Subscribe
Share
Share
Embed
These podcasts are recordings from the Cutting Edge Issues in Development Thinking & Practice lecture series 2023/24, 2022/23, 2021/22 and 2020/21, a visiting lecture series coordinated by Professor of Development Studies, Professor James Putzel and Dr Laura Mann.
The Cutting Edge series provides students and guests with fascinating insights into the practical world of international development. Renowned guest lecturers share their expertise and invite discussion on an exciting range of issues, from climate change policy, to pressing humanitarian crises. In 2020, the series took place online, enabling us to host fantastic speakers from around the world and to stream the lectures on YouTube, opening them up to a global audience. Now we are back in person but still recording the sessions to share with our global audience.
SPEAKER 0
Hey. Welcome, everybody. Welcome. We have a we have a special occasion today as we we've stolen three Senior faculty members whom so just up the road to come and speak to us on a really a topic that's really central to the study of development and to the practice of development, making anti-corruption real, how to stop wasting money and start making progress. This is quite, quite important operational significance. So let me introduce our speakers and our discussant. We're really delighted to have Muskaan. He's one of the leading lights in international development in the UK and internationally. And in recent years he's developed a penchant for attracting loads of money to do research in various centres that have the most incredible acronyms. So the first one, he and Pallavi are co-directors of the Foreign and Commonwealth Development Office funded Anti-Corruption Evidence Research Consortium, which is shortened as Psoas School of Oriental and African Studies show as Ace. So they're doing Ace work there. And he's also joint head of the Fcdo funded. So as you know partnership. And this one is really interesting. It's the research and evidence on Nepal's transition with the acronym rent. Mushtaq has written together with John Malcolm Soon data, probably the most important book on the role of rents in development, rents, rent seeking and economic development. I know it's a I know it's a long time ago since you published that, but I don't think anything beats it to this date. His line of you're reading, those of you who are studying DB 400 and the MSC Development studies anyway, are reading several of Mustafa's to work, including fortuitously, next week we're doing a session on governance and some of your work is there know? Mushtaq has inspired many of us with his work on political settlements, and that has spawned a whole and quite diverse range of research. And he also in relationship. I'm sorry, I'm mentioning old publications of yours today. He edited a book called State Formation in Palestine, which was published in 2004. And this was an early exploration of a lot of the things you'll be talking about tonight, I think. So. You know, for years. And this is I'll stop introducing you in a moment. But for years, Mushtaq laboured away with an alternative and critical perspective to the good governance programs being promoted at the world Bank, out of the stall of Danny Kaufman and Art Cray. And it's wonderful to see now his research being published in world Development and him attracting proper funding to take his critical ideas forward. He's lucky to be working with Private Roy, and who is reader in political economy at the centre for International Studies and Diplomacy at Psoas and Pallavi. Now, I'm really impressed with the range of country projects that you're leading, and I'm sure you're going to reflect on them tonight during your talk. Recently, she co-authored a paper in World Development Anti-Corruption strategies in the Nigerian Health System and has written other papers on the electricity sector, the education sector, etc. and that have been published recently out of the research that they're speaking to tonight. And so thank you also for coming walking down the street. And our third visitor from Psoas is Jonathan DeJong. You'll see him in a moment. He's here down in front. Jonathan, welcome back to the LSC. We miss you Jonathan works in the department with us during the crisis, States Research Centre's time. And Jonathan is senior lecturer in political economy in the Department of Development Studies. That so as he works on political economy of growth and development in Latin America, especially in Venezuela, you should. If you're interested in the resource curse or interested in Venezuela or interested in industrialisation, you should look at his book from windfall Winter, Windfall to Curse Oil and Industrialisation in Venezuela in 1920 to the present or to the recent past. Yeah. And so Jonathan is going to provide some critical remarks on his colleagues research and get us on to the Q&A session. So without further ado, I turn it over to you, Mushtaq. Welcome to the LSC.
SPEAKER 2
The. Thanks, James, for those fulsome remarks. Obviously exaggerated, treated with a big pinch of salt. The talk today will be on introducing you to a framework and approach that we've developed in cyberspace for making anti-corruption real, for making it implementable, and for having a developmental impact. Now, when we talk about corruption, we are often talking about a vast variety of things, because corruption refers to the misuse of power by people who have power to break rules of different sorts, usually bureaucrats or politicians breaking rules to benefit themselves, and much of this harm society to different extents. Some of it harms society a lot, some of it harms society a little. But corruption describes many different types of things. At one end you have predatory corruption, where warlords go with guns and the extract from citizens. You have political corruption where politicians, often in developing and emerging countries, gather a lot of resources illegally and then use that to buy themselves support from powerful people in society to get themselves re-elected or stay in power. So political patronage and clientelism are also forms of corruption. And then you have policy distorting corruption where government policies in health, education, power, skills, industrial policy, agricultural policy, resources are diverted and the result is affected. And we call that policy distorting corruption. And then you have the kind of bureaucratic obstructionism and red tape corruption where you have to pay speed money to bypass obstacles created by bureaucrats. In other words, corruption describes a vast array of different things. And this is one of the problems. When people talk about corruption, they're not very clear what is the problem they're trying to address. And the result is they achieve nothing. So we are going to break down this problem and focus on what we think is a very important type of corruption, which is the policy distorting corruption, the corruption that affects policies in implementation of policies. And another way of thinking about our work is actually it's not really anti-corruption, it's about redesigning policy better so that they can actually get implemented in developing and emerging countries, sometimes even in advanced countries nowadays, where if you can't design the policy that it will be implemented, then you don't achieve the objective. And the main reason why policies aren't implemented is actually because powerful people are diverting the resources of the policy somewhere else. So anti-corruption and policy design and policy implementation are just different ways of looking at the same thing. So that's my introduction. Now let's start with the presentation and why we were commissioned around six, seven years ago and began this research, which is a big research project. It's. Underpinnings are a massive waste of resources on anti-corruption. Over a long period of time, and not just over the last ten years, but over the last 20 years where we've invested a lot of money on anti-corruption. And not only has this not really delivered very good results. A lot of the anti-corruption itself has become corrupted. So you have anti-corruption commissions that are corrupt. You have investments in police and judiciary which are corrupt. You have investments in political parties and democracy and democracy is corrupt. So all of the investments in the things which are supposed to reduce corruption have also been corrupted. So our objective was to say, is it time to give up or are we addressing the problem in the wrong way? And our answer is that we we are addressing the problem in the wrong way. And this connects to a lot of research I had done before this on governance and so on, that James was referring to it. And the objective here is to use evidence. So it's evidence based. It's not normative. It's not saying this ought to be, this should be this would be a good idea. But saying what does the evidence tell us about what can be done and what can be done should be feasible. It should be implementable in different contexts, given the configuration of power and capabilities in that context. And it must have a developmental impact. It must make things better. We're not just doing anti-corruption because we are normatively against it, which we might also be. So feasibility means that it will be implemented or enforced. An impact means that there has to be a social gain from the anti-corruption. And the main analytical point that I want to come across, and then Pallavi will give a 2 or 3 case studies, is that we need to understand that we are doing anti-corruption in countries where the rule of law is weak. If the rule of law is weak, it means that laws are enforced, but they're not enforced all the time and for everyone equally. And that is causing the problem. So the problem in the first place is that you have a weak rule of law. So if you have a weak rule of law and laws are being violated by powerful people, you can't solve that problem by passing other laws which are going to punish those people for violating the first set of laws, because the second set of laws on anti-corruption will also not be followed, will also be violated. So you can't solve a rule violation problem with more rules. You have to ask why were the rules being violated in the first place? And what can we say about what might be enforced in that context? And this is the core point of the whole talk, is that we need to identify sufficiently powerful actors who are going to be our allies, who will implement that specific anti-corruption in their own interest, not because they care about us or they care about the world. Like most people, they will act largely out of their own self-interest. And if you can do that, then and only then will we have an implementable strategy which is also in the public interest. And the public interest is is a second point. So we're not just saying let's identify what the powerful people want to do, because most of the time, those who have power at any level in society are using that power to enrich themselves and not the rest of us. So it's not simply saying, let's work with whoever has power, it's saying, and this is a tricky bit. Can we identify things that some sufficiently powerful subset of people want to do who can enforce it, which is also in the public interest? Once you get your head around that, I think you can start making progress. So this is a very different framing from the way corruption has been framed in the policy literature. In the policy literature. Anti-Corruption is all about achieving transparency, which means we should discover quickly who is violating the rules. So transparency includes things like having a free press, having watchdogs, having democratic discussions and elections, having digitisation so that things are very clear. Yeah. So all kinds of things that reveal violations is transparency and accountability. Other systems about what to do with it. What do you do with this information. So once you have this information. So accountability includes elections. The assumption is citizens will vote out corrupt politicians or systems within the bureaucracy to fire or punish the people who are breaking the rules. And we've invested a lot of money on these things, and that's exactly what I was describing. We now have quite a lot of transparency in developing and emerging countries. We have a lot of accountability systems, but they are not used. In fact, the systems themselves are corrupted. So there is a third factor. And the third factor is who is going to use this information and the systems. And if you can't identify that then you haven't got an anti-corruption strategy. So the third factor which is critical is enforcement. And this enforcement again normally we think about enforcement as some very powerful principal up there who is going to come and enforce. And and here too, we take a very different position. And we say that most enforcement in most societies is not done by horizontal top down enforcement. Most enforcement in most societies is actually what we call horizontal enforcement. And this is a critical idea that comes from the political settlement way of thinking. And I'm going to spell that out in some detail. To get that horizontal enforcement going. What we are saying is that we need to rethink policy design. You can't have policies which look good. And then you say, now I need to implement this policy. So I need enforcement to help the policy be enforced. That's a completely wrong way of thinking. You have to design the policy so that the very design of the policy gives resources and opportunities to people who, in their own self-interest, will then enforce that policy. And that policy serves some public purpose. That means you can't have a separate governance strand, a separate enforcement strand setting aside from your policy, the policy has to be redesigned. And I think this way of thinking is why I said at the outset, anti-corruption, the way we do it is really about policy design to to to enhance the chances of its implementation. So what we are talking about, and this will become clearer by the end of the lecture, is designing the policy with those internal checks built in so that you actually get people sufficiently powerful people, not necessarily the most powerful people in society. Violations are often done by mid-level or lower level people, and to stop them, you need other equivalent people actors, organisations who will check the people who have the right or the power to violate. But they must have as much power as the people who are violating. Okay, so so so that needs to be kept in mind when we are talking about power. We're talking about relative power. So research is necessary to identify these opportunities because they don't. They're not self-evident. But once we do the research, then it becomes a kind of subject for advocacy for for introduction into policy. Okay. Let's think about why this is so different from the way we think about the problem. A lot of the problems in the world come from simple economic models, which people implicitly use, and they don't understand a lot of the context in which these economic models were developed. The top down way of thinking is what economists call the principal agent model. And even if you have never heard about the principle agent model, a lot of your thinking is affected by it, because that idea has infected a lot of the ways in which we think about policy. Okay, so the principal agent model says that there is somebody usually who wants to enforce something. There is a rule or there is a policy, or there is an institution that somebody wants to enforce and that person, that organisation is the principal. And the assumption is the principals can they have the power to enforce and they want to enforce because it's in their interest to enforce the contract, the rule, the policy, whatever it is. But the agent who is being asked by the principal to do certain things can hide information and can do things that the principal doesn't want. And so an example would be citizens elect politicians. Citizens are now the principal. We elect the politicians. The politicians are supposed to deliver public goods. They use our tax money to enrich themselves. We don't observe it. But had we observed it, we could have gotten rid of the politicians or politicians could be the principal. They have a manifesto to deliver health services. They tell the bureaucrats who are now the agents. The bureaucrats get the money, they steal the money, the politician doesn't see it and therefore they get away with it. Okay, so you have nested principal agent problems, one inside the other. And these nested problems are all about the principal being cheated by some agent. And the same person can be a principal in one problem and the agent in another problem. Okay, so that's the principal agent way of thinking. Now how do you stop this rot from happening in the principal agent we are thinking is you need to have transparency and accountability. So as soon as the principal can see what the agent is doing, they have the information transparency and they have the systems for punishing the agent. They have the accountability system. Your problem is solved, right? The voters vote out the corrupt politicians, the politicians kick out the corrupt bureaucrats, and you are delivering stuff in the right way. And here is the challenge. You do it and you find that voters elect known corrupt people, and the corrupt politicians do not fire the known corrupt bureaucrats. Everybody is colluding with everybody else. And then the whole emphasis shifts from saying, this is a principal agent problem, but you're still stuck with that paradigm. You're saying the principals are unprincipled. There is something normatively wrong with the principals they are not enforcing. So it must be social norms, it must be culture, it must be something of that ethics. Have ethical training, tell people corruption is a bad thing. You then start wasting money on that sort of stuff. These might all be contributors, because actually norms are a dependent variable in a society where there's a lot of corruption, people get used to the corruption, and so that becomes part of the normative behaviour. But that's not the cause of the problem. The cause of the problem is somewhere else. So here is where I think we say that this is a completely misleading way of thinking about the problem. Principals and agents are not sitting in air. They are surrounded by a social context, and both principles and agents behave in particular ways because they are under pressure from others in society to behave in particular ways. And that's where the horizontal check idea comes in. And the horizontal checks are the things you are not seeing in this model, the people who are horizontal to the principle and the horizontal to the agent who are constraining their behaviour. And we take that for granted in advanced countries, in rule of law countries, because there is so much there that as soon as a detection of a violation happens, it is corrected. And we think it's being corrected because I have detected it. But you are not understanding the context in which that detection can be corrected. So here is. A thought experiment. Why do people stand in queues? Now, if you think about standing in queue. It's a simple rule. It's a it's a rule about getting onto a bus. And it seems that people in developing countries don't know this rule. So there's a lot of confusion and chaos in getting into buses. How do you enforce this rule? What is it that makes people stand in line to get onto a bus? A lot of people immediately jump to the conclusion that this is about social training and socialisation and knowledge. And if you train people, they will immediately see the benefit of standing in line and they will get on the bus. So here's why this is wrong. You know, we work in developing countries and in the same developing country, in the same city. I can tell you that people stand in queues which are very normatively North European in some places, and the same people don't stand in queues in another place just a few miles away. You cannot explain that with culture. Right. So why are people standing in queues in Gulshan in Dhaka, where I live, and not standing in queues in Milford, which is five miles away? And the same people is not something that you can explain either normatively or with the principal agent problem or failing. And the reason is very simple. People stand in queues if the people standing behind them in the queue have the power and the capabilities and the interest to pull you back if you break the queue. And if you know that if I break the queue, other people in the queue will pull me back. You don't break the queue and you instinctively know which queues you can do that in and which queues you can't do that in. And that is why developing countries are quite schizophrenic. The same people know that. Now. If I try to stand in queue, there are powerful people who will break the queue. No one will stop them and I will never get on the bus. So I'm reasonably powerful. I must join and break the queue. And there are other places. You know how powerful you are. If you break the queue, you get hammered, right? So you don't break the queue. The vertical enforcement is still necessary. You still need a police. So if the fight breaks out and the police come to stop the fight, that's when the vertical enforcement become necessary. But most of the hard work is done in the horizontal enforcement, right? So the difference is when the police come to the place where the the horizontal enforcement is good and someone is is having a fight, everybody will say, this person was breaking the rules and the police have no option but to pick that person up. Whereas if you are in a queue where the queue wasn't there and there's a fight and the police come and you say that person broke the queue, the police will not pick that person up because the powerful people in that queue will not support the police in picking that person up. So the police actually operate when they're under pressure from powerful people. The police themselves are under horizontal pressures, and that is why a lot of the police in developing countries break the rules, because most of the time they're under horizontal pressure from powerful people to break the rules. But not always. Sometimes they don't break the rules. And so the interesting thing is when and why and what can we do about it? So here is what the reality actually looks like. The principal agent thing in the middle is embedded in a lot of horizontal activity, and the principal is himself or herself under a lot of pressure, either to break the rules or enforce the rules. And the agent is under a lot of pressure to break the rules or enforce the rules. And so how they're behaving can't be understood without all of that activity going on around them. So if you think of the principle agent problem in an advanced country, what is happening is that the horizontal relationships of the principle in Asia are constraining them to correct the rule violation as soon as it's detected. Why is that? Well, I don't want to go too far into it, but this is a political settlement's idea that the distribution of power and capabilities in society matters. So if most of your organisations in your society are productive and highly capable, okay, they need to have lots of rule enforcement for them to operate because a highly productive and complex organisation needs a lot of rule enforcement to operate. Okay. So imagine a society where the actors are all productive organisations, and one of those productive organisations has violated a rule which is they haven't paid their tax, and this has been detected by the principal, who is a lowly bureaucrat. And the violator is a powerful company and they haven't paid their tax. Okay. Now what makes that lowly bureaucrat enforce the rule on this powerful company is that while every powerful company individually would love to not pay their tax, they also know that their ability to interact with other powerful companies relies on their reputation for being rule following. And so if you lose that reputation, you can't operate because you have a contract based society, which is a sign of a complex economy. So as soon as that agent is detected, all the other actors that that agent is interacting with other companies, which will say, I don't want to interact with that company because it breaks the rules, even though they would love to break the rules themselves as soon as it's detected. You can't say, I will continue operating with that company, and that principal will say, I am not very strong. Do I really want to have a fight with this company? Will be under pressure from other principals saying if you don't, we will remove you and if you need our help, we will come and help you. So the principal is under pressure to act, and the agent is under pressure to get punished and pay the fine and correct their their behaviour. Right. That is why it's happening. It's not just that there is transparency and accountability. There's a lot of activity happening around the scenes. Whereas in a developing country. And so when this happens across all sectors and it's widespread, you have what we call a rule of law. And a rule of law means that as soon as a violation is detected, regardless of the power of the violator and who that violator is, the violator gets sanctioned. That's a rule of law. Okay. Now that's as many people have said, including very mainstream economists like North, Wallace and Wineglass. A rule of law is tenuously there in a handful of countries, and even in the countries that is there, it's continuously in collapse, in danger of collapse. In most of the world. We are far away from a rule of law, so the rules are enforced. But on some people, some of the time there's partial enforcement and is often entrenched violations. But developing countries are not anarchists. It's not that no rules are followed. There is a structure, but you have to understand the logic behind that structure. And so the way we think of how to make that intervention is first, understanding the structure in a particular area that we are interested in and then saying what is the most effective way of creating more horizontal activity, pressuring principals and agents to operate in socially desirable ways, which means you definitely can't do it in all sectors and all activities tomorrow, which is why developing countries do not become good governance. This is what I've been saying for 200 years. But developing countries can make incremental improvements which take them into in a developmental direction. So. Just to end up now and handing over to Bolivia. Effective anti-corruption, therefore, requires understanding the power, capability and interests of horizontal actors. And that's why we call our approach the DCI approach. And so these three things very quickly, the complex ideas. But I'm going to quickly throw it at you so you can get a sense of it. Power describes the material and organisational capacities of actors. So when you are in a contest over resources, how long you can hold out in that context matters a lot. So how much how long you can hold out matters on depends on how rich you are. The richer you are, the more resources you have, the more of the conflict you can engage in. But it's not the only determinant. The organisational capacity of the actor is also a determinant, which is why the richer side doesn't always win the better. So if you are not rich, you can compensate for that by having better organisations so you can take more pain and you can sustain yourself in conflict for longer. And we are in a world with lots of conflict. So just imagine all the conflicts that are going on. You can have very asymmetric wars between a much richer side and a much poorer side. But if the poorer side has more organisational and ideological power and they can hang up, they can win. So power is that combination of material and organisational and ideological components, which gives you holding power. And in developing countries, many of the most powerful actors and organisations and the richest in society are intermediate classes, networks of different kinds of patron client networks, which are actually very powerful capabilities for us. Very important. We are not referring to enforcement capabilities. When we talk about capabilities, we are talking about productive capabilities. How do you make a living? How does that actor or that organisation make a living? And the reason is very important if you have the capabilities to make a living by producing value, that doesn't just mean in manufacturing, it could be in services, it could be in agriculture, it could be in sports, it could be in music. If you can create value by doing what you are doing well, you are likely to want the rules to be enforced, right? So if you're a very good football team, you support the enforcement of rules. If you are a rubbish football team, you might actually want the rules to be broken repeatedly, right? So your productive capabilities are important because it determines the likelihood that you actually want rules to be enforced. And we see this again and again in our work. The more productive or almost productive an organisation is, the more likely it is to support rules. The less productive it is, the more likely it is to extract resources and want to extract resources through different forms of extraction and capabilities, together with the competitive environment in that sector. Determine interests. So you could be quite productively capable, but if you are the only organisation in that sector, you might become a crony capitalist. But if you are capable and there are lots of you around and you're competing, then the interests of that organisation would be more aligned with rule enforcement and rule following behaviour. So those are the three things we look for. And as we give you examples, you get an idea of why this is important. And we say that power capabilities and interests need to be mapped, and we need to have an idea of how the different actors in a, in a policy arena are likely to respond. And we map their power capabilities and interests, not by asking them, what is your power, capability and interest? We map it by looking at their past behaviour. So we look at the evidence of how they have behaved in the past in different contexts, and from that we deduce what their power capabilities and interests are and from that. Almost towards the end of my my part, we identified three anticorruption strategies. The first is where in that area, there's already a lot of horizontal checking going on. This is the easiest and the most promising sector. The queues are already there in some areas and the QS aren't there in other areas. And then we say, okay, so obviously there is some effective checks already happening in that society. How how can we enhance that in the QS that are not already in line. Okay. And we call that enhancing effective checks. That's the most promising. That's the most immediate quick results are possible there. The second strategy is that the QS aren't there, but there are people trying to get into the QS. So you we map the power capabilities and interest, and we say it is feasible with some feasible policies to actually start creating the conditions for the queues. They're not there yet, but we can see how it could happen. This is a more difficult job, but if you can start creating the effective checks then you can move to enhancing. So you start with two and you move to one. But then there are lots of areas in developing countries where not only are there no queues, we don't even see the queues being formed because the distribution of power capabilities and interests are such that almost everybody in that activity has very strong incentives for violating rules, and no feasible change will make them follow the rules. And here, because some of these areas are really important and do a lot of social damage, we have a third strategy which is called mitigation and transformation, which is that instead of trying to force the stopping of the corruption, we deal with the harmful effects of the corruption. Hence mitigation. And the transformation is a much more longer term strategy of making people exit from those activities. So the steps that follow we will now take you through is how do we identify what to do? The first step is what we call almost an economic anthropology analysis. We immerse ourselves in understanding a policy process from a resource flow, and at each stage of the resource flow, there are different actors trying to influence the resource flow. So there are different levels in that resource flow right down to a policy outcome. So if you think of something like, let's say industrial policy, industrial policy might at the highest level involve the Ministry of Finance, it might involve the central bank, it might involve the land administration, all of the resources that are being given to to firms to improve their productivity. Then those resources come down. And then there are lower level bureaucrats and politicians who are intervening or managing that resource flow right down to the lowest level, where companies are getting benefits, and then they are negotiating and being checked by other companies and other bureaucrats. And there might be links between different levels so that politicians at one level might be protecting violators at another level and so on. So the power of people at different levels also depend on the coalitions and the networks they have, which enhance or reduce their power to, to enforce or violate the rules. And what we are looking for is, is there an effective entry point anywhere here where the horizontal checks are almost there and or already there that we can enhance, which will improve the policy outcome that we want to improve? We do not have the ambition of saying, can we clean up everything and everything becomes rule. Following that is an impossible task in a developing country. What we are asking is, is there some level of intervention that will significantly impact the outcome and significantly improve the outcome? Understanding the complexity of the different levels and their and their interconnections. So once we identify the links and the levels where this is possible, that's the research starting point. But also if there is no link or level that horizontal checks are possible, there is also then the research becomes about mitigation and transformation strategies. This is where I come back to why we are interested in policy distorting corruption. Every type of corruption has different types of horizontal checks going on. If you wanted to, if you wanted to stop predatory corruption. So, okay, if you look at this diagram as you go to the right. Corruption becomes more damaging. And as you go higher up it becomes more difficult to address. So the most damaging types of corruption are the most difficult to address. Predatory corruption involves warlords. They have guns. They're going and looting from people. To stop that, you need a horizontal check, which comes from other armed people who will have the incentive to stop them. Okay, so the horizontal check there is extremely difficult political corruption happens because politicians can win elections by stealing money and doing patronage to stop them. You would have to have other political actors who don't need to do that, but who can still win elections. And when does that happen? That usually happens when your country already has lots of productive firms who are paying a lot of taxes, who want rules following behaviour, who are willing to fund a political party to enforce the rules, and that party can take on parties which are violating rules. Well, that's far away in most developing countries. So this is why we say political corruption is very tough to act. Foreign policy distorting corruption is everything that is about health, education, industrial policy. And some of it is quite difficult to address. Some of it is quite feasible to address. This is where we focus on and the market restrictions. This is a red tape and bureaucracy. This is actually very easy to address because they harm lots of powerful people. And most developing countries have made a lot of progress in removing that red tape type of corruption, but it doesn't have a big impact on development. Okay, so our entry point is that policy distorting corruption, which you can understand once you understand the horizontal check and why it's significant. And that is actually the area where we can have a lot of impact, which will then make fighting political corruption easier as we go on. And all societies have a mix of all of these types of corruption going on. So it's not that, you know, some countries only have policy distorting corruption and others that predatory corruption. I will stop there. And now we will go through the more difficult job of giving you some examples.
SPEAKER 3
No, which is why we make a good team. I always think the theory is the harder part, and the empirical bit is the easier part, because we see how this one goes. So what does it look like now? This is where we actually go and work on our framework and work with our framework. So this was one of the very interesting projects that we did in Bangladesh, which was on reducing corruption and climate adaptation projects. Now Bangladesh, as some of you will know, is one of the most vulnerable countries to climate change. It's also a country with very high levels of corruption. It is estimated that almost 30% of funds that are meant to go into climate adaptation are distorted away. What we did was we tried to find evidence based on the first strategy that was outlined and it looked like this. So that nice, nice, neat little symmetrical diagram with little blobs, green blobs going up and down. This is what it ends up looking when we are brainstorming and have had eight coffees through the day. You might not make sense of it, but actually if you look at where we are, we start at a high level. We have central government contractors, real contractors, subcontractors, and then we come down to small peasants, landowners, traders, and then we see that this is the level where we can actually intervene in and find an effective solution. And I'll quickly take you through how that might be done. But this gives you an idea of the economic anthropology that Mushtaq was talking about. So we've already mentioned it here. So this is a process. This is this is actually a method of very detailed process analysis. We we look at we know which sectors. By this time we've done our political sediment analysis. We work in Bangladesh, Nigeria and Tanzania mainly. And for this phase of it's Bangladesh, Nigeria. So we know the political settlement of the country. We know how power is distributed, how policies are distributed, who gets what, who captures what for what reason, what the relative power of these various sectors are. And we also know therefore what sectors we want to target. So we'll be very clear that in Bangladesh, climate adaptation infrastructure is one sector that we wanted to look at. And here we actually started off talking about. So what's the process. Who's who's linked. What are the levels. Where does intervention seem possible? Where does intervention seem like a dead end. And it's one thing that we want to actually emphasise here. A lot of people think doing anti-corruption is hopeless. Doing political economy is hopeless. There is no way out. You know, it's all the politics and there's nothing that that can, you know, there's nothing beyond the politics of that situation. What we are trying to show is it's actually possible to have solutions. They might not be the kind of big bang solutions people are hoping for. The change will happen overnight. If that's not possible, B that's highly not desirable. We've seen what that's what that's become, whether it's in Afghanistan, whether it's in Iraq, whether it's another context. You just can't have a big bang where you change the governance structure and the political settlement of a country and then bring in rule of law. It's just not going to happen. You will regress to the mean. So this is this is the kind of process analysis that we that we do. And then we build on these to, to at every level, build a picture of the power capabilities and interest. And again capabilities for us do not does not mean enforcement capability. That's that's the power aspect of IT capabilities. How productive am I. Because if I'm not productive, why would I want to be rule enforcing? I will need other means to be profitable and productive. Not. But if I'm inherently a productive organisation, I won't need to break those rules to make money. And then we actually go. The second level of research is where we start looking for evidence of one. What are the different drivers of corruption? Are the different drivers of corruption? And are people being corrupt for different reasons that we can perhaps address through policy in the other, which is what we call enhancing effective checks. We try and look for similar agents or similar players in similar circumstances, but some are actually showing lower levels of corruption in their outcomes and some aren't. But everything else is similar. And then we try and find out what why is it that there are some exhibiting high levels of corruption in some lower? What is the difference? And can we do anything to actually enhance this in a way that we can make rules following the sort of entrenched norm for that particular sector or level that we are looking at. So what did we do in the adaptation project? We looked at adaptation projects by the same implementing agency. Now these are implementing agencies of the Bangladesh government. And we saw what was something that was very interesting, same agency, same formal rules of accountability, rules governing the sector. But in some cases where where this was these were implementing agencies. You had better outcomes. What were they building? They were building embankments. They were building cyclone shelters. Some of you know, Bangladesh is extremely vulnerable to very, very damaging cyclones and also a river erosion and erosion because it's an estuarine country. So these are extremely critical defences against climate change. And therefore these are extremely important for the outcomes to be productive. And in. This case. When we say productive, it means a good quality so that the next monsoon doesn't wash away an embankment, or the next cyclone doesn't blow away the roof of a cyclone shelter. So what we saw was the same agency building, well constructed, good quality embankments and cyclone shelters, and the same agency doing something completely to the contrary. What was the reason that this was happening? And we conducted a survey across 1900 households. We looked at. And I'm just going to this slide to show you, because I know we're short of time. We looked at these four areas and we looked at areas where we had in these four areas, cyclone shelters where these were the outcome was of, let's say, shoddy. In another, it wasn't. And these were very similar in all aspects except that we we said one thing we made sure of one thing that when we were looking at these these projects, we made sure that the good outcomes were in otherwise higher corruption areas, because then you could turn around and tell us said, hey, but that's because the general background of corruption was actually positive. You know, this was a this was a place where there were high levels of governance. So you actually had a better outcome. So we ensure that that didn't happen. You couldn't come back and tell us if this wasn't happening. So why was corruption less in some of these areas. And we saw something very interesting. Let me take you back to this slide here. We actually saw that most people would be looking at the contractors, looking the subcontractors. We'd be looking at civil society because this was Transparency International, which was actually a partner in this in this project and trying to look for enforcement to come from here. We saw something very interesting. We saw that it was the small peasant, in this case, a small present in Bangladesh. It's a relatively well-off peasant, not the landless sharecropper, but somebody who has more land, who, for instance, is in that particular village or that particular town is relatively more powerful, is relatively more capable. And why is it that when that that that person is getting personally interested in the outcome, why is it that the outcomes are better? Think of it this way this is a smaller farmer, but who has more surplus to take to the market? Who wants roads to be? Or the road to his nearest market town to be more navigable? That person will want the embankment to also double up as a road. That person will also want the cyclone shelter to perhaps become a community centre. And here's the interesting thing in in these kinds of strategies where we are looking for differences in behaviour, some are corrupt and some are not corrupt. In very similar circumstances, we actually see that in these cases rules exist, and in some cases the rules are actually being adhered to. In Bangladesh, what was important is all adaptation projects have to have dual use. So it's not just about protecting you from the cyclone or from the river taking out bits of your river bank. It also has to double up as either a road or as as community shelter, as I said. And then we realise that for stronger participation of these individuals, what needs to happen is the dual use needs to become stronger. Why is that going to happen? Why would we use become stronger? It's because of this. The small peasant is actually having tea in the local shop with the subcontractor. The subcontractor in the small peasant appears. Not these real contractors, not the high level politicians at this level. What was the horizontal enforcement that was working? If I want my embankment to be a road so that I can carry my goods to the market, I'm going to make sure that that that real contract that we mapped out uses the right kind of sand, the right kind of clinker uses the right kind of material. And if I'm if I'm not interested in it, then I don't really care what's built. I'm not going to be overseeing the real contract in terms of what what quality of products he's putting in while building the community centre or the embankment. So here the solution was better planning, better planning, which is make sure that the it's supposed to be dual use. Most of these constructions aren't dual use. Make sure that they are actually being used for these dual purposes, and ensure that you get in that constituency that we identified. In order for that dual use to actually have the right kind of outcome, which is sustainable architecture and sustainable construction, even when the worst cyclone hits or when there's a huge flood. So this was this was something that was a very important finding for us using the first strategy that Mustaq mentioned, which is enhancing checks. The checks are already there. We already knew that in some cases the outcomes are better. In some cases they weren't. We identified why the outcomes were better. And then we try. We got evidence of what was it that made the outcome better, and then tried to see where the policy to then enhance what was already happening. So this gives you a sense of what what we were saying visualised. If some of you like these kinds of nice corridor drawings. So there's high corruption in some projects. The relatively. Apply to the diamond. We found corruption during construction, low corruption and similar projects because a relatively powerful interest were actually mobilised and checking the contract contractors actually constructing those sites. So this is this is one example of how to enhance these effective checks. Strategy two is a very interesting one. And here I'm going to very quickly talk about the power sector in Nigeria. Here we will see that we had a lot of other projects. So the first strategy, we also had projects in the electricity sector in in Bangladesh for skills development in Bangladesh. This first strategy that you can read it up on our website, but I'm going to take you through one example, a case study from the Nigerian power sector. And what was the problem here? The problem here was a corrupt behaviour was rampant, but we also stepped back and found that people were being corrupt for very different reasons. And what the policy solution here is to look at the reasons and the drivers of corruption and see are there particular drivers of corruption that creating some kind of policy around them could actually affect? Could we provide an exit for the people who were corrupt for very different reasons? And I'm going to tell you what the reasons were. As you can see here, there are agents violating for unreasonable reasons in the Nigerian case to very, very give you a summary of what was happening in the Nigerian power sector in 2013. It went through a really big bang privatisation. It was supposed to be the biggest privatisation effort in in Nigeria. In the African electricity sector. A lot of private players came in and you'd expect that things would get better, but transmission generating capacity in the country has stayed pretty much stagnant from 2015 to 2023. Average generating and evacuation capacity of the national grid has stayed limited to about 4000MW. Why did that happen? Now, of course, there were issues of legacy corruption, but one of the main reasons was those who were bidders and who became successful bidders and owners of assets in the electricity grid were actually very politically connected bidders. Now, no amount of vertical or horizontal enforcement is going to work on them. The sector in obviously in a country as large as Nigeria is too big to fail. You can't have that sector going bankrupt. So there were a lot of bailouts going into these, into companies in that sector to no avail. They just weren't willing to be rescued. So the sector is now in a very low level operational equilibrium, but that's at one one end, which is the high level political corruption. Then we looked at other kinds of actors within the sector who might be violating for very different reasons. There would be consumers, they could be commercial users. And we looked at a very, I'd say, ignored segment in Nigeria, which is a small and medium enterprise. The SME owner. Really, why did we look at the SME owner? Because in most developing countries it's the SME sector which is the beating heart of employment growth and income generation is one way of broad basing your income growth. And in Nigeria, you might expect that if you ask SMEs at a particular survey, what's your biggest constraint? The answer is in corruption, because that's what people tend to think about. Nigeria is a corrupt country. The first answer that they gave was lack of electricity. And what was interesting is that they were violating for what we would call unreasonable reasons. For in their head, it's reasonable because they know there is intractable corruption at the higher level of these politically connected players. So when they're not paying their bills, when they're stealing electricity, when they're bribing engineers in their head, it's a perfectly reasonable violation because you know how much electricity they actually get from their grid on an average. And we did a survey in, you know, what? What actually was the beating heart of the Nigerian manufacturing sector in the southeast of the country, between 4 to 5 hours of electricity is what they get per day from the grid. For the rest they self generate. So if I'm an SME manufacturer and if this is my only source of livelihood, if I'm stealing from the grid in my head, I'm doing something reasonable. And I think for most of us, if we actually look at it very dispassionately, they're not really far off. Not very wrong. I get five hours when I need 49. What else am I going to do? Because for the rest, I'm topping up with very costly self-generation generators and buying diesel. I'm actually paying off other people to supply electricity. To me, it's not easy for me to be following and if I am really following, I'm actually not going to get the electricity. So we're actually doing a pilot to test the kind of intervention that we have thought of. And our intervention here was dealing from the grid start small 20 megawatt CNG or hybrid plants, which will only supply to the established clusters, the clusters. Interestingly, we were not talking about setting up clusters because that is an entire new can of worms where you have an SME business zone, zoning regulations, rent, capture of a different sort. We said there are some clusters, fully functioning SME clusters in India, sorry, in Nigeria, which have existed for about 20 to 30 years, let's supply them with embedded. Plans dealing from the grid 20MW, and they will, in that case, be an incentive to actually pay up. And why would that be? We established two things from our research. One, the willingness to pay because they were already paying very, very high prices for the electricity that they were getting from the grid and that they were actually self generating with their generators, with the smuggled diesel, with the bribing of the engineers. And they were already finding some informal ways ten SMEs would get together and say, let's hire a generator and let's try and see whether we can solve the problem. So there were these informal mechanisms already happening. So for us it was about then creating a horizontal check, which is we create this institutional mechanism, a collective electricity supply mechanism where the incentive to free right is then less because if nine out nine of us are happy paying our bills, the 10th one ultimately not paying well, in that case, I can form a very effective horizontal constituency, which can then monitor the 10th person who is actually not following. So that's the kind of change that we were trying to build. So I'd just like to draw your attention to this. This little phrase on the generator at the side is a little machine over there. Now, you could think that this is what defines Nigeria, but that's not true. And hustling is is tiring after a point of time. I mean, it's good for sort of short, snappy blocks on the country, but honestly, people would much rather be following if they just had the right kind of incentive mechanism to be following. And this is, this is what we are actually trying to do here. And this in fact, is the cluster in Abuja. It's it's a small furniture manufacturing, small cluster that we are trying to test this. It's both a technical solution, but it's also an institutional solution. Just providing the technical solution, which is this hybrid plant isn't going to be enough. Is the collective monitoring mechanism going to be successful? Because ultimately, for us, anti-corruption is sustainable and successful when we have self enforcement and when there is horizontal enforcement, there is no point having vertical enforcement when you have the money, when you have somebody coming in for two years, and then the person goes away and the money dries up and we go back to being corrupt. It's only when I'm following the rules in my own interest is corruption or anti-corruption successful? That's the important thing that we want to kick off with this, with this pilot project. And this is also this is a published paper. The adaptation project is a published papers. We've given you the links in case and if you're interested. So what's happening here. So if you solve this we have a series of very nasty looking red blobs. But here we're actually seeing some of the red turning green. Because what we are doing is there are some violators who are coming under pressure and the ones who are actually breaking rules for reasonable reasons. If we address the reasonable reason, if policy is possible, if it's possible to create a policy that can address these reasonable reasons, then a lot of these agents then become rule followers. And that's the transition that we are seeing from red to somewhat green over here. Then we have the most difficult kind of corruption, which again, was was the case study that I'm talking about is from Nigeria, which is what we call networked corruption. And this is actually what political corruption is all about. When corruption is so rampant that everybody is benefiting from it. And this is this is a situation where we can't look at creating horizontal checks because we don't find differentiated behaviour, high corruption, low corruption, we don't find reasonable or unreasonable reasons. There might be a hierarchy of benefits. Some people are getting higher payoffs from being corrupt and some people are getting lower payoffs. But everybody in the end is benefiting from this kind of behaviour. That's intractable corruption. That's what we call network corruption. So as we said, high corruption agents, ineffective tracking by horizontal actor actress. This is this is just a corruption ecosystem here. And what we studied here. I don't know how any of you are familiar with how much you're familiar with Nigeria, but there is an interesting subsector within the extractive sector in Nigeria, which is called artisanal oil refining. There's trust me, there's nothing artisanal about it. There's nothing nice about it. There's nothing clean about it. It's criminal. It's underground. It's very, very polluting. And it has immediate externalities on the people who are both within that artisanal oil refinery. And this is actually it's called they refined oil. What is done in these big mega factories is actually done in these kinds of kitchens in Nigeria, in the Niger Delta, where you actually have a lot of oil. And if you know of the Niger Delta, this is where a lot of the violence also takes place in terms of how oil benefits are distributed or not. In the case of the states in the Niger Delta. So they literally cook stolen crude. That's what it's called. And at various points where the crude condensates, they make it into kerosene, into diesel, into into petrol for your cars. But it's done at this cottage industry level now you think but you know, this is polluting. The water will do for you. The air will be terribly polluted. It must be so easy for civil. It'll just, you know, get people together and create a kind of collective action to stop this. Or, you know, it would be easy for the states to enforce taking the Army have have some really tough sanctions on them, criminalise them. Everything's been tried. And suddenly collective action efforts have also been tried. They haven't worked. Why haven't they worked? And that's principally because everybody is gaining, not just in terms of money, but they're finding employment. This is an area which is highly impoverished in Nigeria. In fact, in Western Africa, they're finding employment not just in the artisanal refineries. And this is something that is well known. But what our research actually found is that there are jobs linked to the artisanal refineries, hospitality, catering, supplying of materials. Now, if I shut this down, I'm actually taking away a livelihood in one fell swoop. Also very critically linked to what I just said about the electricity sector in Nigeria. These are all areas which are completely reeling from the grid. I'm going to buy diesel or kerosene to power my generator so that I can watch the next English Premier League match, because we all know the Nigerians love their football and I wish I had that photograph to show you. But there's a there's a great photograph we have of a generator with a sign in front which says the latest EPL match being shown here, and this is a great source of demand. I can't shut it down. What has happened, though, is anti-corruption as as violent sanctions is bringing in the Army Securitising anti-corruption as as successive Nigerian governments have done, has actually created a spiral of violence because it's only pushed this underground, made this a more criminal activity. So that's a real unintended consequence of anti-corruption. But what do you do here? Something has to be done. Something's got to give here, as Mushtaq said. And I also explained, this is corruption that's networked. Everybody is benefiting across the hierarchy. The governors are in some way. And the person involved in in this distillation that's going on, this is actually going to be sold off at various retail outlets. This is how you can buy petrol or crude oil products in Nigeria. They're benefiting. The intervention here is for us to mitigate. We know that there are pollution externalities which cause a lot of health outcomes. So can we link the primary health care networks in these areas to at least provide some kind of treatments for, for the waterborne diseases and for the airborne diseases that are happening in terms of supplanting the demand that these create in terms of fuel. Can we work with solar power? And, you know, this is something where Nigeria is actually pushing quite a bit in terms of solar generation. So can can these artisanal alongside the demand for artisanal products, can these be replaced with solar power. So that's the mitigation aspect of it. The longer term aspect of it is absolutely very difficult, but that's the only way out for us, which is the transformative aspect of it. Livelihoods. The reason why artisan and oil refineries still exist, because livelihoods and incomes are generated through them. And can that can that happen? That's the transformative aspect of it, really. So going forward, I'm just going to quickly sum up what we've said. And I know we've given you a lot in just 40 minutes, but hopefully you've got something to take away from here. Feasible and effective anti-corruption requires identifying and improving incentives for actors who can, which is the power and who want to and the want to is also important because the want is when I'm productive. So that's my capability and that will make me interested in the self enforcement. So all three are very important. I can and I need to want anti-corruption needs a radical rethink. It's what we call it has to be baked into governance policies and one one often if you're going out and soon some of you will be going out and working with a lot of donors out there. You know, donors will have a governance policy, and then they'll have a policy for the health sector. They'll have a policy for SMEs, they have a policy in climate. And then they'll say, oh, but we've forgotten governance. There's a lot of corruption in all of these. You cannot have these as two separate pillars. Anti-Corruption has to be part of whatever sectoral policy that there is. You can't have an electricity sector policy that doesn't think about how to deal with policy constraining corruption. So that's why we absolutely insist this is the new way of thinking. It needs to have to come together. And this is also, if you think about it, what is the problem of policy implementation in developing countries that there is no enforcement? This is one way of getting in that horizontal enforcement of getting in that self enforcement, which is going to make policy implementation successful. It's not just about anti-corruption, it's much wider. That's how we look at it. When the rule of law is is weak, when there is rule by law, what we need is horizontal enforcement. Absolutely. Vertical enforcement is required. We are not saying transparency and accountability are not things that aren't required. Absolutely not. But they are not enough. They're never sufficient in these kinds of contexts. So you need to design policies. Keeping this embeddedness horizontal enforcement in mind and operational strategy will be different. It will take time. We have to be patient. I'm I'm saying we need patient capital, but not in the. Sense in which you normally know of patient capital, but we need a lot of patient capital to deal with corruption. It's going to take time. There are different ways of seeing how you can address the corruption. It could be by expanding rules that rule following behaviour that already exists. It could be splitting up the drivers of corruption and addressing the driver that's most amenable. And if that's not possible, then the mitigation aspect is what we need to look at. And anti-corruption is long term. Sometimes people actually misinterpret us. And certainly Mushtaq has been interpreted, misinterpreted quite a few times. And people have said, what are you saying? Corruption is good for growth. No one, no one is absolutely saying that. It just means that the results will not come immediately. Corruption cannot be good for growth. But what comes from from an understanding of the fact that corruption is not good for growth doesn't mean that you go and implement something that's big bang. It's incremental. It has to be incremental, but it's additive. And that's how you get this long chain, which will ultimately give give us the rule of law kinds of societies that we're looking for.
SPEAKER 4
That was fantastic. You've been privileged to hear two of the best political economists that work on this in the world today. We all have. I'm going to make a few comments about where I think this framework makes tremendously important contributions to both theory and policy. And I think that's the relevance of this work, is that it has resonance and makes contributions and is innovative both in the terrain of theory, development theory, but also policy making, and particularly. And particularly how policy. Can make a difference to power balances in an economy. I'm going to discuss that. I think the first thing is that on a very broad level, this will work. Addresses one of the classic problems in developed, which is the transferability problem. That is why does the education system in Finland not work somewhere else? There's a lot of lessons learned in development. Follow South Korea's industrial policy or follow whoever your favourite model is and whatever sectors policy. And we know that is problematic. And this work points out that policies work because there are underlying political power balances and systems of accountability that make enforcement possible. It's easy to design a policy on paper. Their work points out how difficult it is to implement it. It's kind of if you watch Roger Federer play tennis or Novak Djokovic play tennis, you could study them all they want. But the constellation of factors that made them exceptional is not necessarily something you can copy. And the development community is full of these kind of they neglect the importance of how transferable institutions and policies are. I think the second important thing that this work does is it looks at variation within a country as a source of identifying policies that might work. So instead of asking the question, why do some other country do everything better? It asks, why are there differential outcomes within a country? And they look for clues as to why that might matter. And I think that's quite important, because people can relate to what works and does in inside their own country, and pointing out that some model across the other side of the world works is not very helpful, particularly because the political arrangements across the world are very different from where you're studying. So I think it makes an incredibly important contribution in that regard. It also emphasises that an old point in development, which is that stage of development, makes a big difference. Alexander Gerken It's one of the hallmark features of development economics is that the appropriate policies in countries have to be tailored to what stage of development and economy is, what's the fiscal base? Does it matter if the country collects $50 per person in tax versus $30,000 per person in tax, they make a very big deal out of that. And they also point out that the starting conditions in a country are not just about power balances, but also that a lot of the political contests happen in the informal sector, and there's not a lot of rule following. So they're urging us not to use Denmark as a guide to understanding policy of transferring good governance from a country that has a high fiscal tax base and has formal, productive sectors that operate. But let's start with a country that has a low tax base and has a lot of informality. And rule breaking is not a pathology, it's a norm. And when you see documents of the world Bank and a lot of other NGOs, you don't see that often as a starting point. This research has urged us to look at the world like that as a starting point, if we're going to think about making effective policies. It also is research that's developing a strategy. It's not a wish list like the SDGs, you know. You have to have all these outcomes that we desire. They're giving us a feasible strategy of how to get from A to B. The SDGs are a set of incoherent wish lists of things that if you actually achieve results on one, you're actually going to have to do badly on other things, not a coherent strategy. It attempts to make strategy at the incentive at the centre of linking research into policy use. Now, I think one of the things that this research brings back, and it's an old it's an older idea and I think it's important to highlight it. Manka Olson once pointed out a long time ago, the collective goods, a street light is provided often not because we have collective action, because we have a dominant player, maybe a millionaire, that lives on the end of that street and will provide the street light for themselves, but has the externality of providing light for the whole street. This research is coming back to that idea that we need to identify. We need to look at social differentiation in society and figure out who are the actors that have incentives to provide public goods, to enforce them when they're provided. And that's something that's been lost in a lot of the abstract discussions of governance. Terms like accountability and transparency don't mean anything when you're not talking about the real social differentiation and who has the interest to enforce rules. It's not everyone. And this research makes that quite clear. Um, you know, and the work they didn't mention the work about the importance of understanding the productive capacity of an economy is quite important. They also, the Ace program has done research on vocational training programs. And, you know, similar to the logic about similar to the logic that was presented in the Bangladesh case. They find that there's billions of dollars wasted in vocational programs where people where there's not productive firms that can take advantage of it. So people finagle systems to steal money, but where there are productive firms that actually can use it. You see these things work. This is a fundamental issue of political economy and not using buzzwords like transparency and accountability without specifying what are we talking about and who's going to be interested in enforcing these things. And that's why political economy is so central to understanding how governance works, of which this work identifies. I think it's also important because it's trying to identify solutions without revolutionary changes in political power balances. It's trying to look at certain levels, entry points where you're finding actors that might need a prod with already have an interest in doing this and and really informing the government about these things. Which leads me to sort of a few queries. One of them is, is the problem of the cases you studied one of information and not political power values. That is what prevents a relatively local, powerful elite, not a rich person, but a rich peasant from from emerging or figuring out this problem without external advice. They live there. They have an interest in enforcing things. Why do we need. It's just a broad question of political economy and consultancy. General. Why do they need an external actor to tell them that? The second issue is. The contractors that are corrupt, they're probably aligned with higher level elites. So why are higher level elites allowing these lower level but locally powerful people? Why are they allowing these folks to get away with this stuff? Because if we can identify policies that allow this to happen, then maybe power balances are a lot more malleable than we think. Because one thing about political economy that's a straightjacket. If you say power balance determines outcome, South Korea's political settlement determined why it could design and implement industrial policy better than, say, an India or Indonesia or Nigeria. This is opening up the possibility that there's more room for manoeuvre within the power balance. It's not like colonialism prevented everything, or that civil wars always lead to good outcomes or bad outcomes. It's looking at nuance and variation and saying there's a role that interests play in, and there's a role that policy plays in this world. Second question is how scalable are these solutions? Hirschman Albert Hirschman talks about the demonstration effect. If you can show that in certain sectors, building productive capacity matters, will that feed up the system? This is just a question now higher up the system to say we actually need to build productive capacity. It's more important than we thought. Export industries are more important than they thought because reputation and quality matters to to an export sector. If things start to break down, or if there's rotten food that gets exported, that ruins everyone. So we're learning also that exporting itself creates important horizontal checks and balances, because following the rules of the game makes everyone more profitable. And it's another angle to think about where the export sector matters. I'm going to stop here, but I just wanted to ask one question, which is, do we really think that this framework can't tackle political corruption? We know that political corruption has very differential outcomes. As Max Weber informed us a long time ago, you know, China is different from Nigeria. They both have rampant cronyism and corruption would somehow work. So is there ways of thinking about how the framework can attack political corruption and not just the policy distortive corruption? Okay, I'm going to stop there, but thanks very much.
SPEAKER 5
I really want to thank.
SPEAKER 1
All three of you, and I hope the audience is not by now saying, oh, I should have gone to study it so as instead of you. So. Before I ask you to respond to Jonathan's. Provoking thoughts and also reminding us about Hershman possibilities, which this is all about, it seems. I want to go to the audience. Can I? Can I see the hands? Okay. How about. How about this woman in the pink? Sweater.
SPEAKER 6
So much for the presentation. That was really fascinating. I worked recently. Project funded by us did. That was, I believe, looking at the policy distorting. Corruption. It was certainly trying to address.
UNKNOWN
The transparency accountability issues.
SPEAKER 6
Didn't see this framework being used. So my question is, to what extent is this claim by the international community? And is that happening in a way that you would like and or what would you what would you sort of hope to see happens moving forward?
SPEAKER 1
But over on the other side there.
SPEAKER 7
Very much for your presentation. My question relates earlier to one of the questions posed regarding power structures and power settlements. And in Bangladesh, I wanted to ask, where did the power that elevated these individuals to the same level of the subcontractors come from, such that they were able to have horizontal enforcement on these subcontractors?
UNKNOWN
Good question. See.
SPEAKER 1
It looks okay here.
SPEAKER 7
A very generalised question, but I think it's very important. What role do you think that media plays in this whole line? Is the media vertical actor instead of horizontal actor? And really, how can we fix it? How can we especially considering developing countries.
SPEAKER 1
Okay, let me start with those three.
SPEAKER 2
Should I should I start? Okay, so I think there's a question about the enforcement on the local contractors, which is also aligned with Jonathan's question about how the power structure is constructed.
SPEAKER 1
So I think.
SPEAKER 2
I think the point that I want to make very strongly is that the distribution of power and capabilities in countries is always changing. It's always malleable, but it's not totally malleable. Right. So policy and politics is the art of the possible. So we are always trying to shift things, but we're not trying to shift it in such a way that the blowback will be so hard that we get killed in the process. So I think figuring that out is an art is not a science. It's an art. And I think that the aim always is actually to shift, to shift the power balance, to shift. Now you're. So if I come back to your question about the local landed peasants and the subcontractors, that messy thing that Balawi put up is the starting point is to actually understand that detail of the process through which contracts are given and implemented in a specific sector, in a specific country. In the case of Bangladesh, what happens for these embankment type projects is that the money comes to the central ministry, and it is contracted out to politically connected people at the very top. Those politically connected people are very powerfully connected. And the standard. Mechanism there is that the contract value, around about 30% of the contract value comes straight back to the bureaucrats and the politicians who are giving the contract. Okay. That is a very high level political corruption that we don't think we can stop right now in Bangladesh. Right? So all contracts are given by the people who are getting the contract and the high level. And I'm only talking about infrastructure contracts. Those people at the high level who are getting those those companies are actually shell companies. They don't build anything. Their whole function is to get the contract, give a kickback, and then they subcontract the subcontract is technically illegal, is not allowed in violation. Okay. But the subcontract is how every contract is, is done. And then it's subcontract and subcontracted right down to the village level. The village level subcontractor is not actually very powerful. They are simply paid to get a contract and they're building the damn thing. Now they are actually constituted of people who are locally powerful in the village, but they come from the same social background as the small peasants. They sit in the same tea shops and they sit in the same. They belong to the same party. They know the party officials at the local level. Okay. So this these people are also corrupt. But but at the level of implementation, there's a huge difference between an embankment which is well constructed. Even with the 30% has gone at the top, you can still make a good embankment or you can make a rubbish embankment, right? So our point is not to as I said when I was looking at this policy diagram and all these circles, is not to ever say, let's stop every corruption from top to bottom, because you're absolutely right. Some of this corruption is connected with very high level politicians who are using this money for their own kickbacks and their political purposes. If we try to take that on right now, we will get massive pushback. But if you take if you say actually, it makes a huge deal of difference to Bangladesh, whether your embankment is is surviving and if operational or whether it gets washed away in the next monsoon, then I think we have there are solutions and the solutions are to actually mobilise the checks and balances at the local level, because there there's a more equal distribution of power between the contractors and citizens. Now does this actually is it scalable? So which is a question that Jonathan asked. I think scalability is one of the really big questions about all evidence based policy. Whether you do an RCT or whether you do our kind of work or anything scalability, you will not know the scalability answer till you scale and but but I think that we have a better understanding of scalability than a typical RCT because in our case, it's already happening. And when it's happening and settling in disparate areas, we don't see any attempt by the higher ups to stop it. And this is just one example. There are examples which are even much more sensitive, where in the power sector you have power plants which are, you know, 30% or 40. Overpriced and other power plants, which are relatively cheap, operating in the same country, and no one is trying to stop the ones which are low cost, because the point is that even the low corruption ones are sufficiently profitable for everyone to make money. And if we are not trying to say you shouldn't make money, but what we are saying is you shouldn't make excessive amounts of money and there is some amounts of kickbacks and so on, which is politically necessary given the political settlement, and we are not trying to address that. So the question you had, Jonathan, very important one is other aspects of political corruption that we can address. Absolutely. But they're they're the ones where the, the, the politicians engaging in this corruption face challenges from other politicians or other actors who have the same level of power. So I cannot stop the political corruption of the Prime Minister, but I can stop the political corruption of the local level officials in the In the Embankment project, who are now under pressure from other people who belong to the same party, who want the embankment project to be to be done. So, so these are really important questions. And the scalability question, whoever asked it is a really important one, because maybe it was you because I think that's policy, evidence based policy. You know, it has to have validity. That is it's your research method is internally valid. It has to have external validity, which means it can be transferred somewhere else. But the most important thing is it has to be scalable and scalable. You only know when you scale. Right. But I think that a lot of this I suspect, is actually scalable. When you go. Was there any other question?
SPEAKER 3
I think there was a question on. I think if you ask whether it's a problem for information, and I'm going to plug that with the media one. I think because both go together, if we don't think it's a problem of information problem out there, there is information available. The problem is power asymmetries. I mean, you talk to your best source for any kind of political economy, not its most developing countries is the person who drives you to your favour. The drivers know the best. They have all the information, including where the president of the country is making his or her money from. The information is out there and this includes the media. The media knows there's very little that we can do to act on just in terms of the raw information. Information is not power. What we are dealing with is a power asymmetry, which is very different from how we think of information as power in, in the more sort of average rule of law country. So that's that's one. And in most rule by law politics, media itself is politically captured. Media is actually a part of political corruption and part of that entire network corruption that that I spoke about. So media. Yes. The counterfactual of not having the media. Absolutely. We do need a free and fair media, which we see across the world, that the space for free and fair media has been being shot down. It's not just on corruption, it's on on all other lines, but certainly on corruption. If you are, you know, basically hitting it somebody's political interests. Media is either a complicit player or media is powerless enough, is far too powerless to actually act to do something about it. So I think that's the information question. And the media question I think, is can.
SPEAKER 4
I just add.
SPEAKER 2
To this? I think Jonathan was saying, why don't why don't those villages figure this out from themselves and do it? So here is a very interesting point. We researchers have power, right? So that person at the village level is only seeing their own little world, and they're not conscious of the fact that this is a well constructed embankment. It just happens to be a constructed embankment because they're taking actions. It's only someone who has the capacity to look at multiple embankments and look at the whole process from above can make that assessment. So that little village and the villagers, and they really don't know, and they don't have the interest in changing the policy framework in Bangladesh to make other villages have good embankments. So I think that's the role of research. The role of research is to identify what is feasible, and our normative values are important. So we are not just, you know, we have our own power capabilities and interest in the in the pot. Right? So we are not we are not neutral actors. We are we go with our own interests. We have our own capabilities. We have our own power. And but what we want, what we should do is come up with a solution that becomes sustainable after we exit the sea. In other words, it shouldn't just be there when the NGO is there or the academic is said. This is a problem with a lot of RCTs. The RCT has the academics sitting there doing the treatment and then the treatment works. And when you do the same thing without the academics, the treatment doesn't work right. So our job is to is to have a look for treatments which are happening without us. And that treatment is much more credible than a treatment that we have designed.
SPEAKER 1
I'm going to ask you a question, which you don't answer now, but in answering the students questions, you can come back to it because and it's linked with what you just were explaining. So there's there's a striking absence of the state in your story and a striking empowerment of researchers who can be there on the spot, etcetera. So it seems to me that there is a story here that's quite important, which is the what kind of political situation at the level of the state can call up Mushtaq and his and Bellerby, right. And their team of, of, of corruption breakers, you know, and come into the scene and look at the impact. What so. So I think that has to be more.
SPEAKER 4
They're in the.
SPEAKER 1
Study. Okay. Let's go.
SPEAKER 4
Just wanted to make one point about the media.
SPEAKER 1
Come back with the next one. Okay. Yes. Here, please. All right. How about how about the two women here? First one and then the next.
SPEAKER 6
Is Princess and not that I would justify corruption. But I just wanted to ask like the fact that attention is more towards. The people with less power. The subcontract. The small business. Is it it like. Is it like a bit of a negative light? Is linked with people at a higher power than. Like more than the. Know, because when we look at the people in power, they are actually helping each other. To an extent, like I said, I'm not justifying it, but then I'm just trying to like, find out. As if that didn't like shifts attention away from the bigger, bigger fish.
SPEAKER 1
Could you pass that just in front of you? That's a efficient move you want.
SPEAKER 6
No question about how to link between the micro, the individual and the macro. But I saw that your model adopted. As far as what? Further? The stories.
SPEAKER 8
Okay? Okay. Nicholas.
SPEAKER 7
Quick question about that. So when we. When you have, for instance, a kind of. Non-productive firms in the country and we know with our, let's say, treatment, we can only make very small changes, this kind of incremental changes towards that. And this is a long transformation process. Is the government doing that at all? Because when I think I get to donate and say, hey, we have a seven year program, we will make a very, very small change. Is that all happening? I think that's a real political problem in that. So it is my perspective and.
SPEAKER 1
Okay, great. So if you're not, if you answer those rather concisely, we can go to the audience again.
SPEAKER 2
So let me take the last one which is also connected to your one. And you can do the.
SPEAKER 3
I can be.
SPEAKER 0
Vulnerable. Yeah. Yeah.
SPEAKER 1
So I think there's a.
SPEAKER 2
Really important question about the state and politics in developing countries. And I think that varies greatly across countries. So I think the issue is that you can't generalise, you know, that the role of the political leadership and of political ruling coalitions and of the state. So I think that if you look at really successful countries, if you look at the Chinas and the South Koreas of the world, a lot of this discovery of what works was internal to the interests of the ruling coalition. So if you anyone who knows China will know that the whole history of China is based on experimentation. And post 1980s, they tried different things in different places. They shut down what didn't work. They scaled up what worked. This is because the state. So that brings us to the higher level of political settlement question in the country. The high level political settlement, which you haven't touched on here at all, is some of our other work is that if you have a political ruling coalition, which can take a long time horizon and has a lot of enforcement power internally within the coalition, it can do those experiments and it can shut things down, and it can do other things and then scale them up. Generally speaking, we as researchers don't really need to do very much in those countries. They don't need us and they won't ask us, right. So they're doing their own stuff. We are working in countries where the overall political settlement is such that the ruling coalition doesn't have the interest, incentive or ability to do it. They are making short term money quickly, and they are doing it in a way that is damaging maybe their own long term interests, but they have no capacity to to come to that. Right. So they know that what they're doing is not sustainable, but they still do it right now. Their what our job is to actually.
SPEAKER 7
Find.
SPEAKER 2
Feasible things that can be done sometimes, regardless of whether the ruling coalition is interested or not. Right. So a lot of the things that we are doing can be done by local politics, can be done with development partners, can be done with intermediate level politics, doesn't really need the Prime Minister to be interested. And if you some of these things in Bangladesh or Nigeria or the countries we work in, you would not even make them interested, right? They are interested in much bigger rent issues, like for example, the government of Bangladesh might be really interested in building big infrastructure projects with Chinese money, which they are borrowing tons of money for and taking 30% from that. Then they're doing it really well with good infrastructure projects, but it's leaving the country with unsustainable debt because it's overpriced. Now we don't need to advise them on this, right. They're doing that. But if you go and say, can you build some embankments properly, this is not worth their time. They're not interested in such small things. But we have entry points with local politics. We work with the local government engineering department or we work with development partners, if you can, if you can. So Bangladesh, for example, has a massive pipeline of adaptation funds, which donors are not giving Bangladesh because they're not sure that Bangladesh will not be able to use it with low enough levels of corruption to make it sustainable. Okay, so so that's the starting point. So if you go to donors and say, yes, Bangladesh has a very poor governance, but here is some research which shows if you give your adaptation funds and say the condition of releasing this money is that the contractor at the top will have to have five different positions for the location of this embankment, and you take it to the local community, and the local community decides the one that is most convenient for it. And by definition, that selection of the embankment which the local community wants is the one where the most powerful people are behind that. We know that that's the case. How local politics works, that itself reduces corruption. Now, this is something we are trying to sell to the development partners, because instead of that, this is the connection with the policy design. We are saying you should design your adaptation policy so that the checks are built into it. For example, by putting into the contract design that you have to have three or 4 or 5 locations for this embankment. Remember, Bangladesh is a flat country. You can build the embankment here for 200 yards, there for 500 yards there. It makes a hell of a lot of difference to where the landed peasants are. Who can best use this land if you manage to build it at the point where they are interested in the quality of the embankment, suddenly the quality improves. The Prime Minister is not interested in this. They are interested in much bigger things, right? Or the kind of research we are doing on power or foreign skills. But there are constituencies which are interested. So we are talking in the countries where the time horizon and the enforcement capabilities of the top people is low. So this is really important to understand. We are not talking in South Korea, we are not talking in China, we are not talking. And the Prime minister of Bangladesh or Nigeria, it was not possible for me to advise. We are we have to do the research and find the most useful place to take it to, and the feasibility of it is that we have identified such places, whether they pick it up or not. We have, see, but I think that it is imaginable that someone will pick it up and it's imaginable that it will work, whereas if we ask. My mother's government. What do you want us to do? They will set us tasks which are completely pointless. Right.
SPEAKER 1
Thanks, Jonathan. You wanted to come back before and then?
SPEAKER 4
No, I just wanted to make a point about the media. I mean, well, one thing. The way corruption, the dominant theme of corruption, is that it happens because of policy distortion. And if you let this give the state authorities too much authority and discretion to allocate resources, it'll be abused or to be captured, which generates bribery. But what's often forgotten in about a political settlement's economy framework reminds us that corruption is a tool of political competition. The scandal itself, through the media and the media is often captured, is a way to discredit people before an election. It's not something that just happens. People actively use it. And so when you see corruption scandals, like in Italy in the 1990s or Lava Jato in Brazil, these were these were part of political competition to wipe out parties. They're not just things that happen. And so we talk about the media. If this is just a whistleblower. That's not the only role the media plays. And it probably not even the most important one in these type of countries that we're stuck in, but they're study and research.
SPEAKER 3
In our countries. The media will actually work for party power, and we know that that's pretty clear. As as John said, I want to quickly raise the point about bullying. And I completely take your point of this next question, but I'm just going to flip it. If you if you think about the situations that we are working in, whether it's the embankment or we worked in healthcare, for instance, which sees a lot of anti-corruption in Nigeria and in the electricity sector, when you think of the standard anti-corruption rules. Who is it actually impinging on the most exactly this vulnerable category that we are working on? And what is the kind of anti-corruption? It's sanctions. It's a kind of securitised and corruption that I spoke about in the case of the Niger Delta. These are the most vulnerable people in that in that so-called extractive function in Nigeria. The anti-corruption that we are talking about is mitigation and transformation. We're focusing on them actually, because they're the most vulnerable and actually because the standard anti-corruption, which has to do with accountability and transparency and enforcement, is really that vertical enforcement, you will get the most powerless over that. The doctor, who has a very reasonable reason, for instance, a female doctor in Bangladesh, this is what we found, that a lot of them are absent because they genuinely they have genuine security concerns about where they're posted it, far away from where they are. They need family support. So there are and there are those who are politically powerful and are absent because they have good protection. But you're not therefore going to enforce anything on on the person who's politically protected. You are going to enforce on the person who's not there, which finds being there, you know, and insecurity. We are actually trying to move away from that discourse. Anti-Corruption always targets the vulnerable, but in the way that we are saying the outcome is going to be very different, it's also efficiency enhancing for them. Somebody the Niger Delta hasn't had anybody speaking to her. You know, the lady who runs, whether she did on brothels and it's a way of making money. And the police will be the first one to go there and say this is morally wrong and take her out. Nobody's really talking about, well, what kind of livelihood alternative livelihood can can be provided. So yes, we are working with the vulnerable and those who seem the most powerless, but also in very different ways, so that the outcome is actually not the one that standard anti-corruption provides. And I hope that answers your.
SPEAKER 1
Very good. We have time for one more round of questions. Give up your hands if you have one here, one there. So how about there? And then here. I'm always shy to ask my own students. Okay.
SPEAKER 7
Go ahead. I just have a quick one on political decentralisation and what role that.
SPEAKER 9
Plays in a lot of what you're talking about, I guess works better if you have either more power at a local. Of all or more delivery, at least at a local level, which can then be pressured. So I guess, is that one of the larger scale policy recommendations that all indications which might work.
UNKNOWN
Yes. I just had a question about global level power checks.
S11
So looking at the.
SPEAKER 7
People who look at benefits for.
S11
The right kinds of soil, the local people checking that may lead to a reduction in corruption.
SPEAKER 7
Have you noticed any.
S11
Association between a reduction in corruption and associations. So the local level contractor, maybe they share an ethnic identity with people watching them. They share a local regional identity, a religious identity. Is there any association that you notice in that way? Very good.
SPEAKER 8
All right.
SPEAKER 6
So I have a question that kind of relates to your older work on rents and political settlement. And in some of the more recent literature, the kind of key thing is whether there's learning taking place. And I'm wondering, you know, if we take the example of the kind of Bangladeshi people building embankments is the kind of transformative element that the people who have pressure on them learn how to build embankments better, and that therefore they become more. If, as contractors, I was a little unclear if the kind of horizontal power is just there in that area and they work in that area, or whether those people can get jobs in other areas, and whether that is the kind of learning is the source of the scalability.
SPEAKER 1
Okay. I'm going to collect whatever last questions there are. So after.
S12
Question. I was just wondering about the limits to this incremental approach to finding. Because it seems to me that, for example, if you save the work of the local actors, which. Build a bank base with goods in and provision of then probably the people selling the rubbish and in the first place will have also pressure and power to put some horizontal pressure to be corrupt, ironically so that the next election will be with rubbish. So how did you manage these concerns or is it something that came up?
SPEAKER 7
Can you elaborate what you mean.
S12
For example? Because incrementally, they're still going to be in the other interests that will be in place. So you have to start at the local level. So that stays in place, which you can move to higher hierarchies etcetera. So we're just wondering how do you sort of fight against this sort of, let's say corruption type, whereby putting one story at the local level, you sort of hold off all of the other interests that will put pressure on the. On the lowest level they're asking for. It's really, really good. Sorry.
SPEAKER 3
So in some sense, leakage is behaviour. Yes.
SPEAKER 1
Okay, I think that's it. So I'll give you three the last word. Okay. So.
SPEAKER 2
Decentralisation is no panacea for anything, right? The example that we gave was about a very decentralised process. There are other areas of anti-corruption that we are working in where the corruption is very high level, but there is still horizontal checks at that level for us to see a difference. So we have a great paper on the commissioning of private power plants in Bangladesh, which are very high level corruption. But even there, there are ways of increasing the horizontal checks at the bidding process, which can bring in politically unconnected bidders, for example, with some kinds of finance which reduce the risk, and those kinds of finance which reduce the risk of politically unconnected bidders coming in, have a magical effect on the bid price and on the corruption. Right. It's incredible. So even at the highest level, I think that you can you can figure out ways of increasing these horizontal checks. Now, I think that so this is connected to the question of with the higher level, stop this and so on. I think you have to understand that, you know, corruption is opportunistic. Corruption is.
SPEAKER 1
And if.
SPEAKER 2
There is an opportunity there will be corruption. It becomes more difficult. There is not enough corruption. If you go to the most corrupt countries, there are lots of things which are happening quite well, and there are lots of things which are happening with a lot of corruption. There are lots of queues and there are lots of chaos. So it's not that as soon as there is a queue, the people from the chaos will come and break up the queue because the queue is actually working for the people around it. This is the whole point that we are making. The whole point that we are making is that the queue will only survive if the people who are in the queue and the people around the queue want the queue, and they have the power to enforce it. So someone will have to come and spend a lot of force to break up a few that the people want to be in. And I don't see any evidence of this. Right. So I don't see any evidence that people are deliberately going out and breaking all queues just because they are in a place where not having a few advantages to them. So even with the power plants, which have massive millions of dollars of theft, they don't come and stop the power plants which have known millions of dollars out there. Right? Because those cheaper power plants are also making lots of money for the people, and it's sufficient for them to do their business and the sufficiently powerful to stop other people from breaking their business. So you can't break the business of a of a politically unconnected investor who is making lots of money. So I think that's what we are trying to get at, is the sustainability and the interests of the people around, which is going to create from the bottom up, lots of bits of rule following behaviour, which ultimately will connect up. And there is a big debate between us and those who say that this incremental approach will always collapse, because at some point someone is going to come along and break it all up. And so there are other people who believe in what they call big bang anti-corruption that unless you, you, you have a significant shift in the equilibrium from high corruption to low corruption, it's not sustainable. And we strongly and deeply disagree with that. And we have we actually doing a project right now, which is which is in great detail, outlining what happened in Bangladesh in 2007 and eight, when we had a big bang, exactly. This type of top down anti-corruption, which not only did not work, actually change the political settlement in a very adverse way and created an authoritarian regime. Right? So a kind of unintentional consequence of that. And so these big banks and I was mentioning even more dramatic examples of, you know, Americans going and doing democracy in Iraq and so on. But even on a smaller scale, when when you have these big bang anti-corruption, the evidence is very clear. They are not sustainable. And they actually you move backwards when you try to do these things. So the incremental is not guaranteed success. It might be pushed back and it will be pushed back. Out of ten things we do, five might survive or three might survive. But that is the way of progress everywhere. Can you guarantee that everything that's being tried in the UK will not be pushed back? The next government might come and so on and so, but that's how progress happens. It happens with gears and you move forward. So the question about environmental pressures and so on I think, and someone else had a question about ethnicity and identity and so on. All of those things are important. But for us the most fundamental thing is the self-interest of the actors. And that connects to also the behavioural economics question. A lot of behavioural economists don't have a materialist approach to interests. A lot of behavioural economists are really talking about interest as kind of preferences. No, I prefer this or I prefer that and so on. And I think the three of us.
SPEAKER 7
And so as Jemmy.
SPEAKER 2
Perhaps even James. We come from a very materialist background. So the materialist background is we look at the materiality of interests that if I really don't have, if I if I don't have a material interest, I'm not going to go out in the midday sun with an umbrella, checking how an embankment is being built. I will go out there. If that embankment affects the speed with which I can take my grain to market, then I will go out there. I will not go out there because it's the right thing for me to do. I might do it for 1 or 2 days, but I won't do it in a repeated, sustained basis over long periods of time. The materiality of interest is very important. And then actually ethnicity and all these things matter less. What you find in developing countries is that people deploy identity, ethnicity and all those things for extractive rent collection. I want to construct my own ethnic group, my own caste group, my religious group. That gives me what I was telling you earlier about organisational power to contest for rents with other groups. But, but and those because they are so deep rooted in developing countries, it's a real curse. We have all these different ethnic castes, tribal and other associations.