this a lot. So organizational culture became a very central piece of how I began to think about this. That hierarchical organizational culture also shaped the structure of the classroom. It isn’t that policy makers have never cared about how students learn in class and what happens. There’s been a lot of debate about this. If you go back to our policy documents from the 1950s onward, you will see a sameness in this puzzle. But the classroom itself is interpreted within this hierarchical frame. So it’s a very disciplinarian kind of structure. The teacher also only recognizes accountability on the back of very easily identifiable sets of things, which are parts percentages. And syllabus completion. Those are easy to monitor. The teacher is given, a lesson plan is handed over. There’s a syllabus that has to be run through. Manisha Shodiya’s whole mobilization effort was around the idea that the classroom is now a victim to the syllabus, the textbook and the examination, and we need to break this. But this is how the classroom is structured. So the teacher is now getting mixed messages because the order is saying, you can do things differently. The classroom is designed for something else and the middle management is still not clear how they could administer in a more flexible context. So it’s administering exactly the way it’s supposed to. And then there is a larger question of the broader social consensus in which the classroom has also organized. So we were in a school once during a parent-teacher meeting, which is also a really big part of the reform, encouraging parents in my work, we can talk about it later, but I bracketed society because I wanted to understand the black box of the school itself. The biggest issue was that parents came marching into the headmaster’s office and you tested our children. This was done to figure out what…as a learning assessment, where are students in the classroom, but you gave them this test without telling them in advance, so of course they’re not going to be prepared and you never told us the marks, what kind of school is this, right? So the relationship between schooling, the examination is a very much very deeply baked into how we think about the school system and how students do in the exam. So the teachers are also seeing, when we ask teachers what is a good or a bad student, it was always defined in terms of students who could pass the exam and students who were duffers who couldn’t pass the exams. There wasn’t no space to think about how you teach. And in all of this, it kind of creates this very important cognitive dissonance inside the classroom. When you ask a teacher why are children not learning, she will tell you many things and most of them have to do with the system in which she is locked. Some of it has to do with society, the social distance between teacher and student is vast, particularly in the public system. So she will say, ultimately, these are kids from poor backgrounds, they don’t know how to study, they won’t even think about making sure all the textbooks are in their bag when they come to school, although my kid has the same problem, but that’s separate. And so it’s always something else. And then she’ll tell you when she’s done things like these summer camps and changed how she teaches, she sees shifts. But when you switch it back, saying, OK, so what of that can you bring into the classroom? She goes back to talking about the system of the classroom, the societal and the bureaucratic system in which she is. So reforming the classroom with all these teaching at the right level ideas is not fear that you can do through the order, because the teacher has fully not internalized it. Her impetus in the classroom is very different. And then above them is the middle management. The middle management that the reformers in this instance particularly didn’t like because they were also they came from an activist background. The political party itself came through an anti-corruption movement. They wanted to bypass the middle management and go straight to the teacher. But the middle managements impulse is completely different. The context in which they become bureaucrats is completely different. They are also part of a hierarchy in which they have no flexibility. So they don’t know that they can’t give it. They don’t how to give it. To bypass all of this, they created a mentor-teacher structure. Now, a hierarchical system is suddenly told you have to mentor. Mentoring is really hard. I mean, as academics, we know that it doesn’t come naturally for your PhD supervisor to mentor you. It requires a very different set of skills. But the organizational culture is not designed to actually nurture those skills, but even more importantly, it doesn t know how to transmit it. So I’ve now created a cadre of mentor teachers and they all come in for training. I don’t quite know how to train them because I don t know what it means to mentor. So the NGOs come in and they offer some support but the NGOs are outsiders and it takes a very long slow journey for these outsiders or new ideas to slowly start finding their way in. So all to say a set of things are done trying to reorganize the classroom to teaching at the right level to shift the classroom such that the teacher feels more empowered to teach differently, she finds herself stuck because in fact it can’t be done through fear. It requires careful, slow nurturing. To do that nurturing, you created the mentor teachers, but they are also from the same world. That too needs a careful set of inputs and slow nurturing. And we would see in the newspapers, for instance, that the chief minister has now sent these mentor teachers to Singapore or to Finland or to Cambridge. And then there would be a huge uproar. All this taxpayer’s money going to waste. Everyone is going on an easy little gig, and et cetera, et cetera. Where’s the accountability? But when you talk to them, they may not have got anything, but there sense of importance and power. I was sent somewhere, I got to see something. And that kind of slowly starts changing their equation and their own status hierarchy within the structure of the school for them to start owning the idea that they are part and parcel of this process of trying to shift the classroom away from this lock-in of curriculum, textbooks, parts percentages and that’s a slow process, could not have got any of that through an evaluation. It had to be done by actually sitting there and seeing the chaos and boy was it chaotic. So I’m often asked was this successful, you know? But it is, but changes of this nature is going to be messy and it has to be messy in order for outcomes to actually change. I’ll just stop to say the real shift I saw at the end of this whole exercise was in the staff room, teachers were still complaining about all the same things. For an external observer, this was all noise, sound and fury signifying absolutely nothing. But if you paid a little bit of attention while they were complaining, they were also talking about, the kid in the last row is very far away from being able to pass the exam. Maybe I need to think about a slightly different way of teaching the child. That is a huge shift. COVID comes along the way, two years school is closed. And suddenly schools reopen. Some state governments like Tamil Nadu did a lot. Delhi, my children go to a private school, they went back to private school after two years and suddenly they were supposed to behave as though nothing happened. The government schools tried to do something serious, building up on all of this effort that had been introduced, all these small missions that cumulatively came to something. And it really showed post-COVID. When there was nothing to be blamed, except an act of God for the fact that no one was learning. And teachers knew that they had to think about this learning loss inside the classroom. So they then picked up all that they learned over those three years before the schools closed and began to apply it in the classroom. And then you really saw the conversation shift because they were literally talking about the last three kids who maybe need some special ed because they have been coming and despite all these camps they haven’t actually progressed because teachers were not kind of following progress. Of course eventually the exams came and that became the focus and it all went back. But there had been a mindset shift in a little bit of what the teachers were thinking about the classroom and the school. And that, to me, was a real change in the real state capacity. That’s what state capacity is. It isn’t about transferring money into a bank account. It is about engineering these careful shifts in the minds of all who are actual participants and deliverers of public services.

Salimah Samji I hope you get a sense of the complexity, right? We are often, it’s too easy to think of things as success or not success, and that, I would say, is really naive way to think about development. For anyone who’s done any real development, it’s always messy. It is always messy, and it is a process. But having an understanding of this can really help figure out, you use this nice word in your book about finding these spaces. And I’m wondering… If my last question, and then we’ll open up to questions from the audience. You identified levers of change, right, in all of these three years of work. Maybe you can share those, and we’ll then open up to questions.

Yamini Aiyar Sure, in all this mess, there were some important levers of change. And again, these are very different to how the state capacity conversation usually happens, at least in my interpretation, in India for sure, which is, you know, what technology can we infuse? How can we perfect the panoptical so we can monitor better? How we can we bypass these middle layers of the bureaucracy and just to deliver cash into bank accounts. Which is a great help. Some of my new work is around looking at how this correlates with voter choices. So if you’re a politician, it’s a smart move. But in all this chaos, there’s other really interesting things that are going on. And those are what I identify as the levers of change. So one is just consistent, constant, regular communication, which sometimes we say get, you know, and Mia Kalpa, this is just like political brand building that is annoying and, you know, trying to seek attribution for change that may not actually reflect on the ground. But that consistent communication is really crucial to sending messages through the system. What was really important about the way this communication was done in Delhi was that it actually happened not just through social media and radio and all of that, but actually through consistently being present. Now it’s possible to do in a smaller site, like Delhi schools are, it’s more manageable than a large state. But the fact that there was always somebody in and out from the political party, from the bureaucracy, kind of in the schools, talking to the school actors made a very big difference. And the mentor teachers played a really important role in that. So they essentially over time became ambassadors of some of the ideas that were being discussed and debated as part and parcel of what the reform process was to be about. So I think that that’s one, communication. And if you look at some of the literature on things like relational contracting, you see that even private sector firms, this tacit shifts, big reforms are not about tangible, visible things. They’re about tacit shift in culture. They only come through consistent communication. And this is not about having a communication officer that can put out a big government ad in the newspapers every morning, but it’s much more subtle, deeper engagement. The mentor teachers also played, I think, the second visible thing for me, which kind of is what I call the principal agents of change. So this idea of having these ambassadors of change, there are complications here because they often get very closely aligned to the political party. And we’ve now had a change of government in Delhi and the mentor teachers of your are seen as, folks that were part and parcel of the previous government. So they’re all being sidelined. So we have to think this through a little bit more about how these ideas can get also legitimized within the structure of administration as insulated a little bit from politics, but they play a really important role and I think it’s an important one part of this tool for me has been an important lesson. Very often when we talk about state capacity in India, We talk about it from the point of view of the challenges of blurring boundaries between politics and bureaucracy, that the politicization of the bureaucracy is a problem, it needs to be insulated, but actually no change is apolitical. You will always have this blurring. The issue is how do you blur in a way that builds capacity rather than blur in a way that de-capacitates or overly politicized, and that’s a very different set of thinking that we need to bring to the table. And the third lever actually took me back to where I began, which was these missions. So where did I begin? I begin by listening to bureaucrats saying, we know how to do the mission. There’s a burial ground of pilots. My issue is routinizing. But what Delhi actually, where it had the most success was in when it kept infusing missions. So through the course of these three years, teachers were really resistant. They felt already burdened. They felt even though the whole effort was to try and bring them in and enhance their dignity and stature. So one of the stories that the person who was one of the reformers, Aatishi, who then became the chief minister of Delhi used to repeatedly like to tell as a way of positioning, we used to have these trainings for teachers, the teachers used to complain that they would get, you know broken chairs and, you know, bad samosas. And the money that they would get as reimbursements for coming to these trainings would take six months to be delivered. So they changed that. They got tent houses to give red carpets, proper chairs, nice quality samosa that weren’t stale, bottles in little bit slurry, just stature inducing things that, you know, but teachers kind of were looking at all of this knowing, yeah, this is what politics is. Like, you know, our daily lives are in the classroom and it’s very undignified and very different to this and it beats out any desire to actually teach. So they were really resistant to all of this change. And then, you know, the CM one day wakes up as CMs do and said, oh, I need a new thing and said we’re gonna start a reading week. And then everyone is scrambling saying, what is that? And, but it converted into a reading camp that was a two week, three week camp where children were taken out of the class in the zero period instead of doing surya namaskar, they were doing a kind of reading thing. And the teachers, to begin with, really hated it, but slowly, slowly, slowly started seeing change. And then they built onto this layers and layers of missions. These missions were in the context of the school, in the context of a classroom. And each with each one, you saw progressive breaking down of kind of sites of resistance and opening up to thinking that, oh, maybe some something different is happening here. So this kind of the missions have to be consistent. All of this together, of course, requires political will. That political will has to have degrees of consistency. What was really interesting about Delhi is that the political party took its schools as a means to gain political legitimacy, which is rare, particularly in North India. And it kind of became part and parcel of what differentiated this party from other political parties. Whether voters voted because of what was done in schools is an open question and a hard one to answer. But this political legitimacy in short degrees of consistency. But I, you know, and I think that’s what made it interesting because we could bracket political will and then actually look at what’s happening inside, you know beyond that. But all to say, political will is important, but it isn’t sufficient to make change. All these other things are understanding the base in order to create reforms and identifying what I call interstitial sites of change is where state capacity gets built. So I end by saying state capacity to me and that’s the biggest lesson is about the interstetial sites of the state where you identify levers that you can leverage to carefully and slowly motivate. Part of it is to build out the intrinsic motivation. But of course, we are all humans and not everybody is fully intrinsically motivated. But the other part of it is slowly introducing tacit cultures of change within the context of the bureaucracy. I find it slightly embarrassing to go to Salimah and say, you know, I didn’t show up to school 25% of this academic year and therefore get legitimized for not showing up. I know that Salimah will take me a little bit more seriously if I say to her, oh, you know I went to work and I actually wrote a book and can we talk about it? So my account starts shifting and the way in which I seek legitimacy and status within my job starts getting a different identity. And that’s, to me, where state capacity will be built. So the more panopticons we build, the more it is a sign of a state that has no capacity rather than a state, that is enthused with some capacity.