Anirudh Burman on Rethinking India’s Land Regulation

SHRUTI RAJAGOPALAN: Welcome to Ideas of India, where we examine the academic ideas that can propel India forward. My name is Shruti Rajagopalan, and I am a senior research fellow at the Mercatus Center at George Mason University. 

Today my guest is Anirudh Burman. He is an associate research director and fellow at Carnegie Endowment India, and prior to that, he worked at the National Institute of Public Finance and Policy and the Centre for Policy Research, both in New Delhi. He has his master’s in law from Harvard Law School. We spoke about the dysfunctional land markets and the kinds of reforms required in land use policy, land sale and land transfers. We also talked about the various experiments with land pooling and leasing in India, how to think about eminent domain law, land titling, land title insurance and much more. 

For a full transcript of this conversation, including helpful links of all the references mentioned, click the link in the show notes or visit mercatus.org/podcasts. 

Hi, Anirudh, welcome to the show. Thank you so much for doing this.

ANIRUDH BURMAN: Hey, Shruti, thank you for having me.

RAJAGOPALAN: One of the reasons I follow your writing and I follow your work is to try and make sense of India’s structural transformation and your work on state capacity, land policy, urbanization. In a sense, what you’ve been doing while tracking the administrative and regulatory state and how it has been changing has been incredibly useful to me. 

In the last few decades since the structural transformation started with the post-1991 liberalization and so on, there are two major things happening. One is, of course, the exit from agriculture to industry and services, and that is really the sectoral part of the structural transformation. The second is the rural to peri-urban to urban migration, which is deeply entangled with the first part of the sectoral transformation. It’s also its own sort of thing. For both of these, the crux of the problem that we always land on is land policy, and in particular, land use, sale of land, the rules surrounding that. Pretty much everyone in the business is in agreement that India really needs land reforms, but no one seems to explain what that means.

The last time we did land reforms, it was all about zamindari abolition and redistributing land by taking it from large zamindars and giving it to farmers who didn’t have land and so on. The first thing I want to ask you is can you just give us a sense of what the land market in India looks like? Once we understand that, then we can pick that apart and discuss more details.

How Does India’s Land Market Work?

BURMAN: Sure. I think that’s a great way to start off the conversation. I think one insight about thinking about land markets is that when you are looking at a market like land, which is a factor market, a good way to understand it is to see how well product markets that depend on land are working. If you are looking at housing, if you are looking at manufacturing, if you are looking at agriculture, which all consume land, and see how well they are working and whether land or land use is actually a constraint in the functioning of these industries.

In India, we clearly see that all these markets function. It’s not that we have a dysfunctional land market. We’ve grown significantly. We’ve had reasonably high rates of growth since the mid-’90s, till say, mid-2014, ’15, that time. We are again beginning to see some growth now. To the extent that land is used as a factor of production, yes, land markets are functioning. At the same time, we do see other problems of the land market, which is housing is expensive.

It’s not very easy to get land for, say, manufacturing or industry. It’s not very easy to convert agricultural land for nonagricultural purposes. Industries are always on the lookout to get suitable land to set up their manufacturing units and so on. There’s clearly a problem. Our cities are highly dense, highly crowded, probably underbuilt. We are seeing a lot of problems in terms of, say, transaction costs when we transact in land. It’s not dysfunctional, but it’s not working very well.

RAJAGOPALAN: Here, I think you’re absolutely right to land on the transactions cost, but before we get into transactions cost, I want to sort of frame this conversation a little bit more in the context of the whole economy. I guess, if there is a dharma for economists, it is that resources flow to the highest valued use, and anything that gets in the way of that, including transaction costs or other frictions in the market, one would like to remove those kinds of friction so that we can make the most productive use of scarce resources. That’s the sort of the tent pole within which I try and think about the entire problem.

With land, there are two specific issues. One is that land is immovable, so when we say it needs to flow to its highest valued use, it’s not like human migration, or it’s not like factory units or cold storage warehousing where you can sort of move things from one place to another. It’s really wherever that land is you’re kind of bound by that and now the entire conversation is about what that can be used for. Now, this gets into the question of not just are you supposed to grow food to eat and produce on it or are you supposed to build housing on it or are you supposed to build factories and so on. That’s one part of it.

The second part is also how much can one build. What are the rules around that? I want to try and understand, what does that look like? Because that’s also at the state level. What does that landscape look like? Then you can walk us through, is this a high-transaction cost enterprise? Is it a low-transaction cost enterprise and so on?

BURMAN: Well, absolutely. I didn’t want to give the impression that the problem is certainly one of transaction for us.

RAJAGOPALAN: Yes, of course.

BURMAN: I think there are all kinds of restrictions on land use that are equally problematic, and then they create the suboptimal use of land. We are always in this discussion on land is scarce in India, which is a ridiculous idea because there’s no scarcity of land. It’s artificial scarcity, which is created by our regulatory framework. When we look at, say, land use in states, you see many examples.

If you’re looking at cities, there are restrictions on FSI [floor space index], on how much you can build on top of a particular set of land. That inhibits vertical construction in cities and therefore it leads to urban sprawl. Then you have these additional discussions on how land is scarce, and how agricultural land is being converted into, say, land for housing and so on. 

Similarly, you see these kinds of restrictions on the ability to convert agricultural land for nonagricultural purposes. Whether it’s for expanding cities or whether it’s for industrial units that want to create new manufacturing facilities or expand them. There’s this perpetual discussion about can you convert agriculture land. Who can buy agricultural land and who cannot? Many states have restrictions on nonagriculturalists buying agriculture land, on people from outside the state being able to buy agriculture land. This creates this whole perception of artificial scarcity that actually leads to this very suboptimal use of land available in India.

RAJAGOPALAN: Here, I want to parse out two things that you’ve said, or rather, there are two categories that you’ve discussed. One is a set of restrictions or regulations or laws that dictate land transfer. The second is those that dictate land use. These are categorically different in the sense that the first is about sale, lease. You’re actually having some kind of transfer of ownership or possession from one group to another or one individual to another.

In the second one, it could be in the hands of the same individual, but you could change the land use, you could increase the FSI, which is the floor space index requirement—or restriction, rather—and allow people to build. Can you walk us through a little bit more detail on these two broad categories? I understand there’s differences between states, but maybe some examples of what these restrictions look and feel like.

BURMAN: One great example of what you can do with your own piece of land is your building bylaws. How much of, say, one-acre plot can you actually use to build a building? Many building bylaws will say you can use only 30% of the land, and you have an FSI of 101.5. That means you can actually not use 70% of the available land. Again, that’s an artificial scarcity because you own a one-acre plot, why should you be able to use only 30% of it?

If you do that at a citywide level, you are actually wasting a lot of space that could otherwise have been constructed on, and you are creating artificial scarcity within the city. You are actually not using that space at all. Dr. Bimal Patel and others have done a very nice paper on it demonstrating how Indian cities actually waste a lot of private space, which can be used for public use. They’ve shown how actually Indian cities do a particularly bad job of it compared to many other good cities globally.

The other example of it is just plain FSI restrictions. Indian cities have many restrictions on how high you can build on top of a particular piece of land. Compared to most other global cities, Indian FSI regulations are quite restrictive. They’re getting gradually better now, but I think we have a long way to go. 

RAJAGOPALAN: Before we get into transfers, there’s another kind, right? Which is, if you’re in a city, the question is can you use the same property for residential use or commercial use or quasi-industrial use because you don’t have too much heavy industry in dense areas? In rural areas or peri-urban areas, it’s a question of do you use it for agricultural use, residential use or industrial use? That is another bizarre land use restriction. It’s a little bit different from what you were talking about before because that inhibits new building on the same property. Here it’s a question of this capacity has already been built, and yet, we won’t allow someone to use it in a slightly different way should they wish to.

BURMAN: Yes, absolutely. I think zoning is one of the chief ways through which we do it. I think both U.S. cities and Indian cities are quite aggressive about zoning and that’s developed costs for all kinds of businesses, housing, quite significantly in India. What it’ll do is to say if you have land in a residential zone, you can only build residential construction on it, and you can only build housing on it. You can’t set up or open a grocery store there. I think we’ve gone through the historical experience of having that kind of a regime for a long time. We are gradually now incorporating mixed-use plans, slightly more flexible zoning.

Again, we have a lot to do in terms of allowing much more liberal land use in these cases. I think planners also worry sometimes too much about whether, let’s say, activities “causing nuisance” will be developed or set up in areas where people live or reside. I think one thing people don’t consider is that just pure land economics means that very often you can’t set up the kind of industry that would create a nuisance in the middle of a city.

You can’t create a large manufacturing unit in the middle of a city because you can’t get that much land. You would just be paying through the nose for the land and labor would be equally expensive. It makes a rational sense for an entrepreneur to go locate that kind of industry on the outskirts or maybe even outside the city. Zoning has tried to basically restrict the normal activity of the market in order to control the development of the city, but not taken into account the fact that economics itself takes care a lot of the problems that it tries to solve.

RAJAGOPALAN: It’s actually almost worse than what you’re saying. Because normally, let’s take the example of something like now Mumbai, erstwhile Bombay, there were all these textile mills which were right in the middle of the city. In fact, they were in the downtown area. Even after all the residences developed around it, the powers that be never relabeled or allowed different kinds of land use.

Now, they are these tony, posh cafes and malls which can continue some commercial and industrial activity, but it’s basically dead space where you have a cute cafe bang in the middle of the city and it’s caused so many downstream problems, which are in some sense, some of them are nuisance bearing. One is the downtown has shifted, which means you need to dig up more for new roads and you need to dig for metros. Now, it’s not even a joke or a punchline in Mumbai. It’s just a state of affairs. 

The second part of the nuisance value is because of this artificial shortage that’s been created, slums emerge. You either have residential areas that are pure wonderland. Then you have slums which are pure nuisance. I do mean in the classic tortuous sense of nuisance, which is there are problems with sanitation, there are problems with noise, there are problems with smell and all those other things that we consider. It’s almost two steps worse than what you are describing in their failure to truly understand what land markets do.

BURMAN: Absolutely. I think slums are a classic symptom of scarcity. It’s literally a manifestation of not thinking of the fact that land markets in a city have to cater to all sections of society. You cannot have these aspirational or utopian ideas of what you think is good housing only to make it unaffordable for half the city. It’s a travesty that even until recently, many master plans would not actually account for slums or slum development.

We only started doing that once the problem really got out of hand. We need much more organic and flexible planning when it is required because I don’t think you can do without any planning, but you do need much more flexible dynamic planning that is also oriented to the economics of how the land market functions.

Why Do We Have Such Bad Regulation?

RAJAGOPALAN: I love complaining about our current planners and municipal corporations and so on, but I guess just to back up, where did these classifications come from? Because it’s not the creator of the earth and the universe, whoever that may be, or no one as it may be the case, started labeling spaces and saying, this is agricultural, this is urban, this is rural, this is industrial.

Now, these labels have come from somewhere. I imagine in some places we’ve inherited it from a colonial government. In some places, it goes as far back as the Mughal times and so on. Where did these labels come from and why are they so inflexible?

BURMAN: I think there’s always been a question of labeling, and it has its origin in agricultural land because, at one point of time, land revenue was the main or the major source of revenue for all states. There was a huge interest in making sure that the amount of land that is configurable in a particular kingdom is being cultivated and the correct amount of revenue from that land is being collected. There was a lot of emphasis put on mapping agricultural land and making sure that the cultivator is properly mapped with the average amount of produce from that land and is properly recorded and so on.

That has carried on. I think when cities became more and more important to our condition, to societies, I think that classification system carried on. With regard to, say, much more intensive land use restrictions and classifications, I think it’s something that originated both in Europe and U.K., which had its conceptions in this notion of being able to plan cities according to where you wanted people to live and where you wanted them to work and so on.

Then of course in the early 20th century, the U.S. took it to a new level. India also used and incorporated a lot of British notions of land use classifications and planning. One thing that is sometimes forgotten is that residents also have an interest in making sure that these labels of land use classification stay. Because if you become a landowner and the land use classification has a role in making the value of that property keep on appreciating, you don’t want the classification to change, right?

If you bought land in a coastal regulation zone and you were able to get an exemption there, you’re the only person, maybe one of five people who got that, you don’t want that classification to change because the value of your property is directly related to that exemption and that classification. I think there is a mix of this top-down vision of what a city or a society should look like and organic response from people who benefit from it, who have an interest in making sure that that classification stays.

RAJAGOPALAN: The NIMBYism that you’re talking about, as we call it in the U.S., that’s pretty prominent. But I want to go back for a second to the point you made about agricultural land, right? James C. Scott has written about how agricultural revenue was the easiest to extract, pretty much all nation-states are sort of built around how do we maximize agricultural produce and agricultural revenue coming out of it.

One thing we’ve seen since the birth of the republic in 1950 is there’s a lot of continuity in other colonial laws, but when it comes to agricultural revenue, there’s almost a clean break. Indian agriculture was enormously taxed and then a part of the founding of the Indian republic was that that will no longer be the case. Not only will we abolish the intermediaries who collect those rents and revenues and pay those further up into the colonial food chain, we are not going to do any of that.

Now it seems a little bit bizarre to me that that has continued. That’s, I guess, part one of the question. And then part two of the question is given that that’s continued and now we are 75 years almost removed from that moment, we still don’t tax agricultural land. What is going on in that system?

BURMAN: Yes, absolutely right. We basically decided not to tax land revenue because of the entire legacy of colonialism and the fact that the land revenue was seen as exploitative. And one of the promises of becoming sovereign was that this exploitative system would not continue and, to some extent, we basically delivered on that promise. I think why it has continued till this date is because agriculture continues to be nonremunerative. I don’t see any other major factor behind it. We still have a large class of small and marginal farmers.

We still have fragmented land holdings, and therefore, there is no political capacity or will to actually invest seriously in land revenue administration in agriculture. To the extent that while doing a study in Rajasthan, some years ago, I was told that the administrative cost of running the office of the Patwari who goes out to collect land revenue is higher than the land revenue collector.

We’ve basically done away with it for political reasons, for the reason that it was exploitative under the British. Our failure to make agriculture remunerative has meant that we can’t actually think of taxing it seriously again.

RAJAGOPALAN: My question is, given that we’ve continued the system and we’re still not taxing agricultural revenue in any meaningful way, why do we still hold on to the label? Isn’t it in everyone’s interest to allow for more change of land use, let people switch it to different uses that then can be taxed under the various other things that we’re talking about, property taxes, or it can come under GST in the case of any commercial or other kinds of use, so why aren’t we doing that?

BURMAN: I think the reasons are similar. We haven’t thought about how we are going to ensure that we can maintain the income security or the economic security of the cultivator when the land is transferred to a noncultivator or an outsider. That creates this political risk that if you allow free transfers of land from agriculture to nonagricultural, you might conceivably see a group of farmers or groups of farmers who are angry with this change in regime.

Because we are so worried about political stability, no government seriously wants to take that risk. If you look around, some governments have taken that risk. For example, Gujarat, Karnataka, Haryana, a few others have actually made these changes where they’ve allowed outsiders to come in. For the most part, it is seen as a political risk. I think there are also other reasons. For example, a lot of hill states, there is this idea that you don’t want outsiders to come in, buy property in the state.

That’s got nothing to do with agriculture or urban or whatever. It’s just an idea that you don’t want local land to be held by outsiders. It’s not purely a question of whether the economics makes sense, but I think there’s a lot of the political that goes into these decisions.

Land Transfer and Consolidation 

RAJAGOPALAN: This brings me to the second part of some of your writings. The first problem is this land use labeling and mislabeling and that creating transaction costs and frictions. The second is just the crazy rules surrounding land transfer. You’ve written about this before, but here we are talking about not just who is allowed to hold the land. Like you were talking about in northeastern states, they don’t want someone outside the domicile but also within those states, you can’t sell it to a nonfarmer or a nonplantation owner and things like that.

An individual or a family because we have the tax category of Hindu undivided family, there are limits to how much land or property each individual or each family can own. The numbers there, or rather the ceiling there, sometimes is ludicrously low. I live in Virginia. I have neighbors and friends whose backyards are larger than the restrictions that the Indian states puts on some of these farmland sale requirements or holding requirements. This is a second part of the puzzle, right? The first part of that structural transformation is the resources need to go to their highest valued use.

A subpart of that is even agriculture could become more productive if the landholding sizes got bigger. Landholding sizes aren’t bigger because we put in these crazy artificial restrictions of smaller landholding size plus we redistributed and broke up large zamindaris into smaller farms. Now, two or three generations later because we don’t have rules of primogeniture, you keep getting smaller and smaller landholdings.

I believe the average holding size is about two acres, just over a hectare. It’s pretty small now. Why isn’t that another place where you change? I can understand that there’s this fixation with agriculture. We don’t want to have outsiders come in. We have a romanticization with villages and rural areas. You don’t want to convert them. The same thing could also rearrange within itself. People could actually exit the village and go work in industry or services or take the IAS exams or whatever it is people wish to do. At the same time, the resource which is land could be put to a higher value use even within agriculture.

BURMAN: That’s a great question and I’ve been thinking about it. I don’t have a great answer but let me give it a stab. I think you’re completely right. Actually, land fragmentation has increased over a period of time and largely due to succession. I think Dr. B.R. Ambedkar wrote a very prescient essay on the same topic. He basically foresaw what is happening today.

He, again, made the very simple argument that if you want people to prosper, stop focusing so much on what inputs they need to cultivate their small plot of land, develop alternate avenues of employment and opportunity and the land and agriculture will take care of itself. I think to some extent that is the problem. We have not been able to create those alternate avenues, and it’s sad that agriculture is one of the fastest growing sectors of the economy in the recent years. 

After a long period where the Indian economy grew in services and manufacturing now, we are disturbingly seeing a reversion to agriculture. I think there is a problem in the sense that we are not creating the kinds of jobs that will wean people away from agriculture. I think that is in turn creating this stagnancy where people feel that they need to depend on the land that they own, even if it’s a small plot of land. It’s an insecure existence and one shock could push you back into poverty and you need the only asset that you own, even if it’s small, even if it’s inefficient to cultivate it.

You basically need something to fall back upon and I think that’s the basic problem that we are stuck with. That how do you create these alternative systems where people can actually exit agriculture efficiently? That’s why I think the quality of urbanization also matters a lot. Can you make cities much more affordable? Can you make it easier for people to find a rented accommodation rather than to live in a shanty on the outskirts?

Can you get them someplace where you don’t have to commute 25 kilometers? Can you create a similar social security net in urban areas as you have created in rural areas? I think the entry and exit from agriculture to nonagriculture in terms of what economic opportunities and avenues you have, if we don’t think that through clearly, I think we are going to be stuck with this feature of the economy where everyone has small land parcels that they are cultivating inefficiently.

RAJAGOPALAN: This is where it starts becoming a little bit crazy when I look at it. You’re absolutely right. People try and attach a lot of value to this one big asset that they have. They don’t want to sell because the urban landscape, forgive the pun, is a little bit complicated. There’s a lot of uncertainty. There are lots of informal employment and labor markets, informal housing. The only true fallback is that one asset that they hold back in the village or in the agricultural area.

Now having said that, the value that is attached to that asset entirely depends on the use of that asset or the permitted use of that asset. I grew up in Delhi and in the outskirts of Delhi, Gurgaon was developing. Of course, that was not the most productive agricultural land. It’s pretty barren land in the outskirts of Gurgaon, but there were farmers, people were doing some dairy and produce. There was some farm activity going on there. Initially, the chief minister of Haryana’s office and the government really, started consolidating land, giving it to big developers like DLF and ANSALS to develop large tracts.

As that process continued, the farmers got wise and they understood that what’s happening is a bit of a scam. What was really going on is the value of that tiny land holding, when it is dubbed for agricultural use, let’s say that value is X, the moment you can get a change of land use certificate, either for residential or commercial or industrial activity, then the value of that land shoots up. Normally, in most peri-urban areas it’ll shoot up, say four or five times. Now, in the case of Gurgaon, things were happening so quickly, that sometimes the difference between the prices were X versus 30X.

What happened is that the stage two and stage three consolidation efforts should have actually been very smooth. People were dying to sell their land and exit unproductive agriculture. The developers are dying to buy land because every Fortune 500 company wants to set up in Gurgaon, but they’re not able to connect with each other. Plus, no single person has a large enough tract of land.

Now, the government has to step in, use eminent domain, take that land. The compensation they pay for the eminent domain will be based on the circle rate of the agricultural land, not after change of land use. Then you get all these protests and then you have a bunch of sociologists writing papers about how developers are extracting from farmers using the political system. This is truly bizarre now.

To me, in my head, and maybe I’m being too simplistic about it, if you allowed the person who already holds the land to convert the land use easily, or to transfer the land or to buy up larger tracts of land, if you didn’t have all these land ceiling laws, then none of this would be necessary in the first place. This entire mess has been in some sense created by the regulatory framework.

BURMAN: Yes, completely. I think one additional point is also that capital investment is not easy. You generally see outsiders coming in assembling land parcels because the farmer, himself or herself, has too small a plot of land. You’re always in this situation where because they don’t have sufficient capital or the wherewithal to go and borrow, you will see someone who has that ability to either borrow or to invest, come and assemble land and basically buy out existing landholders.

That’s a uniform occurrence wherever this kind of transition is taking place. I think what is important there is what is the role of the state as an intermediary and are there clear rules about it actually intervening and is it doing so in a transparent and fair manner? The problem that has occurred in India always is that land acquisition has not gone off well. Often, it’s happened in the situations what you’re describing where a private land assembler or an aggregator or developer finds the cost of land is too high if privately negotiated.

Therefore, it gets the state to intervene to acquire the land because even if they pay the state the cost of acquisition, it’ll still be lower than having to pay the negotiated market price. I think that’s where we’ve not done well. In that context, I think the 2013 Land Acquisition Law that was enacted was actually a good step because it made the cost of acquisition that much more expensive.

To that extent, it forced people to do private negotiations much more than to get the state to intervene. It both basically made public purpose acquisitions more difficult by narrowing down the definition of what the public purpose is. It also said, “We will give you more compensation, you’ll have to go through a longer process and so on.” It actually forced entities in the market back with a negotiating table where acquisition was often seen as a way out.

RAJAGOPALAN: I want to talk about acquisition separately because there are a few other pieces of the puzzle, and land use is one of them, and transaction costs is one of them. To stick with land use, do you think one way of solving the problem is just right now most of the rules in India say that the person who buys the land is the person who has to petition for change of land use?

Is it possible to have slightly more uniform rules? Not label very, very specifically, but have general uniform rule saying roughly residential, roughly agricultural, roughly industrial, roughly commercial. The person who is selling the land can also petition for the change of land use. Then if the buyer further wishes to change the land use, they’re always allowed to do that. Right now what’s happening is to change the land use, you need to submit near-perfect plans of what is eventually going to be built in that place.

What happens is that the farmers who are trying to exit agriculture can never capture the rents that you can get from the change of land use certificate. That’s actually the biggest holdout. The holdout is almost entirely a regulatory holdout. It’s not an actual holdout where one person is being very opportunistic and trying to extract a pound a flesh from a developer or an industrialist. Am I being too simplistic in thinking that something that simple could change things quite a bit?

BURMAN: I would amend that a little bit by saying, I think that information asymmetry is a problem. It can be overcome if you actually have land use plans that are clearly transparent. For example, if you say, over the next 20 years, we intend to develop X, Y, Z areas as an industrial cluster. Everyone in that area knows that there is going to be a change in land use from agricultural land to an industrial cluster. People are going to be coming and looking to buy that piece of land.

That information asymmetry is sorted. If you are in a situation where you are saying, we are going to allow a private developer to develop this 10-acre piece of land tomorrow, then we’re going to allow him or her to develop that 20-acre piece of land tomorrow, that ad hoc is detrimental to creating information symmetry between the buyer and the seller. Because then the seller has to actually try and understand what the buyer is buying the land for.

I think that is one way to solve this problem—because I think you struck at the heart of it, which is if the negotiating parties have good information, they should be able to negotiate a proper price where the landowner is able to capture the rents from that transaction.

RAJAGOPALAN: Yes, or the rents disappear, basically. That’s the bigger part of the puzzle. One way of characterizing it is as regulatory rents. The other way of thinking about it is we’ve almost created two separate markets in land: one with the change of land use certificate for the new use, which is of higher value, and one without. Farmers are actually selling in one market, which is already depressed because you can only sell to farmers. You need to depress the prices in the market because of artificial reasons. Then in the other market, the sky’s the limit on how high the land price can go sometimes, especially in places like Gurgaon. It’s almost a question of either rents can completely disappear or dual markets. 

I have a further question on this. The way you phrased the answer was we can have better information and transparency. If the government or the municipality or state government, whatever is the appropriate level of government clearly announces that we are going to develop this area, isn’t that the upside-down way of thinking about it because areas do just develop? Thane developed. Thane used to have dairy farms and used to have orchards and now all of it is startups and fintech companies and things like that. Areas develop bottom up. Isn’t the better thing to do to say that everything will be open land use? You can wish to build anything in any area, except if you really want to be build heavy industry, or chemical industries, or let’s say you want to build a zoo, which is going to have tigers and pythons, then you need an extra layer of permissions and things like that.

Otherwise, it’s going to be super easy to have any land for commercial or whatever. Basically what I’m saying is, why isn’t all land at a default level mixed use, with very few restrictions for certain kinds of nuisance uses? You’re smiling…..

BURMAN: Yes, I really wish that were the case, and I agree that is the utopian ideal of what the regulatory framework ought to be like. My answer was through the fact that we have a particular regime in place right now, so how do you transition more effectively from that so that your farmer can actually benefit? I agree there should be a negative list of land use plans, and I think that is the position of the common law and tort law where nuisance-causing activities are basically not permitted in this core of the city or next to residences and so on, and they have to move outside.

It also goes back to that idea of zoning where people used to zone areas and you can’t have an industry here, but actually, under tort law and common law you got the same answers. You can’t have industries there but we don’t need zoning for that. But to get back to the point, I think the problem is that we have a particular regulatory regime today that says that this land is agricultural land.

The question is how do you transition in a way where the person who owns that land can actually get a fair bargain? I think one way to do that is to give ex-ante  information, saying, “We are going to develop this area.” The other is, of course, to remove sources of regulatory discretion and rent seeking, make some processes more clear. 

For example, some city land use certificates, that I was looking at, say that your redeveloped area has to have access of a road which is 33 feet wide—now, every municipal administrator in that locality will know which roads are 33 feet wide. You don’t need a site visit to see whether that’s 33 or 31. That’s a requirement which is just a custom built for rent seeking. A lot of these can be digitized, they can be done visually on a map. They can be done online, you don’t need any scope for the inspector to go and “inspect” the road and so on. I think those processes will also help in curbing a lot of the problems that come with this change in land use certificates which have often become a problem.

Transitioning to an Efficient Land Regulation System

RAJAGOPALAN: I completely appreciate your point that you are thinking of this as a transition problem. Like, how do we solve today’s problem given a particular regime, and then transition to the next thing and the next thing? I’m thinking of it slightly differently, almost as 100-year problem. Let’s say today we figure out a system where top down, we say this is going to be an industrial area, or this is going to be developed for a certain kind of public utility or electricity or solar or whatever the government plans.

My problem is given the pace at which the structural transformation is taking place—because India’s not growing like the developed world at 1.5%, 2%. India’s growing at 6%, 7%, and we think that’s too low. We want that to also go into some kind of Chinese miracle rates or hyperdrive. That will mean that very quickly that same thing is no longer going to be the highest valued use.

What you’re going to get is Thane will become industrial activity, but then you’re going to have the same problem in a place like Kalyan or Thane. Which is what happened with Phoenix Mills and things like that, which is tomorrow when you need to transition out of industrial activity to services or to, I don’t know, GPU centers, I don’t know what the next use even is, what happens then?

In some sense, I feel like if you really think about this as 100-year problem for a very fast-growing economy, there’s almost no other way to think about it except reset the default. You need to dismantle the whole damn thing and start ground up. Sorry, there are too many puns in this question of land, but it just feels like that’s the only way out. Do you feel a little bit more optimistic than me?

That, let’s solve today’s problem, people will become richer, they’ll become more productive. They’ll get integrated in a different part of the economy and the global economy, and then they will also come to realize that these kinds of transitions are good. We’ve done it once before, we’ll have some institutional memory of doing it again. Is that where you are coming from or is it just forget about the 100 years, we just need to solve this right now? Which way are you thinking about this problem?

BURMAN: To some extent, I’m optimistic that the way we solve these problems will help us solve future problems. For example, if we take the case of change in land use, I think one thing that happens is the value of land and real estate becomes so high that no city administrator can ignore that value anymore. 

Fifty years ago, you could afford to say, I’m going to have this whole plotted development where everybody will have a lawn at a small plot of single-story houses with a backyard and a front yard, and so on and so forth. Now you are basically saying, how can I build a 20-story apartment on it? That flexibility of land use will come even from the fact that the market value of land and real estate is increasing because cities are growing, becoming more productive. The demand for land and real estate is increasing.

I think part of the optimism comes from that, that if you can solve the problem of today’s transition, you are going to create land markets and real estate markets that are so much more valuable that you can’t ignore the economics of it to the extent that you have done over the last 50, 60, 70 years. Having said that, I do take your point, that it’s not a given that over a period of years, people will not have bad ideas.

RAJAGOPALAN: Like Phoenix Mills, right?

BURMAN: Correct.

RAJAGOPALAN: To me, every time I look at that, it’s almost the blood boils. We’ve known this is a valuable piece of land. This is surrounded by some of the most valuable property, not just in India, but in the world. We have somehow still managed to make a mess of it.

BURMAN: Yes. I think that goes to, again, the point that you are making, which was that some of these problems might not be solved even if land values escalate, which is if there is a Phoenix Mills coming up, or if there is a row of redevelopment on that road, why isn’t the planner or the person who allowed that development to take place, why hasn’t that person or that institution made sure that the road infrastructure keeps up with that development?

RAJAGOPALAN: Yes.

BURMAN: Other aspects of, say, sewage, sanitation, drainage, you’ll see water flooding up to knee-length every year there during the months. When that redevelopment took place, it was a golden opportunity to solve those problems. It was somehow not taken or not given adequate attention. I think that gap between the state capability and the growth and dynamism of the market is something we are still figuring out. I don’t have a good answer to that, but I agree with you. I think it is a 100-year problem, and it’s not easily solvable.

Eminent Domain and Land Titling 

RAJAGOPALAN: I want to now come back to the topic that you were alluding to a little bit before, which is about using eminent domain for land consolidation and so on. I’m a little bit obsessed with this eminent domain question. We have a very bizarre problem. Again, I’m now discovering that this is a 200-year-old problem. Now, we’ve never had good land titling in India, so land titles are always questioned. You’ve written about this. Some of the problems are that it’s not a conclusive title. It’s a presumptive title, and that’s, again, been a historical legacy. We never really switched out.

Now, when it is a presumptive title, the alternative is you have really good land registry systems that the government maintains, even if they’re not perfect. Then you can layer some title insurance or something else on top of it, and you can get somewhat functional land markets or property markets. Instead, what we’ve had is, we’ve had at every level—and this goes back even 200 years ago with the East India Company—when people can’t understand what belongs to whom, or if you bought it from someone if it’s not conclusive, it can be challenged anytime, immediately that title is a little bit uncertain.

They started using eminent domain, first, to build railroads and then to build other kinds of industry, including TISCO and all sorts of things because the one way you can extinguish any other interest in a piece of property is by using eminent domain. The moment you use eminent domain, no one else can come and challenge that title tomorrow and so on and so forth. If you are in an economy where there’s very quick structural transformation, industries obviously need land and sometimes, for heavy industry, that land is a very particular place. It needs to be near a river or it needs to be near a particular kind of port or something like that. So it’s not like you can just move a mining industry or an aluminum smelter somewhere else. These are very particular problems.

They prefer to buy the land from the government even if people in that area are willing to sell it and they know that that area is industrializing. We have a second bizarre problem. People want to sell land, people want to buy land. One is the transactions costs and change of land use that we spoke about earlier. The other part of it is there’s just so much uncertainty surrounding the title that you can’t put a multibillion dollar project, throw it into complete litigation and legal uncertainty because one or two people will challenge the transfer of the title. The best way to do it is let the IDCO of that particular state, which are typically this industrial development corporation of Odisha or Gujarat or Maharashtra or whoever or the infrastructure development. You let those people consolidate the land and then sell it. That’s how we’re doing it.

I understand that the Land Acquisition Act of 2013 can get people better prices, but it still doesn’t change this fundamental problem that the reason why eminent domain is being used in the first place is not just about price and transaction costs, but it’s about some other uncertainty which has to do with titling. How do we solve this titling question? Sorry, that’s a small question.

BURMAN: I think one of the things we see is that the land titling problem is not just a titling problem. It’s also a political problem, especially when you think about acquisition. Let’s think about, say, Nandigram or Singur. Basically, a dispute over compensation can easily escalate into a political problem that makes the entire enterprise at risk. What would you expect in a normal environment that you acquire, say, 100 titles of land, which let’s say occupy 100 acres, and let’s say there are disputes in about five of them.

In any normal enterprise, you would be able to price in the cost of that risk and get some kind of indemnity or insurance for it and basically say, “Okay, there’s a likelihood that 5% of the land titles will be faulty and I’m going to solve for that problem either by bank insurance or whatever else is available as insurance.” The problem in India is that if 5% of the land titles are faulty, the local farmer group, the panchayat, the politician, everyone gets involved and your entire enterprise is at risk. 

RAJAGOPALAN: The litigation risk?

BURMAN: Exactly. You have this curious problem where what should ideally be a normal financial risk becomes a disproportionately large risk to viability. That’s the problem that acquisition is trying to solve. It’s not actually trying to solve the 5% of land titles problem. It’s trying to solve the problem where your entire unit can shut down tomorrow because 5% of land titles are faulty. 

Ironically enough, it’s given rise to the bigger problem where the state often acquires land at cheap prices for private use. At least till 2013, that was happening on a larger scale. I don’t know the extent to which it’s happening now. I think it’s become much more viable for farmers to sell now. That’s broadly the acquisition risk that we see in India. 

Let’s say in the context of acquisition, I think sometimes a titling problem itself is overstated in the sense that we overestimate the amount or the number of titles that are bad. I think the problem is not that there’s a huge number of titles that are horrible, it’s just that the number of titles that matter that should be fine, are not fine.

RAJAGOPALAN: I would actually further add to that. It’s not that a huge number of titles are bad, but the problem that if you roll the dice and it’s bad then it’s protracted litigation for 30 years or giant payouts or enormous amount of uncertainty, which for an asset like land, it can just kill the deal basically. I think that’s the bigger issue, right?

BURMAN: Yes. Also, the other thing is that we actually don’t have very granular data on what is going on with land titles. We have perception surveys that a lot of disputes in courts are property-related disputes, but to say that those are all land title-related disputes is a bit of a stretch. 

RAJAGOPALAN: It could be inheritance, it could be something else.

BURMAN: Exactly. You haven’t paid stamp duty correctly, you are squatting on my property, whatever it is, right?

RAJAGOPALAN: Yes.

BURMAN: We don’t have that kind of granular data to say, “Okay, all our litigation is because of bad land titles.” Let’s assume for the fact that we do have a land titling problem—and I won’t deny that there are problems because of land titles. What we see is that because we have this presumptive titling system where anybody with a better title or who can prove a better title can come and show that, actually, you are not the rightful owner of the property, and so on, and therefore, their sale deed or their claim holds greater value than yours in a court of law. That often leads to problems because it’s hard to then ensure that there is no risk to the title that you’re buying or selling.

The problem in India is because you are actually creating these two systems of registration and mutation while at the same time, we are going through this period of transition. You first have to register your land deed or your sale deed with the subregistrar’s office, and then you have to update the record in the Land Revenue Department’s land record, the mutation process. 

Now, some people often forget to do one or the other, or mostly the second part, which is the mutation. That creates a problem because until that is complete, your name is not reflected in the land record, and therefore, you are not considered the presumptive owner of the piece of property. That is a problem, and yes, land records need to be updated and corrected.

I don’t think doing this mass land record digitization is the only answer to the problem. In India, we’ve been doing this for 20 years. We have a Digital India Land Record Modernization Program that we’ve been running for 20 years. The idea is to then, once this digitization is complete, is to move to a system of conclusive titling where if your property details are mentioned in the land record, that is conclusive proof of your ownership of that piece of property. That’s the ultimate agenda, but we haven’t reached that goal yet. I think we need other approaches to actually solve that problem. 

RAJAGOPALAN: Before we get to the other approaches, just to understand this digitization process or what we’re trying to do, to me, it seems like it’s a slightly confused enterprise because with digitization, what they’re doing is simply that there will be a digital register which will reflect the physical register as accurately as possible, right? The problem is that the physical register is not accurate, and we haven’t figured out how to plug those gaps. There are a few different issues. One, most rural areas don’t collect property tax or agricultural tax so there’s no incentive for the local government. Even if they collect it, it’s not by the local government. We have that other problem.

Local governments would have an incentive to know what belongs to whom with very little error if they could collect taxes on it and actually pay for the registry themselves, but they don’t have that incentive so they don’t do it. Then you pointed out that people always forget to register mutations or sometimes they try to avoid registration and stamp duty so they don’t actually have a clean thing. Plus the registrar itself is probably one of the most corrupt offices in the country. There are massive rents to be made there so they can make a lot of money by tampering with the register without paying any personal liability. This is where that system is. To me, digitizing a flawed physical register doesn’t solve the problem. It just makes it easier to figure out that it’s a flawed physical register. That’s part one. 

Transitioning Land Title Systems

Part two, is the transition from a digital register, however flawed, when you move from a presumptive title to a conclusive title to a Torrens system, which are slightly different things. Can you walk me through that process? Now that we know that digitization itself is complicated, how do you go from digitization to Torrens, or should we even bother?

BURMAN: I think the design of the program is that once you’ve digitized and computerized all land records, then the computerized or digitized system becomes your system of land records in steady state, so any future changes happen online or on the digital system and that any future updates to transactions happen online. Because it is all happening online and you are also updating the cadastral, the map itself, you are creating much more foolproof and authentic records. That’s the design of the scheme. I don’t agree with it.

In fact, as in many states, exactly what you described is happening. They’re basically taking existing land records without updating them and digitizing them. Some states have taken the initiative to make sure that they actually survey that relevant part of the state before they update it. I think Gujarat was one of the few states that did that. They actually resurveyed the state and then updated the land record. Many states are not doing that so there is a problem that you are basically taking bad land records or old land records at best and putting them into the system. It’s going to be a problem when you make that system a conclusive system and you see glaring deficiencies of that.

RAJAGOPALAN: It’ll lead to more land disputes in courts. That’s the part two of that problem.

BURMAN: Also, any system design deficiencies that you create today are going to show up in litigation tomorrow. For example—and I’ve seen this—if you have a system where unpartitioned land is owned by three brothers—under Hindu succession law, you each have a notional one-third share of the property but it does not belong in a one-third share to each brother. It belongs to the family. But when you input that data into the land record system, because of the way this was designed in the state that I was looking at, you had to put in the names of the three brothers and say that they owned one-third of the property.

Now, what happens if one of the brothers dies early and the other two decide to share it 50-50? If you come to the Land Revenue Department and say, “I want to make a mutation or make a change,” they’re not going to be able to do it because you originally, as per for the system, owned one-third of the property and not half of the property. If we actually continue doing what we are doing now in some states, we’re going to end up with these kinds of problems. And if you do conclusive titling with a system like this, it’s going to get even worse because in a conclusive titling system, the state actually has to indemnify you if there are any errors in the land record system.

Very briefly, a conclusive titling system is actually supposed to reflect in the record a complete and accurate and updated version of what is actually existing on the ground. Theoretically, you should not even have to go and do a site visit to be able to figure out whether you want to buy that land or not because everything is there in the conclusive titling system. The flip side is that if you get things wrong, the state has to keep indemnifying and the question is, will Indian states indemnify people if these errors keep popping up? That is the big if about the conclusive titling system and if you were to say, “We won’t indemnify,” but it is still a conclusive titling systemthen there’s no point.

RAJAGOPALAN: It helps with the Evidence Act in court. A lot of the current disputes will get moved along faster and get thrown out and things like that. You don’t have to relitigate the title. You can have conclusive titling without indemnifying which gets you maybe 30%, 40% of the way to clear out bad cases and get things thrown out and things like that. But you’re right, it doesn’t get you all of the way, especially for big projects where there’s enormous amount of financial risk.

BURMAN: Exactly, and there I think products like title insurance are way better where you try and basically buy insurance for any future defects in title.

RAJAGOPALAN: Yes.

BURMAN: I think we need that kind of a system to grow in the Indian market as well.

RAJAGOPALAN: I have a question here. Now, let’s say this digitization actually works. Let’s assume that that’s a good idea and let’s say based on that good idea, we switch to a conclusive titling system. Can there be a limited time period when the state says, “We will indemnify everything,” and then there is some kind of statute of limitation that kicks in at, say, seven years or 10 years, then we say, “This is it”? For these seven years, put up disputes, we’ll figure out who’s the correct owner. There will be an emergent bottom-up process by which the register starts getting updated because it’s a dispute resolution. And there, gram panchayats will come in and there will be a period of slight chaos and uncertainty, but after seven years, forever hold your peace. That’s the way we do it.

Then you know that on that day, fair or unfair—I’m sure there’ll be some poor widow who gets left out of something, which is terrible—but for the larger system, there is certainty beyond that. Right now, there’s a secondary question which you already raised on whether India has the state capacity to keep the register maintained because, in the future, it has to still indemnify everyone. 

That’s question one. Is that a good way of thinking about solving the uncertainty and the financial risk problem? If not, then should we have something like a spin-off of the LIC, which says that for a period of 20 years, we will provide title insurance? It is helping develop an insurance market and we’ve already liberalized insurance markets to some extent.

LIC takes the lead because the state needs to absorb some of these losses, but it won’t be as chaotic as indemnifying in cases where the state has no capacity and poor land records. These problems always go in clusters, right? The same state that has no capacity has poor records and will need to indemnify, and all of those things. In that 20-year period, while an LIC-esque vehicle starts doing land titling, you will have private people who come in, there will be some market of lemons situation where the private players don’t get into the terrible states, but over a period of time, the problem gets solved.

Is that another way of looking at it? Are both of these terrible ideas, we just need some third way? What is a good way to solve the transition problem? Because that’s really the problem.

BURMAN: Yes. The way I’ve thought of it, and this is not a fully formed idea, is that it makes a lot of sense to follow land prices and transactions. Instead of doing a statewide land digitization or conclusive titling program, focus on areas that are seeing a lot of transactions, focus on areas where land values and property prices are highest, and focus on cleaning those areas up because clearly, there is a demand for looking at accurate land titles in those areas. You could not serve society better than by going and solving the problem in those areas, and then you can move on to your stage two and stage three. I think doing conclusive titling in these smaller units also makes sense, and then you can gradually expand your footprint. 

For example, if you are doing conclusive titling or land digitization in a village 200 kilometers away from the state capital where you see probably 10 transactions in a year. According that the same priority as the city periphery where you’re seeing 10 transactions in a day does not make sense right now, especially when we take into account state capabilities, the question of whether the state will pay indemnity and so on and so forth.

RAJAGOPALAN: Absolutely, and the financial risk.

BURMAN: And the financial [risk], exactly. It’s very different, and I think it’ll also be much more valuable for people in a positive externality sense. If you’re actually looking at economic growth and development and you want to promote that, doing this program on a priority basis in areas where land markets are vibrant and dynamic makes a lot more sense. That’s how I think of it. 

I think the public insurance option is an interesting option, but the answer to that is that I think we already have the regulatory enablement for private title insurance in the market. I think the problem is with the cost structure, and there’s something about how we’ve done it which is not really affordable for the retail buyers, but maybe for the large developer, it’s much more affordable. I think the question is how do you solve that problem? I think then you don’t have to think of a public insurance, LIC kind of thing. You can just let these providers into these markets.

Land Leasing

RAJAGOPALAN: They will also likely go to the more sophisticated markets first, where there is a big demand from, say, big banks who are bankrolling the mortgage and they want to make sure the collateral is solid. There are other supporting institutions that can come in that cannot just derisk the problem, but also provide the information and the clarity required for this kind of market to actually start getting better or the institutional environment in this kind of market to start getting better.

The second question I have has to do with leasing. You’ve written about leasing. You’ve written it in the context of solar power and things like that. To me, land leasing option seems like another very interesting in-between way between eminent domain versus do nothing. If you’re thinking about the big question of land consolidation, land leasing, if the government is leasing land, let’s say for a period of 25 years, then you know that it’s a pretty solid title. Because when the lease is done, the government will give it back to an individual or a family, or a firm, and you know that that title is pretty good. It’s not as good as an eminent domain, IDCO-title where all past things have been extinguished, but it’s about as good as it gets. 

Is this another option going forward for large pooling of land where we don’t completely subvert property rights? The government comes up with some arrangement to consolidate land through a lease, and these leases are fairly long. You can think of a factory setting up there or a commercial enterprise setting up there, and then after 30 years or 50 years or however long the lease is, people can figure it out.

Meanwhile, because the government is an intermediary and there are other kinds of democratic pressures and so on, you know that the little guy that you know, the one we’re always trying to protect, ‘the farmer with under one hectare of land’ kind of situation, you know that they are not going to get completely ripped off in the transition, nor will they lose all the rents from the development. Because at the end of a 30-year or a 50-year lease, that land is going to be worth a lot—

BURMAN: Exactly.

RAJAGOPALAN: —in case of a sale. Is that another way of looking at it? I know you looked at it from the point of view of how do we set up large parks for renewable energy, but it also, to me, solves this other problem which you’ve been trying to solve in your other writing.

BURMAN: Yes. Absolutely. I was also informed a lot by looking at what China did in its transition years. A lot of what they did was enabled by land leasing. The Chinese legal framework does not create this elaborate Western notion of property rights, but it still manages to create property rights without calling them property rights. The way they did it was to basically do long-term leases. From a first principles’ basis, it makes a lot of sense. For all the reasons you pointed out, the stakes are not as high as outright buying and selling. You do not do a one-time negotiation. You can renegotiate after 20 years, 30 years, 50 years, 100 years, whatever it is.

Second, if your future generations are not going to do any good, then you’re a landholder and you’re not sure about their prospects, that land can either revert back to them or they can continue to earn rental income from them. For people who are at a stage where you are beginning to move out of poverty, but you’re not sure you’re going to really make it and you’re still on that border and one income shock could push you back, I think it makes a lot of sense to do leasing than to be dispossessed, so to speak, of your land through sales. I think the stakes are a little lower when you’re talking about leasing as opposed to buying and selling or definitely compared to acquisition. The problem is that in India, a lot of states do not allow land leasing, especially agricultural land.

RAJAGOPALAN: Walk me through that. Why is that the case? Why are they not allowing leasing even to the government, which seems so crazy to me?

BURMAN: Again, I think historical reasons mostly to do with creating tenure security for small farmers and tenants and to basically make sure that the powerful zamindars in the village cannot repossess the land that has been redistributed to the form of tenancy and leasing and so on. I think because that economic structure has shifted in the last 60, 70 years, it’s high time that we revisit those regulatory regimes. No one will say that there is no land leasing taking place. Obviously, people are doing all kinds of informal transactions in leasing land and dis-own land, but by pushing it underground, we have just made it suboptimal. 

RAJAGOPALAN: Increase the chances of squatting and trespass and all those other kinds of downstream problems.

BURMAN: Exactly. We need to liberalize that regime. I think in the last government, NITI Aayog had pushed out a model land leasing law, which was a pretty reasonable framework to at least start with. I think a lot of states should adopt it ideally.

RAJAGOPALAN: Did anyone adopt it? Are some states better than others when it comes to land lease?

BURMAN: I’m not sure. I’ll have to check. I’m definitely sure that some states do allow leasing. Again, I think there are some reformer states, I think Karnataka, Maharashtra, but I’ll have to check.

Land Pooling

RAJAGOPALAN: Yes. There’s one other really nice idea that I learned about from your work, and this is land pooling. Again, the big problem is fragmentation. We have this broken land record system. It’s not easy to change use or to have land transfers. So we still need structural transformation, we need these big projects. How do we go about it? You’ve written this lovely essay on what happened in Haryana. I want to walk through that experiment with you and just understand that a little bit better.

From what I understand, there was one experiment where the Haryana State Infrastructure Corporation, they were already in the process of using eminent domain to pool a large piece of fragmented land. Then partway through that process, they said, “Look, this is like doing deals by holding a gun to someone’s head.” They said, “We will go through with the eminent domain use or the land acquisition if you don’t agree to pool your land.” But then there’s this other process by which you could pool your land and then you could actually have a stake in the development in this particular situation.

There are these versions of it, which is you put in one acre, you will still own 25% of it, and for the remaining 75%, the state is going to build power lines, sewage, roads and all of those sorts of things. That’s a straightforward, simple way. And what ends up happening there is you have the development of that land, which is desperately required. There is a derisking of that development, but if that development truly takes off, then all the people who pool their land will be the most direct beneficiaries. In some sense, it solves multiple problems. One way in which this was done was by holding gun to the head. Are there other ways of doing this without holding the gun to the head? Are people willing to pool land? Is this a workable experiment?

BURMAN: Yes. This is some ongoing research that I’m doing. Basically, different states have tried different models with varying different degrees of success. You mentioned the Haryana model, which over a period of time, I think there is a clearly visible strand of experimentation and adaptation that is visible in the Haryana model. I think the earlier models were, to put it politely, more crude, and I think the more recent models are far more beneficent and generous to the landowners. I think it also shows that the state is confident, and the landowners are confident of the proposition that is being offered.

That’s one model where basically the state coopts the landowner into the process of development, and says, “We will give you a portion of the land,” and in some cases, even gives you incentives in addition to the redeveloped piece of land. For the landowner, the value proposition is that the redeveloped plot of land will have better amenities, infrastructure and so on, and the market value of the property will be much higher. The similar value proposition is used in almost all schemes across the country, so Gujarat and Maharashtra has something called the Town Planning Scheme, which Gujarat especially, has used very successfully first in Ahmedabad and now in other cities in the state. 

What distinguishes the Town Planning Scheme or the TPS is the degree of state intermediation or intervention in the process of pooling or redevelopment. Usually, they’ll have a very elaborate process of planning that area, of doing public consultations, of explaining to landowners what the reconstituted plot will look like, where you’ll get your property, whether it’s the same in terms of market value as the earlier plot and so on. A lot of that is done by your state officials, your government officials. In other models, less is done by the state. Delhi has a land pooling scheme which largely relies on private land assembly to do the same job. It relies on developers to basically go and talk to landowners to convince them to reconstitute and put in the land under the land pooling scheme.

There are these different models, and I think where we are right now is at a stage of experimentation nationally to see what models will work better. I think one critical variable is state capability and the degree of adaptiveness of these institutions. Gujarat has one model, can other states work that model well? I’m not so sure. Haryana has one model, will that work equally well in say, Karnataka, Odisha? I don’t know. I think what is good is that because land acquisition has become so expensive and also was normatively unfair and did not work out well, people are trying out these models which are much more consensual, much more participatory. It will be interesting to see how it plays out.

RAJAGOPALAN: Yes. There are a couple of really nice things you yourself point out in your essay. One thing I found is you’re absolutely right, it’s not clear the Haryana model can work somewhere else because Haryana landowners have also seen this process for the last 30, 40 years. They’ve seen their uncles and their neighbors and their friends and their family lose land to developers, they’ve seen them protest. They’ve also become wiser and understood that you don’t want to block all development and at the same time, you don’t want to get ripped off.

It’s not clear to me that let’s say, in the outskirts of Arunachal Pradesh, you have a similarly sophisticated seller because that kind of urbanization and development has simply not happened in that area so far. There is some learning in the political economy that’s already taken place or for instance, you talked about the NAINA project, this is the Navi Mumbai Infrastructure Airport Authority, something like this long acronym, right? One of the interesting things going on there is the more people who sign on, the more they will adjust the FSI. That’s not obvious in most places.

In Mumbai, the average seller is highly attuned to FSI, the developer is attuned to FSI, they understand what that means. Again, it’s unclear to me that 200 kilometers outside of a big city in Karnataka, you’re going to have a similar understanding of, oh, we can pool land by increasing the amount of FSI. Maybe it happens in Amravati, but it may not happen in other places. Who should be doing these pooling schemes? I guess that’s the big question that emerges from your writing on this. Should it be something as specific as the airport authority, as in the case of Navi Mumbai?

Literally, for each infrastructure project, we float a new vehicle, and that does it. Should it be a more generalized uniform law that NITI Aayog suggests? Something like the Haryana experiment number 2, which says, if you pool an X amount of land, and you’re more than X percentage of the overall land, you get a board seat in the new development. Should it be the government that is doing the development? Should it be the private case like Delhi? Who should decide this? What level of government should decide this? What level should we set the regulation?

BURMAN: I think there are two or three ways to cut this. One is to save, what is the role of the state and what are the principal functions involved here, and who should be performing which function? Clearly, there is a planning function involved, which is somebody has to say, “What will this area look like after we’ve done the pooling and put in the amenities and the infrastructure? Where will the shopping mall be, where will people live, where will be the hospital?” and so on. Somebody has to do a little bit of that, even if it’s a private endeavor.

Then there is the assembly function, someone has to go and convince people to do it. They have to convince them of the value proposition, give them some credibility that you will actually get your plot back, or the market value of your plot back, which is a very important part of the entire process. Because you can’t afford holdouts, you want to get as much consensus as you can, and so on. Then someone has to make sure that the infrastructure is developed, at least the trunk infrastructure, if there is a major road or a major sewage pipeline that has to come in. Finally, there is the oversight and dispute resolution thing.

RAJAGOPALAN: That whatever was promised was actually done.

BURMAN: Was actually done. You can cut the other three, four functions, depending on what your preference is, but I think that oversight and grievance redressal, contract enforcement function has to be done by the state. I think that is a constant no matter what scheme you adopt. Then it really depends on the degree of state capability for the other four functions. Can you do Ahmedabad Urban Development Authority level of town planning and land assembly in Delhi or Gurgaon? If you can’t, if the DDA can’t, then maybe it’s better for a DLF or ANSAL to.

Therefore, the schemes will look very different depending on what capabilities the respective state institutions have. Also, their degree of credibility, if landowners in Delhi don’t trust the DDA for some reason, it’s going to be very difficult to run the Gujarat kind of TPS in Delhi. I think those are the critical variables.

Other than that, I think there is a lot of scope for adaptation based on where you are placed and how you view yourself and what the local market looks like. I don’t think there should be a massive central push to say, “Take this model, do that model.” I think it’s more important to think of what are the functional roles and responsibilities and how do city planners, urban development bodies think of who can do this best.

RAJAGOPALAN: Here, I feel like there are two additional issues and these are the big political economy issues, like if you solve this for land, you can solve this for so many other things. One is sometimes you have an overcapable state. This is like the regime uncertainty that is introduced because, in a city like Mumbai, the politician or the corporator is totally in bed with the developers, but what happens if the government changes? What happens if everything gets called into question? The supreme court or the high court realized this was totally corrupt, it was a scam. They invalidate all the contracts. People who actually donated their land to land pooling, it gets stuck in litigation for 20, 30 years. It all comes to naught.

One is those kinds of political economy problems for which there’s no solution, but they get in the way. These political economy problems are also the sorts that present themselves in the most valuable areas because that’s where the cronies are attracted, where the honeypot is. That’s one part. 

The other part is, is there a model in states like say Bihar or even like, say, a small state, Himachal or something like that, where we don’t have very high level of capability for this kind of structural transformation or at least it hasn’t been demonstrated yet, but there could be some vehicle that comes in, either a state-owned corporation or a union government or a private. How does that get stitched together?

Sometimes it gets stitched together in places like Bombay and Ahmedabad a little too easily and then that also is a problem. Then sometimes it can’t get stitched together at all because it has potential, but the land is not valuable yet to attract the crazy characters involved.

BURMAN: Yes. I think in some cases it’s what you’re describing, which is there is overcapacity and land is not the main problem or planning is not the main problem. You can do brilliant town planning schemes in, I don’t know, Itanagar or something, but there are other factors there that just don’t make land that valuable. You can have a beautifully planned city. It’s just that the city won’t be very dynamic and it’s not a fault of planning or whatever it is. The labor market just isn’t as dynamic as— 

RAJAGOPALAN: Exactly.

BURMAN: —Mumbai or Pune or Delhi. I think smaller cities and towns face that problem, which is the main problem is labor markets. The problem that cities that are growing fast is that you don’t see that inflection point. You don’t see where a city goes from a small town to a big city, or when a city goes from a moderately sized city to a metropolis. If you look at Pune, or Jaipur or Udaipur, I don’t know, a lot of these cities that have grown very rapidly in the last 10, 20 years. The problem is that at one point of time, the planning capabilities were fine for that kind of labor market. Now, it’s just a completely different value proposition.

You have international real estate consultants who set up offices there who are advising conglomerates on how to run or how to do real estate acquisitions, and the planning capacity has just not kept pace. I think the problem we are seeing there is basically the degree of adaptiveness of state institutions that we need is just not there. I don’t know how to solve that problem. I think people who are working within state institutions are more cognizant of this, they’re trying to do it. There are all kinds of good organizations who work at the grassroots to try to help that capacity building.

I think one of the problems is also the fact that if we overdesign and overplan so much that we don’t have basic amenities. For example, if we say that if you have a nine-meter road, and we are going to allow two-story houses on this, and tomorrow, we are now going to allow eight-story houses on this, and we’re going to make this nine-meter road into 18-meter road. We don’t have the basic capacity to say, I’m actually going to take that amount of margin from those plots to make it an 18-meter road. What ends up happening is that you have an eight-story house on a nine-meter road.

RAJAGOPALAN: Then every congestion and nuisance problem that that brings with it. Yes.

BURMAN: This is not a small subtle problem. It’s a pretty big visible problem. It’s not a question of whether your setback was one meter or 1.5 meters. You’re basically talking about doubling the size of the road, and a lot of municipalities just don’t have the capabilities to do that kind of job. I think that’s the big problem.

RAJAGOPALAN: Yes. I think Ahmedabad is an exception, maybe BMC is an exception, but everyone else is really struggling with this. The last question, Anirudh, is there anything in this system that makes you hopeful or any experiment that’s exciting that you think is doing something great? What excites you about this space and is not depressing?

BURMAN: I think one is that the way planners are looking at cities has changed or is changing quite significantly, even from the time when I started thinking planners are the problem or planning is the problem. I think the more I talk to them, the more I realize a lot of them agree with the fact that we have problems in how we plan cities. I think also a lot of city-level administration, the more administrators you talk to, the more excited you are by just how impressive the things they are trying is.

I think there is a peculiar problem that when you go to very large cities or cities that are already built out and developed it’s harder to make things there. If you go into a new part of the developing part of the city or a younger city, there you can actually see administrators and planners who are experimenting a lot, being innovative, trying to do new things, because I think it’s also a function of the kinds of interest groups and activism that you see in a city. It’s more difficult to make changes in a city where the residents are going to complain about every small thing.

I think that is going to be the bigger problem in India going forward, is if your big cities are your growth engines, as they are in India, how do you redevelop them? It’s not a question of just development. It is you have to redevelop the core of the city to make it look like, I don’t know, Manhattan or Hong Kong or whatever it is, to make land use more efficient, while at the same time, expanding into the peripheries and taking over agricultural land. I think that expansion is still easier, but overcoming all the problems in the center of the city where you have to redevelop, where land prices are already very high, where the interest groups are very vocal and sharp, entrenched—yes, that’s the right word—I think that’s going to be a very key problem going forward.

RAJAGOPALAN: Hopefully, we’ll have you to tell us how we can solve these things. Thank you so much for doing this. I’ve learned a lot from your writing on this. We’ll throw in all the links, and this was such a pleasure.

BURMAN: Thank you. Thank you for having me, Shruti. I really enjoyed the discussion. Thanks.