Shruti Rajagopalan and Milan Vaishnav on India’s Delimitation Dilemma

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 we are releasing a webinar recording from April 24th where Milan Vaishnav and I had a conversation on delimitation in India. Milan is a senior fellow and the director of the South Asia Program at the Carnegie Endowment for International Peace. He is also the host of the excellent Grand Tamasha podcast where this recording will be simultaneously released. 

We talked about the failure to pass the 131^st constitutional amendment bill in Parliament, the government’s intention to reapportion and increase the size of the Lok Sabha, the delimitation freeze pegged to the 1971 census that has survived five decades, why the fiscal bargain between states matters as much as the political bargain, 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, Milan. It's great to see you.

MILAN VAISHNAV: Hi, Shruti. Thanks for having me.

RAJAGOPALAN: Yes, this is, I think, going to be a lot of fun. I want to first welcome all the people who are joining as we speak. I believe we have a few hundred people who registered for this conversation, which makes sense because I think delimitation is one of the great fault lines of our times. In India, it's the political bargain that will set the tone for all other political bargains to come in the future, including constitutional, federal, and fiscal.

I can't think of anyone better to have this conversation with other than you, Milan, because when I first started writing about this five years ago, and I was looking for who has written about this, there was only one paper by you and Jamie [Hintson], and everything else was a little bit outdated. I think of you very much as my go-to person when I want to think through the politics, the numbers, and everything that's happening with delimitation. It's wonderful to have you here and thank you so much for doing this.

VAISHNAV: Thank you, Shruti. I have to be straightforward and say that when you asked me, I said yes on one condition, which was that this was going to be a conversation, because I've relied on your work on delimitation, but I also think we bring very different perspectives. As a constitutional scholar and economist, you have a perspective that I simply cannot have as a political scientist. Actually, hopefully, bringing these together will be useful for us, but also for the people who are paying attention.

What is Delimitation?

RAJAGOPALAN: Yes. Maybe a good place to start is: What is delimitation in general? How does it take place in India, and what are the controversies and fault lines for those who are not very familiar with this topic and steeped into writing it for the last few years like you and me?

VAISHNAV: Quite simply, delimitation is what most of us would call ‘redistricting’ in other democracies. We're having these debates in the United States on a daily basis, literally. It involves two linked processes. One is about apportionment, and the second is about delimitation in the narrow sense of constituency or district boundaries. The first apportionment is really how we allocate seats in the parliament and the legislature across states. The second is within states. If you have a collection of parliamentary constituencies or state assembly constituencies, what do those boundaries look like, given the overall pool of seats that you might have in the state?

In India, there is a fairly clear constitutional basis for how this is done. You have Article 81 of the Constitution, which says the allocation of Lok Sabha seats, the lower house of parliament, should be allocated across states in a way that is proportional to the population. The ratio of every member of Parliament to the population they represent should be as practicable as possible, roughly equivalent across states.

Now, there is an exception here, which is for very small states. Those that have a population of 6 million or below in union territories, those don't have to be proportional. Parliament can decide how they want to apportion those, but broadly in the main, that's what we're talking about. Now, Article 82, which follows, says there should be a readjustment after every decennial census. In India, it's done on the ones, 1991 and 2001. Not 2011, for reasons we'll come back to later. After that census enumeration takes place, parliament can authorize the setting up of a delimitation commission, which will essentially do this work, again, both in terms of thinking about seats across states as well as how to do boundaries within states.

There's a parallel requirement, which I won't get into in the constitution, as you might imagine, about how this should work for state assemblies. It's essentially the same process. There's another part of this which I think is important, which is that the size of the Lok Sabha is constitutionally capped. The cap on the number of seats the lower house can have has evolved over time. In the latest amendment, it is 550 seats, maximum authorized strength. There are 543 seats currently being occupied.

The ceiling, again, has fluctuated by amendments over the years to Article 81. Any expansion of the size would require a constitutional amendment. There is, as I mentioned before, an independent delimitation commission authorized by parliament that is meant to essentially do the nuts-and-bolts work of this apportionment of this redistricting. Its orders have the force of law. They are not questioned by the judiciary. It is set up for this purpose by an express act of parliament.

I believe the last delimitation had an act that was passed, I think, in 2002, or roughly that timeframe, to set up the delimitation that happened, at least dealing with intrastate boundaries in 2007-2008. This is now getting into the weeds, and we'll have time to get into some of this stuff. The last thing I just want to say very quickly is, because I think it's very relevant to the current debate, which is this question of reservation.

Under the Constitution, one of the things that also gets decided by the commission is which constituencies will be reserved for two types of groups, for Scheduled Castes and Scheduled Tribes—two historically disadvantaged marginalized communities. Together, they have a quota of roughly a quarter of parliament and state assemblies. Those are meant to be given to areas which have Scheduled Castes and Scheduled Tribe populations in the greatest numbers. There's a little bit of nuance there, but broadly, that's true.

The new wrinkle, of course, is in 2023, this government managed to get, through parliament, a constitutional amendment, the 106th Amendment, which would reserve one-third of seats in the Lok Sabha and the various Vidhan Sabha state assemblies for women. In that amendment, as it was passed, essentially tied the question of the commencement of the quota with the next delimitation, the delimitation we still have not had, because that delimitation was supposed to be taking place after a census, which we also have not had.

That, I think, essentially is the basic nuts and bolts. Shruti, maybe I'm going to turn the tables on you for a second, because one of the things I haven't gotten into but I think now is the right time to do it, is this process has not worked as cleanly or as neatly as I have just laid out. The constitution is very clear. Every 10 years, you have a census, deal with the delimitation, and figure out this reapportionment and redistricting. You have written this excellent piece—which we should have for our listeners in the show notes—recently in Indian Express, where you trace back this history because there's a constitutional angle and there's a political angle. I don't know if you want to maybe elaborate on that for a second.

RAJAGOPALAN: Sure. That was an excellent primer. Thank you, Milan, for getting us to this point. The question you're asking is how exactly did we end up here in 2026? One of the quirks of the Indian Republic is typically bicameral legislatures have one of the houses which is apportioned by population and the other house which is apportioned based on some other principle. It may not always be population.

Now, in India, both the Lok Sabha and the Rajya Sabha were always apportioned on population. Of course, one has direct representatives and the other has indirect representatives. The Rajya Sabha was never the US-style Senate or some of the other federal republics, which have a very particular number of people from particular ethnicities, minorities, and so on. That's the beginning part of the problem.

Now, what that means in the Indian Republic is if you lose population share, even if the absolute numbers are growing as a share of the total population of India, if any state loses share, it loses some political voice in Parliament proportionate to that, which makes sense because that's how democracies work. Of course, in India, it loses it in both houses. Now, a second quirk of the Indian Republic is right at the beginning, India started as a very centralized union of states, not the typical fiscal federalism one would expect, and it only got worse as central planning kept increasing.

India's purse strings have always been very tightly held by the union cabinet, and all the allocations have been decided by the union cabinet, which is, of course, based on Parliament. We reached a situation in the 1970s where the richer and more developed and the southern and western states were actually losing the proportional share in parliament based on every decennial census. That meant that the fiscal pie and all other major decisions of plan expenditure, welfare entitlements, everything was now going to be decided in parliament, and they had a smaller voice. It's useful to note that in the 1970s, most states were actually Congress governments, either because they won the election or they got rid of opposition to the president's rule. This was all crazy times during the emergency. The 42^nd [Constitutional] Amendment, which was really infamous in 1976, which did a number of things.

VAISHNAV: Lots of other things other than this.

RAJAGOPALAN: Exactly. Lots of things. The one defining feature of the 42nd Amendment was it was centralizing India more as a republic and placing more power, not just in the hands of the Union Parliament, but actually in the hands of the Prime Minister's office and the Union Cabinet. That was really the structure of it. Where did that power get taken away from? It got taken away from the judiciary and the states, the Lok Sabha, Rajya Sabha, and so on.

Now, the 44th Amendment undid most of the damage, but one of the things it didn't undo was this delimitation freeze. Now, what does that mean? In 1976, the amendment decided that we're going to have population proportional representation in Lok Sabha and Rajya Sabha, except instead of the latest decennial census, we are going to freeze that to the 1971 census. It was for a time-bound period of 25 years because the expectation was that poorer states that have higher fertility rates will catch up in both economic growth and the fertility rates would converge, and 25 years is all one needs for it. They kicked the can down the road and said, “In 2001, we will revisit the question of which census Parliament will be apportioned on.”

Now, come 2001, we have the Vajpayee government. This is already the third Vajpayee government because the first one lasted a couple of weeks. The second one lasted about a year and a half when AIADMK took away support.

VAISHNAV: It was 13 days, 13 months.

RAJAGOPALAN: Exactly. Then the third Vajpayee government, when they took office, was a very strong coalition where the strongest partners were from the south and the west. They included the DMK, Telugu Desam Party of undivided Andhra Pradesh. It includes both Andhra Pradesh and Telangana representation, Shiv Sena in Maharashtra, and so on. Basically, in 2001, it was untenable to revise to 2001 numbers because one of the things that had happened thanks to liberalization was that the southern and western states grew even faster economically and their fertility rates dropped even more.

They would have really lost a larger number of seats or proportional political voice in Parliament. India was still redistributing revenues with formulae from the past. In the '70s, the union government would devolve maybe a quarter of the revenues back to the states. Now, I think it's 41%. In the early 2000s, it was something in the middle of these two. That's how this whole thing happened. Now, a couple of nuanced additions to what you were saying about reservation. Now, there are two parts to the puzzle. There's uneven population growth both between states and within states.

What they decided was that, between states apportionment, we would freeze to 1971 numbers. Within state apportionment would be based on the 2001 census. This is to avoid problems. There was a point in time when Gurgaon was an incredibly malapportioned district because it was growing so fast and there was a huge influx of migrants. Same with Thane more recently. It was the single most malapportioned electoral district in India. To avoid problems like that, that would be fixed on the 2001 numbers. The same for Scheduled Caste and Scheduled Tribe reservations, which are again based on population.

All these things got adjusted. Another thing that happened was that three new states were carved out of the erstwhile Hindi heartland states. You had Uttarakhand getting carved out of Uttar Pradesh. You had Jharkhand out of Bihar, and you had Chhattisgarh out of Madhya Pradesh. Because within-state numbers were based on the 2001 census, the new states were less severely malapportioned than the parent states. Where we are as a result of this today is that we still have a system where they kick the can down the road another 25 years.

The Constitution currently reads that the next delimitation or redistricting exercise will be based on the first census after 2026, which should have been 2031. Of course, as you pointed out, we've not had the last census. It could be any census between now and whenever. That's where we are now. That also means that we are apportioning our parliamentary seats based on a census that is 55 years old, which is frankly just extraordinary.

Populations have grown very unevenly, but also economic growth has not only been uneven—in India, it's diverging. The richer states and the poorer states are both growing economically, but the richer states are growing faster than the poorer states. We don't have the typical catch-up growth or convergence. This is where we are now. I guess I should hand this back to you now.

VAISHNAV: Just to add one thing, Shruti, I think it's important to re-emphasize that the total kitty of 543 didn't change, number one. Number two, the statewise proportion could not change. Uttar Pradesh had its 80, Kerala had its 20, Tamil Nadu had its 39. None of that could change until after 2026. What did change, and it actually came into place in 2007, 2008, is within Uttar Pradesh's 80, you could redraw those boundaries such that every member of Parliament roughly presided over a similarly sized constituency in terms of population.

We're bent out of shape in two ways now, not just the malapportionment across states, but it's also been quite a long time since the 2001 census. The size of constituencies—Gurgaon's a great example—is also now totally out of whack. I just want to add maybe one comparative data point to this discussion, because it's worth thinking about it. There are other federal democracies out there where we can think about how malapportionment in India's lower house compares.

One of the things that we've done, and this is a forthcoming paper that Louise Tillin from King's College and Andy Robaina—who's one of my colleagues—have coming out hopefully in a couple of weeks, is to try and do many things. One of the things is look at this comparatively. In the political science literature, there is this standard measure of malapportionment. It's called the Samuels-Snyder Index after these two political scientists, Samuels and Snyders. Basically, it captures the share of legislative seats that are "out of place."

Basically, you take the absolute difference between each state's share of seats in the legislature and each state's share of the overall population. What that tells you, if you put federal democracies on a spectrum in 2025, is Argentina turns out to be the most malapportioned. Belgium turns out to be the least malapportioned. India is towards the more malapportioned side. The number that comes out is 8% on the Samuels-Snyder Index, which means that roughly 8% of Lok Sabha seats, either over- or underrepresent a particular state or union territory.

Obviously, within that, the two most underrepresented states would be Uttar Pradesh, Bihar. The two most overrepresented states would be Tamil Nadu and Kerala. What that means at the end of the day, in practical terms, is a vote in a southern state carries more weight than one in a northern state because you've really departed from this foundational principle of one person, one vote. India is not the most malapportioned, but it is on the higher end of the spectrum.

RAJAGOPALAN: Yes. What I think that percentage also masks a little bit is where it is malapportioned. It is malapportioned in the lower house, which is typically not where the malapportionment takes place in most federal republics. It's typically in the upper house that you have severe malapportionment. This is the place where you have direct democracy, where you have the most severe malapportionment. The numbers, as you pointed out, Milan, if I remember correctly, I think it's 3.1 million per MP in Bihar and 1.75 million per MP in Kerala. It's almost twice as much.

Of course, these are all made-up/projected numbers because I don't have current census numbers. Everything we discuss after 2011 is a projection. That's an extraordinary discrepancy. All of this is based on fertility rates, and the poorer the group, the later the fertility rates dropped. It is more likely that a poor person is in a severely malapportioned state versus a wealthier person. The same is true for Scheduled Castes and Scheduled Tribes because their fertility rates dropped later than the upper castes.

The same is true for the youth. That's fairly obvious. The later you were born, the younger you are. The groups that saw fertility drops later are more likely to be in severely malapportioned districts. That number, which you gave, is very useful as a comparative, but India may be worse on other margins than just the average percentage of malapportionment.

VAISHNAV: That's right. I think what you said earlier about the Rajya Sabha is important. The upper house doesn't ever get any attention in this conversation because its seats are not subject to delimitation. It's dealt with in a separate schedule of the constitution. I'm sure we will come back to this as we think about future scenarios, envisioning a different Rajya Sabha is very much potentially part of the menu of options.

I think you can apply the same kind of methodology to looking at malapportionment across upper houses, and what you see is that India, among federal democracies, has the least malapportioned upper house, because it is proportional to population. Most upper chambers are not. The US Senate is a great example. Every state has two seats. That’s it. On that spectrum, Argentina, the malapportioned statistic is upward of 40%, and in India, it's 12%. What that means, if you step back, is that the Rajya Sabha reinforces majoritarian outcomes to a greater degree than other comparator upper houses.

You end up with two chambers that largely mirror the population. If we think about the Rajya Sabha as a council of states, which is, in theory, what it was meant to be, this raises some interesting questions, which, again, I think we can come back to. Shruti, maybe now, since we've cleared our throats with all this background, we should say a word, maybe about what the government tried to do. I think that's why we're all here on this call.

We wouldn't be having this conversation on April 24th had the government not called a special session of Parliament abruptly, with very little notice, and presented three bills which have massive consequences. The bills, of course, did not succeed. But, they have set the terms of the debate. Do you want to start and I can fill in a little bit on what actually the government's stated intent was through that legislation?

RAJAGOPALAN: Yes. I would not frame it as the government's stated intent or outcome. The outcome was that the bills failed, the stated intent, I can't know. I can base my remarks on the text of the bill, which I think is the only thing people like you and I have to go on at this point. Basically, in the special session of Parliament, they introduced a Constitution Amendment bill. This is the 131st Constitution Amendment Bill, which lost the vote, and two other bills, which would have been tabled and passed, if the constitutional amendment bill were to pass. Right now, none of these bills are active law. That's the beginning.

What did the 131st Constitution Amendment Bill try to do? Again, constitutional interpretation is a complicated and almost fatal business in India. I'm going to try and be very careful, but maybe I’ve personally misunderstood many things. The first thing it did was to actually increase the number of seats in Lok Sabha from 550, which is the current limit, to 850, of which 835 would go to the states, and 15 would go to the union territories. Now, this is not a terrible idea in itself, because in addition to being malapportioned, India also has some of the worst population-to-elected representative ratios. The original Constitution capped it at 1 MP for 750,000 people. We have gone very far from that. It's three to four times that number in most constituencies. This would help stabilize that a little bit, so MPs can actually serve their citizens better. That's one. 

The more complicated exercise was not just increasing the number, but also what would be the basis on which these 850 seats would be shared between the states, which I guess is the big question, given the delimitation freeze on 1971 census numbers. The original phrasing of the Constitution was that the delimitation exercise would be conducted after each decennial census. Then the numbers would be adjusted to have roughly equal constituency sizes as far as practicable.

Now, this removes that link between the most recent decennial census and delimitation. That's a really important change that I want to flag. Now, what does it replace it with? Now, that is a little bit less clear. It replaces it with a delimitation commission, which was one of the other bills had it passed. This commission would be a new constitutional commission. The Bill had a number of requirements of who would serve on that commission. It's typically a former Chief Justice of India or Supreme Court Justice, a former Election Commissioner, the Lok Sabha Speaker would appoint a few MPs, and so on.

There's the delimitation commission, which would then determine which census would be used to apportion the seats. If you now switch gears from the constitutional amendment and the delimitation commission bill, the delimitation commission bill simply says the most recent census, which happens to be the 2011 census. It's important to note that in India, bills can be amended with a simple majority, whereas constitutional amendments require a special majority, which this incidentally failed to get.

By pushing the decision of which census numbers to use from the constitutional mandate to the delimitation commission, which is based on a statute of Parliament and can be changed with a simple majority, we've really switched who makes this decision from a constitutional mandate to more of a political decision. It also reopened the question of intra-state constituency boundaries, which would again be based on whatever the delimitation commission decided.

This was all true, not just for Articles 81 and 82, which are for the parliament, but also Article 170, which is for the state legislatures. The other major aspect was to link the women's reservation to the current exercise of delimitation. The original women's reservation constitutional amendment, which was the 106th constitutional amendment, amended the constitution to add Article 334A. 

VAISHNAV: Wow, I'm impressed that you could say that.

RAJAGOPALAN: That was supposed to be after an exercise of delimitation was undertaken for this purpose. Now, it's no longer going to be based on the most recent census. It'll be on whatever the political compromise is. Now, the third bill in this mix is one on the union territories because we've so far discussed the states. Union territories usually tend to be quite tiny, but there are two big ones. One is Jammu and Kashmir, which is now a union territory, after all the politics of Article 370. There is Delhi, and to a lesser extent, there is Puducherry.

Delhi and Jammu & Kashmir really have the lion's share of the parliamentary seats. Everything else is just one or two seats. That's the other thing that needs to be reapportioned. Those 15 seats need to be reapportioned. This is what the new bills tried to do. My TL;DR on this is that they really shift it from the constitutional and parliamentary levels. You know the way we have delegated subordinate legislation-making, now we have delegated apportioning in some sense. It's delegated to the delimitation commission, which is not the most robust organization if it's based on what will be introduced, and it can decide what census it's based on. It's not even going to be the most current numbers.

A second thing that is hidden in there, and you need to see quite carefully, is that it actually limits whether delimitation orders can be challenged in court. The language very specifically says in the delimitation bill that the new delimitation cannot or shall not be called in question in any court. That's the other part of the constitutional drama. The last part of the constitutional drama is, of course, how we apportion the reserved categories, which, as you pointed out, is both Scheduled Castes and Scheduled Tribes, but also women.

The Scheduled Caste and Scheduled Tribes part is pretty straightforward. It's based on the original Constitution as it has been going, based on whichever census the Delimitation Commission would have chosen had this bill passed. Now you have to also accommodate the women's reservation.

VAISHNAV: Let me just see, Shruti, I'll give you my TL;DR and see how well it matches up. Just because there are a lot of moving parts here. Take the constitutional amendment bill. It does three things mainly. One is it raises the constitutional ceiling of the Lok Sabha to 850 seats, 815 for states, 35 for union territories. It says that the number of seats for each state in Parliament is going to be based on its proportion of the population of states.

Number three, and I think this is important, and this gets to the point of who is the arbiter. It removes the automatic post-census trigger for reapportionment. It now gives Parliament the discretion to decide when delimitation will happen and which census to use, which is something that had been hard-coded. When it comes to the delimitation bill, obviously, in the main, it's providing the operational framework for delimitation, but there is this interesting ambiguity in there, which it says that the seat allocation, how much UP will get, how much Kerala will get, as well as the redrawing of seats within states is going to be based on "the latest census figures published"—question mark—because we don't really know what that could mean.

Then there's the UT bill, which I think is doing one thing: essentially ensuring that the framework for reapportionment that would happen at the state level also happens to these three important UTs, which also have state legislatures. Then, of course, there's the commencement or the kicking-in of the one-third quota for women. I think, just to take a step back, there was a deliberate decision made in 2023 when women's reservation was introduced to tie it to delimitation and, by extension, to the census.

Maybe we should just pause and talk about this for a second, Shruti, because I'm not a constitutional scholar, but it strikes me that if you go back in time to 2023, there was no real need for that to happen. You could have come up with an alternative legislative framework which said, “We are going to take 543 seats, we're going to reserve a third of them in time for the May 2024 general election.” I think everyone's reading of that was like, nobody wants to do that because it means people, mostly men, who have those seats are going to be out of a job come next election. Is that a fair assessment, do you think, of what went down in 2023?

RAJAGOPALAN: I think they were setting this up in 2023 as part of some long game. Now, let's think through the arithmetic of it. Let's say that you're right, and it's really self-interested men who don't want to give up power. It's easier to pass a legislation introducing reservation than a constitutional amendment. The very fact that they managed to whip the votes for a constitutional amendment tells me that there was some other backroom bargain given to the very same men that “Hey, if you agree to this constitutional amendment, and we link it to delimitation exercise in some way, we will do this later in a way that benefits the same groups.”

I'm exactly with you. There was no need for women's reservation in parliament or anywhere else to be hard-coded in the constitution unless they thought future governments would undo it. That's literally the only reason to have done that. It was much harder to pass the constitutional amendment. It's a little bit odd. Now, in hindsight, it seems like that's exactly what they were setting up, which is some way to bring in women's reservation and then say “Oh, the women in Uttar Pradesh are now going to be fewer than women in Tamil Nadu or in Kerala,” and make this a hot button issue as opposed to what the other hot button issues are. I have a feeling it was something to do with narrative building. I still can't figure it out.

The Politics Behind the Constitutional Amendment

VAISHNAV: I think that brings us to the question that probably everybody has joined this conversation to ponder with us, which is the politics behind this special session and these bills. Why now, and what is going on? I just want to say at the outset that I think it's fair to say you and I are both a little bit puzzled because there are a lot of theories, a lot of hypotheses. I think maybe to break it down to first principles.

Clearly, if you are a party and you are a government in search of a narrative because many other things perhaps are not going as you envisioned, the economy is not firing on all cylinders. Some of that may be because of your own doing, some of that may be because of geopolitical factors like the Iran war, Trump's tariffs, and what have you. The emotive issues that have powered the BJP and the Sangh Parivar like Article 370, Jammu and Kashmir, Ram Mandir, are taken off the table. You still have the Uniform Civil Code; you have the National Register of Citizens. You have other things, but the outlook for those is not quite as clear.

If you are looking for narrative building, women's reservation is certainly an issue that you could pick on, especially because the government and the ruling party have made it a priority, certainly in terms of their rhetoric—and some would say even in terms of their policies over the past 12 years—to try and cater to women and make that a centerpiece of their broader political strategy. There are lots of people who've actually done some really interesting work on how and why the BJP has decided to do this.

Now, again, let's just assume for narrative-building convenience, this was what they settled on. In their minds, they're setting up, I think, a very nice trap. Because you put out there this very constitutionally vexing question of delimitation, which has huge political ramifications. You attach it to women's reservation, which is, I think, broadly popular and all parties in theory support. Then, when this package fails, you can blame it on the opposition as saying “They never wanted women's reservation to begin with.”

I think one of the things we were discussing before we started, but I think is true of the BJP historically, and maybe other political parties as well, is oftentimes the narrative is really about the journey as much as it is about the destination. In Ram Mandir, there's a great example of that. That was an emotive issue that powered the party from having essentially two seats in parliament in 1984 to what it is now. Interestingly, once they actually got the temple built, they didn't really see much political payoff. It really was in the agitation and narrative.

RAJAGOPALAN: They lost Ayodhya.

VAISHNAV: They lost Ayodhya. This gives them a juicy narrative for the state elections, which are ongoing. Voting has just started in Tamil Nadu and West Bengal, as leading all the way to 2029. There's a second reason why they may have wanted to do this, which has to do with the concept of delimitation as a tool to redraw constituency boundaries within states. There have been relatively few allegations prior to the past five years that delimitation has been used as an expressly political tool to redraw either state assembly or parliamentary constituency boundaries for political reasons.

There have been a couple of political scientists who have looked at this. None of them have come away with any convincing evidence to suggest there was some kind of tinkering going on. However, we have seen two states in recent memory, Assam and Jammu & Kashmir, which never had their boundaries redrawn as part of the overall package in the mid-2000s, for different reasons, given the sensitivities of those states.

In both of those states now, we at least have qualitative, suggestive evidence of gerrymandering of different kinds in both places. If this is part of what you want to try and embed when you think of all 28 states, this is a useful thing to do. Here is where I've reached the limits of my knowledge, which is what the plan was when it comes to statewide representation in parliament. It seems to me that they were saying and doing two entirely different things.

You have the text of the constitutional amendment bill, which says very clearly, we are going to allocate seats according to population. We're going to rip off this band-aid of this freeze that's been set in motion since the 42nd Amendment. We're going to see that massive divergence. The north's going to gain a lot. The south's going to lose a lot. Coupled with an oral guarantee, “No, no, no, we don't want to interfere at all in the statewise allocation of seats.”

The Home Minister said that. The Prime Minister said that. Every party spokesperson said that. Even to the point that in the final hours of the debate, Amit Shah, the Home Minister, said, “If you want me to, I will leave, come back in an hour with an amendment that codifies this.” Whether that was bluffing or that was serious, is for the historians to tell us. This is where I'm confused about what their real intention was.

There isn't one argument which says, “This was just all trolling.” They essentially knew this wasn't going to pass. They knew they didn't have a two-thirds majority. You could throw out this promise of an amendment and paint the opposition as unreasonable, “See, we said we would hardwire not messing with the statewide shares, and they didn't go along with it. Obviously, they must have known that these guarantees were not going to be sufficient in the way that they were communicated to really placate the states that would be on the losing end.”

If you step back, maybe one thing they were simply doing was maintaining negotiating flexibility, painting a narrative that puts the opposition on the back foot, testing political reactions. You got some very useful information. You understood whether the opposition was going to hang together, how strongly the southern states were going to fight back, and whether some of your own allies like Chandrababu Naidu and the TDP, who represent a big southern state, were going to stay with you. They didn’t stay with you.

This sets the terms of the debate. We'll get to this at the end of the conversation, like future scenarios are going to use what they put out as some kind of baseline. Anyway, I would love to hear what you think about any or all of that.

RAJAGOPALAN: I have one question, and I think you're the perfect person to answer it. To me, it seems like they're in a bit of a bind. They want to reapportion the seats in parliament based on the latest census, which means we know which states will gain. It's really the Hindi heartland states, which are much poorer. We know that West Bengal, Tamil Nadu, Kerala, and Andhra Pradesh are all losers in this bargain. Maharashtra could lose one seat, and so on.

Now, to me, it seems like they've been trying to gain a foothold in Tamil Nadu and West Bengal, and Kerala for a long time, and the BJP has never really succeeded in those states. They didn't want to have future election campaigns to be based on, “BJP can't win in the state because at the union level, they're going to quash our language, they're going to take away all the state's resources, and redistribute it to the Hindi heartland.” Their way out was to build a larger umbrella campaign around women, which clearly didn't work out.

What do you make of that argument? Because it seems like the BJP always tries to build the broader coalition. They tried to do the same thing when it came to caste politics, which was “Let's unite all Hindus under one umbrella.” Is that the thing going on here? That because we can't win these state elections based on language, finances, and those kinds of discussions, let's try and build a different coalition?

VAISHNAV: I don't know, Shruti. I find it very hard to believe. There is this conventional wisdom out there, which is this would all be great for the BJP because their strongholds are in the central states—the northern states, which would gain. They could lose everything else. If you think about their long-term project, and this has been true for decades, and I think it's very much pronounced. As we speak right now, their long-term strategy is to build inroads, and they have built inroads. Maybe not as significant as they would like in the east, in the northeast, in the south, where they continue chipping away, increasing their vote share, and winning seats some years more than others.

I would be surprised if they're willing to forsake all of that work for this kind of a victory, which is why I have a hard time believing in their heart of hearts, they really wanted this to happen. I think one of the things which seems more likely, and it comes back to something we said earlier, is you have the text of this bill, but you also have this newly super empowered delimitation commission. You make them the arbiter, and you seek to influence them. They are the ones who are going to carry out your guarantee of statewide shares, or not. But you appoint those people, you control them, and you exercise influence as the executive.

Now, as a constitutional scholar, I ask you. This strikes me as an incredibly legally dubious strategy, because how will that be held up eventually by the Supreme Court if the text of the bill itself is so black and white in terms of how apportionment should be done?

RAJAGOPALAN: Yes, it is totally legally dubious, which is why they've also embedded the “can't be challenged in court part,” because I think they expected it would be challenged at court. Even before the delimitation bill, I think the constitutional amendment would have been challenged in court, for which there is no immunity against that because of Kesavananda Bharati and the basic structure of the constitution. I mean, the Supreme Court can at any point reopen any constitutional amendment question that it wishes to reopen. There, I do agree with you that it is a little constitutionally and legally dubious. In all fairness to them, we have been for 50-55 years, in a constitutionally dubious place. Why should we place any higher constitutional morality or imagination to this crazy political bargain that happened in the middle of the emergency—the worst time of India's democracy in some sense—and actually give weight to that and continue it for over five decades? It seems to me, we were already in some political compromise, except the political compromise was made by our grandparents’ generation.

The only real difference is now the political compromise will be made by our generation. I guess we are always more skeptical of how this will play out now because it seems a little less clear in terms of the legal text. If this was always going to be political, then past a point, what difference does it make? That could be one kind of attitude.

VAISHNAV: I think it might be useful to discuss what the state of play is now.

The Political Arithmetic 

VAISHNAV: The government bills stand defeated. Basically, the existing ex-ante constitutional framework remains intact. That imposes clear constraints on what can happen next. For now, the size of the Lok Sabha remains capped at 550 seats. A long-standing freeze and reapportionment across states continue. Crucially, there can be no fresh delimitation exercise until the results of the next census, which is currently underway. It formally started on April 1st of this year. Until those results are published, the map will remain frozen unless the government is able to come back and broker some kind of compromise.

When that delimitation happens, after the census results are published, it will be population-based unless the Constitution is amended. Again, that applies both to the allocation of seats across states as well as within them. That same sequencing, that schedule is going to apply to women's reservation. Again, that implementation is tied to the completion of the fresh census delimitation. Even though the government has formally notified the enactment of women's reservation—something that happened as this debate was happening—nothing can really happen until this current census is completed and a due delimitation is carried out.

In the short run, the opposition can claim a victory. It has stymied this rushed controversial legislation. The underlying issues, these cross pressures between democracy, one person, one vote, and how you think of the federal compact, those issues are essentially just been deferred. It's the next census—again, assuming nothing happens in the interim—that's going to set in motion a highly consequential process that's going to determine this negotiation between and within states.

RAJAGOPALAN: Can you give us a sense of what the current map of India looks like if we use numbers from today, which are obviously projections based on the 2011 census and the 2019 projections that were based on the 2011 census and so on? Do you have a sense? We don't have time to go over every single state, but the greatest hits, the biggest winners and the losers.

VAISHNAV: Let's assume, for instance, that you have two baselines. One is the baseline of the 2011 census. The second is a baseline of what the population looks like today, which we don't know for sure. In 2020, the government actually updated their population projections. We have some sense, it's only about six years old, of what the numbers would be across the country within states as of 2026. If you use the 2011 baseline, again, assume that the 543 remains untouched. The three biggest net losers would be Tamil Nadu, Kerala, and Andhra Pradesh. Tamil Nadu would lose seven seats. Kerala would lose five seats. Andhra Pradesh would lose three.

Then if you look at the flip side, who were the net gainers? Uttar Pradesh with eight. Bihar with six, Rajasthan with five. A pretty significant transfer. There are only two states, according to our calculations, that would be untouched, Assam and Chhattisgarh, which essentially stay the same. Otherwise, everyone loses at least one seat or gains one seat.

What happens when you use more current projections of the 2026 population? Obviously, just those numbers get inflated. If Tamil Nadu lost 7 before, it loses 10. Andhra Pradesh loses five. Kerala loses six. That's 21 seats just with those 3 southern states that goes away.

Then on the flip side, Rajasthan, Bihar, Uttar Pradesh, they gain 26 seats as opposed to 19 seats in 2011. These are really significant shifts of power amongst each of these states, which are significant, large, consequential states.

The Financial Bargain

RAJAGOPALAN: I will add here that this is one of the biggest reasons why you haven't seen the states that would gain the most, say very much about it. For instance, you don't have the chief ministers of Uttar Pradesh, Madhya Pradesh, Bihar, and Rajasthan building the narrative that their people are getting ripped off, there isn't one person, one vote, and they're losing their voice in Parliament. The simple reason for that is the states that stand to gain the most are also some of the poorest states in India. They rely heavily on fiscal transfers from the richer states.

What we really see is some kind of political bargain that has been made in the past where it says that India is very fiscally centralized. Parliament decides how the money pie or the fiscal pie gets split. The larger the voice in Parliament, the more say you have over how to split the fiscal pie. But the money is not being generated in these states. In fact, these states are net recipients from other states. It is really Andhra Pradesh, Telangana, Kerala, Tamil Nadu, and Maharashtra—who are on the losing side of the seats if we reapportion with current population numbers—who are the biggest contributors to the national kitty or to the fiscal finances.

This again plays into the politics of it. The narrative building has not been north versus south, because that automatically also sheds light on the rich states versus poor states. If you want to continue taking these transfers, then you have to, in some sense, I guess, give up some political voice. The core of it is not just that the numbers change. It's the economics behind the numbers. The economics are simple. The less developed you are or the later you joined the development party post-liberalization, the later your fertility rates fell, which means the more malapportioned you are today based on a census from 55 years ago. That's the simple math of it. We can get into the more complicated math.

VAISHNAV: Yes. It's worth maybe saying, “What are the scenarios that are out there and that people have talked about?” We've talked about a couple already. One is simply extending the freeze. Passing an amendment to the Constitution saying, “We don't want to touch this hot potato.” We deferred this decision from '76 to 2001. We deferred it again. We're just going to keep deferring it for another 25 years, which actually wouldn't be a surprising outcome in many ways. That's one.

The second is you move towards some full population-based reapportionment. You could do that with the current size. I've given you those numbers. You could also do it with some expanded size. You could take the 850 seats the government envisioned and imagine what the interstate allocation would look like. There's also one idea that's been put forward in the past, what Louise, Andy, and I call in this paper “moderated expansion.” This is what they did.

I believe the principle they followed in the delimitation exercise after the '71 census where they said, “We're going to expand the Lok Sabha just enough to ensure that no single state loses seats.” Underrepresented states are going to gain additional seats to bring their representation closer to the population share, but we're not going to take away any seats from any state. We're just going to enlarge the pie. If you use the standard Webster method—you use your updated population projections, you do the math—that requires expanding the Lok Sabha. It's about 775 members.

Smaller somewhat than what the government envisioned, but still obviously 200-plus from what it is today. Now, what are the pros of that approach? One is, it's political. No state would lose a seat.

RAJAGOPALAN: They do lose seat share.

VAISHNAV: They do, of course, right. No person currently occupying the seat would… 

RAJAGOPALAN: Occupying, exactly, which is important, actually. That's an important element.

VAISHNAV: Which would get thrown out. The second, I guess, is a little normative where you're moving towards population parity without fully resetting the system. You do get closer to one person, one vote, but it's not the sharpest discontinuity. By stopping short of that larger expansion, you maybe also avoid some of the unintended consequences of moving to a really jumbo house with 850 members. Some would say 775 is still pretty jumbo. You know what I mean? Now, obviously, the downside is what you said, which is “Is this really going to satisfy the concerns of southern states?” They want to preserve the federal influence as it exists today.

Even if they're not losing seats in absolute terms, certainly the shares would diverge in relative terms. There is another set of options which people have thrown out, which are hybrid formulas. One option is what people have called digressive proportionality, where basically more populous states receive more seats, but with diminishing returns. You don't make the adjustment linear in proportion to the population. Again, that's a systematic compromise. You could come up with a scheme that says, “There's a floor. No state is going to fall below the floor. Then, on top of that, we're going to move to some kind of proportionality.”

There's also something which I know that has come into discussion in the Finance Commission context, which is: Let's use a blended census formula rather than going from '71 to 2011 or to 2027, whenever the census is done. Let's use some complicated arithmetic to smooth out some of the rough edges.

RAJAGOPALAN: None of them really gets to the heart of the problem, I think, even now, which is India in 1976 was a socialist state, but it wasn't a welfare state in the manner it is today, mainly because it didn't have the money or the revenues to be a welfare state. India really became a welfare state after 1991 when state revenue started pouring in and economic growth took off. A very large part of the reason to be in political power is to be able to redistribute. A very large part of getting to be in political power is making promises, or keeping the promises of redistribution made in the past.

If you just look at what is happening in each state, especially the poorer states—and Arvind and Devesh's most recent book, A Sixth of Humanity, is fantastic at documenting how much we rely on welfare transfers as opposed to building good old physical and human capital. Every interest group starts getting a welfare transfer. Women are the latest and largest of the interest groups, as you pointed out earlier in the conversation. To be able to do that, you need money. To be able to have money, you need economic growth.

It doesn't look like some of these larger and poorer states have really gotten their act together in recent times or even in the last 30 years. I still think we need to pay attention to the economics of it, because if we look at the current Finance Commission devolution, and this is not the proposals that we might have, this is literally what the commission says, there are two ways…

VAISHNAV: The report has just been tabled.

RAJAGOPALAN: Exactly. The Finance Commission has two methods of devolution. One is vertical, and one is horizontal. In terms of vertical devolution of the divisible pool, to which the center and the states contribute. That pool then gets split up between the Union and the states, and that's not equal. The states get about 41% from that divisible pool, and the Union gets the remaining. That's one part of the split. The second part is the horizontal devolution.

VAISHNAV: How it gets shared by the states.

RAJAGOPALAN: Exactly. Now, that 41%, how it gets shared by state is based on a kind of weighted average formula. The weights are poverty and population, the two things that basically stack the odds in favor of the poorer states. When I say poorer states, I just mean they're the ones who rely on intergovernmental transfers. From the overall kitty and from the richer states, you have this movement of revenue going to these poorer states. It's going again in proportion to population and poverty, which is exactly what the malapportionment problem was about, too. It was based on population and poverty because poorer groups saw their fertility rates drop later.

The whole thing seemed a little too interlinked for us to ignore the economics of it. I would put one more proposal on the table. Today, I think most countries don't have states with equal representation in Parliament. As states lose population numbers, they have lower representation, and that seems fair. In most other federal republics, you don't have this kind of centralized distribution of finances. If you allow most states to keep their finances—if the Union-State vertical devolution was 90% devolved to the states and 10% left with the union—you would not have the southern states kick off such a big fuss because their people lose, too.

Let's not forget that they also have poorer and marginalized groups that suffer from having constituencies delimited from a 55-year-old census. They are the more progressive states. I would also want to put that very much on the table because I feel like if we don't solve the fiscal bargain in some way, we can't solve the constitutional bargain in a way that would reflect the principle of one person, one vote.

VAISHNAV: Yes. I think that's exactly right. It would also require, without going into too much of the details, that divisible pool that you mentioned, which is essentially the money that then gets divvied up. The union government has also worked assiduously to try and shrink that divisible pool by having cesses, surcharges, and other things which go around that. You would have to have a cleaning up of all of that, including maybe increasing the number. I think you raised the larger point, which is you have at least four crosscurrents going on at the same time. We have been talking about them largely in isolation, including much of this conversation.

One is this question of political representation and how that squares with population. That's a ticking time bomb in some sense. The second is this feeling of resentment around fiscal transfers and rich states put so much into the kitty, and they only get so much out. It's unfair because we're subsidizing poorly governed states that can't get their act together. Clearly, that sense is building up. The third is the nature of taxation has changed. You now have a unified goods and services tax, which has limited the degree of state autonomy in terms of tax reform and tax innovation, because now you've cast your lot with this new system where you don't have as much autonomy or sovereignty.

You have a pooled sovereignty where you share tax jurisdiction with the center. Then the last thing is, of course, interstate migration, where you see a lot of people moving from poorer places to richer places, whether that's Maharashtra, Gujarat, Tamil Nadu, or Karnataka. That has helped the economies of certain states, you could argue. But it's also put pressure on them in terms of inadequate infrastructure, welfare, transportation, which led to a lot of calls for job reservations for locals, for “sons of the soil.”

There are all of these things, and they're all inextricably linked. It seems to me that we are at a moment akin to another—people have said this—of a States Reorganization Commission-like moment, which is not just about state boundaries, but really has to be part of this broader conversation.

RAJAGOPALAN: I could not agree more. I think in addition to the GST Council and the splitting of finances and so on, there is another conversation to be had about portability of benefits and what kinds of subsidies we give. If most of the subsidies are being given to agriculture, which are highly local, trap people in a particular state with portable benefits, then what you end up seeing is that the money is leaving the richer southern states. It's remaining locked and trapped in these poor agrarian regions in the northern states. Plus, you have all these seasonal workers and so on moving to the southern states.

That's what causes this NIMBY-ism. If we fundamentally adjust and make these benefits very portable, maybe the southern states would also feel very differently about who is coming to their states. Yamini Aiyer has made this point about making welfare entitlements more portable. Lots of people have talked about this, but that seems like another piece of the puzzle where maybe migration will take care of it. But for that, we need to solve the fiscal puzzle.

VAISHNAV: Can we just go back for one second to something I wanted to circle back to? It's come up a couple of times, which is the Rajya Sabha. I think it is also important in this conversation to think about what the objective of this house is and how to make it a more effective representative of states' interests. After all, it is meant to be a council of states. I think most people would agree that is not the way it's turned out. There are things that one could do there. Number one, you could think about—it's a radical reform—changing the composition of the Rajya Sabha to reflect the equality of states rather than their relative size.

We take it out of that proportional ball game, like the US Senate. The downside—Madhavan has a great piece on this in The Hindu—is you can't just give each state two, because remember, there are times when you're going to have a joint sitting of Parliament when, all of a sudden, you're going to have a really shrunken Rajya Sabha and potentially a really large Lok Sabha.

RAJAGOPALAN: A huge Lok Sabha.

VAISHNAV: It's fine if the Lok Sabha has the upper hand, but it can't be so out of whack. You'd have to think about what that equalizing number would be.

RAJAGOPALAN: Even more than just the equalizing number, Milan, I think it's also that we can't give states equal numbers because in India, you can just create a state very, very easily. We created states by voice vote in the past, which is Article 3. If we're going to go the US Senate way, let's say we give them 15 seats each to solve the Madhavan problem. You still need to adjust how states can be created and destroyed. We can turn states into union territories, and we have. I think those sorts of things also need to be baked into the constitutional design. 

VAISHNAV: Then another thing which, again, people have talked about what I think is really important is there was a 2003 amendment to the Representation of the People Act, which essentially dispensed with the notion that if you were a Rajya Sabha member, you had to be domiciled in the state that you represent. Now, that had already been hollowed out. I think famously, it was-

RAJAGOPALAN: Manmohan Singh.

VAISHNAV: Manmohan Singh was an Assam Rajya Sabha MP. You would have to not just reinstate the domicile requirement but do something even stronger. One solution to that, it's indirect, is to say, “Actually, make these people directly elected and make them accountable to you as opposed to just indirectly chosen.” There may be other administrative ways of doing it.

There's one final thing to mention, and this actually relates to a question that somebody has asked in the audience, which is about the Tenth Schedule, commonly known as the Anti-Defection Law, which most people on this call would know is an amendment to the Constitution passed in 1985. It says, in short, that if you are a legislator at the state or national levels, and you defy a party whip, which is basically to say if you vote in a way that's contrary to how your party has ordered you to vote, not only can you be kicked out of the party, you can be disqualified from Parliament. 

One of the things it's done over time is really concentrate power in the hands of party leaders and has taken away agency from ordinary representatives. The question that was asked is, what does it matter if we increase the number of MPs if they really can't vote for themselves? I think that this is particularly an issue in the Rajya Sabha, because if you want it to be reflective of state interest, these people then have to be able to say, “Actually, it’s in Rajasthan's interest or Punjab's interest,” and so on, and right now, you really can't do that because you will lose your seat.

RAJAGOPALAN: The purpose of being an elected representative is not just voting in Parliament and voting your conscience. There are also a lot of things done on the ground. There are things like MPLAD, which is the money that is disbursed, given at the constituency level. It is also about aggregating preferences. Let's say they're not voting their conscience in Parliament. If they're self-interested politicians, at the very least, they're taking the preferences within their constituency to the party. I do think it is important to have better representation, notwithstanding the problems caused by delimitation.

To go back to your Rajya Sabha point, I think the most important reform we need in the Rajya Sabha, and this is irrespective of the delimitation stuff, is we need to require money bills to pass in Rajya Sabha. This is essential. Money bills can just bypass Rajya Sabha. Of course, the downstream consequence of that is whenever the government doesn't have the votes to pass something, they add a budget or appropriations element to it, and turn it into a money bill, which is what they did with the Aadhaar Bill. Remember, Aadhaar was passed as a money bill, which is a little hilarious.

If money bills under Article 109 are not required to be tabled and passed in Rajya Sabha, then we have a very big problem when it comes to the representation of states and how we split the fiscal pie if we are this centralized.

VAISHNAV: I think the other wrinkle in that is the fact that what people have designated money bills, at least to a lay person, surely don't seem like money bills. The Supreme Court of India disagrees with that assessment. Maybe it's time to open up to questions. I think people have already put in some questions in the chat. We will try and take as many as we can in the remaining 18 or so minutes. I think I’ve answered the one on the Tenth Schedule, which is, I think absolutely right. If you increase the number of MPs, you're not maybe getting as much bang for the buck because people can't really vote for their interests.

Shruti, there's another question about safeguards and delimitation. How do we ensure that the next delimitation, once the commission is set up, does not become a tool for partisan boundary drawing, especially given the concerns that we mentioned already about Assam and J&K? I don't know if you have thoughts on that.

RAJAGOPALAN: I think it's always going to be political. This is true not just in India; it's true across the world. The question is, can we minimize capture? I don't think we can eliminate the political nature of the exercise. I do think the way the current Delimitation Commission was floated through the bill that did not pass or wasn't even put to a vote, it says that the speaker would pick five people from Parliament or the state legislature, in addition to the election commissioner and Supreme Court judges. I think we explicitly need to create room for opposition.

VAISHNAV: They have to be cross-party.

RAJAGOPALAN: They have to be cross-party. I think that's an important piece of the puzzle. Whatever we're talking about at the parliamentary level is quite symmetric and mirrored at the state level. The same reform would have to happen at the state level. That would be my major concern. For instance, you have a quarter of the seats reserved for SC/STs. You have a quarter of the seats reserved for women, but that is not reflected in the Delimitation Commission. Now, do we need to have this kind of reservation or representation all the way through or not? I'm not saying one way or another, but at least it's a topic of conversation. It should be put up for debate.

VAISHNAV: I would just say, Shruti, that the reason this is such an interesting issue to me is India, for most of its post-Independence history, has actually managed to create a kind of independent technocratic way of redistricting. You just look at the shenanigans in the United States, where it is so broken down. Now we're just seeing a kind of arms race with Democratic and Republican states trying to one-up each other because that's the incentive. I don't blame the Democrats at all for doing what they're doing, because if you're not going to have mutual disarmament and you just have unilateral disarmament, then that's politically disastrous.

I think there are certainly, when you pass the enacted delimitation bill, some hard legal criteria that you do want to encode. Obviously, population equality is one. Geographic compactness is another. You want to have contiguity. You want to have respect for administrative boundaries. If you look at some of these maps that are put out in Assam, you have natural geographic barriers, or you have islands of one enclave of a constituency that's fully enclosed by another. The other thing they did when they redistricted, I believe it was in Assam (someone could fact-check me), is they randomly introduced population density as a measure, which was not something that they had done before.

RAJAGOPALAN: Which is complicated in a hilly state to begin with and is very different in the plains. It seemed very targeted to get a particular kind of outcome.

VAISHNAV: Correct. One, there are certain things you can do and insist upon. The second is, obviously, there should be some modicum of transparency around the publication of draft maps, justifications for close calls, and decisions you had to make. One thing I don't have the answer to, Shruti, and I could be mistaken here, is that I think it has been a norm that the Delimitation Commission essentially gets a thumbs-up, thumbs-down vote in Parliament. It can't be amended—the judicial review piece, I'm not sure.

You highlighted that as something in this recent bill, which clearly is problematic. You don't necessarily want the judiciary to get involved in influencing outcomes, but you certainly do want them involved if there are violations of process.

RAJAGOPALAN: That's the thing. We have so much delegated legislation in India that if you look at the delimitation bill, it's pretty sparse. Then you'll say the Delimitation Commission or the relevant authority has the power to make rules. Now, those rules have to come under challenge if they violate rule-of-law principles, and if they're arbitrary or discretionary. Removing the courts from that ambit is a problem. I completely agree with you that I don't think the courts should be involved in everything.

In fact, I think Indian courts talk too much about too many things and literally don't do the job they're supposed to do, which is resolve disputes and conflicts, and speak on constitutional matters like electoral bonds or demonetization in a five-year period. Having said that, you do need some check. My recommendation to Parliament would be: If you're going to leave the courts out of it, then don't have delegated legislation when it comes to delimitation. Flesh out the bill at a level of clarity that when we pass it in Parliament, we have a pretty good sense of what is being passed.

VAISHNAV: Yes. I think, bottom line—and there are a lot of questions on this, which I think we've answered in terms of this question of judicial review: What does a good-faith Delimitation Commission look like? I think you and I both agree that you want it to be independent, but you need to have some constraints on how the commission exercises discretion, which means you have to have something hard-coded. You also probably have to have some process where you have representatives across the political spectrum who are giving input.

There's a question here, Shruti, which I think you went over in your response on the fiscal federalism side, which is: What does a good-faith bargain look like between the BJP and southern regional parties? Is it enough to say, “Look, we are going to come up with some way of making representation more proportional, but the quid pro quo here is that we're going to take devolution from 41% to 71%,” or whatever the number is. Do you think that is a bargain that southern states would go for?

RAJAGOPALAN: Yes. I would actually call this a bargain, not just between the BJP and DMK. I think this is across the board, because remember, Haryana is a net contributor. Gujarat is a net contributor. There are many states that are not dubbed southern progressive states that are part of this puzzle. I do think they need a larger bargain in Parliament across the board. The BJP tends to represent the most number of people from the poor states or Hindi heartland belt. I think they are involved in that end of the bargain. The opposition has shown that they can come together when required.

I think even within Tamil Nadu, you will see all the parties get together, because whenever one of them comes to power, there has to be money to actually govern. You'll see a lot more solidarity when it comes to fiscal bargaining than we see in all the other kinds of bargaining over Hindi, Ram Mandir, or any other hot-button issue. I do think they need to come up with some bargain. I think the fiscal devolution is probably the best one. Another thing is to hard-code fiscal devolution into the constitution. Maybe have an eclipse clause that each year there is going to be greater devolution over a 20-year period and simultaneously increase proportional seats in Parliament.

Again, we don't want to leave the UP-ites and the Biharis without any revenues. This is the poorest group of people in the country. They do need some investment in health, education, and physical infrastructure. That would be the nature of the bargain, I think.

VAISHNAV: There's a question here about whether there is a way to use sub-state data of a more granular kind. I think it's a really interesting question. Certainly, given the kinds of satellite data, terrain mapping, and stuff you have, that could be used as an input to ensure that these maps are done in ways that are contiguous and don't do things, like cross over rivers and places that are unnecessary and so forth. I think that's certainly one. One thing we didn't talk about, Shruti, but it gets to the politics a little bit, and I thought about this in thinking about the sub-state data, is you could make a plausible argument that one reason the government did this now is because they want to avoid dealing with the caste census.

You don't want to be in a world, perhaps if you're the BJP, where you now have to consider reservations in Parliament for OBCs and others if that cuts against your desire to keep Hindus united as one bloc. Obviously, there have been a number of views within the BJP on reservations, but quite a lot of segments of the party have been anti. This was one way perhaps to nip it in the bud because now, assuming nothing happens and status quo prevails, that is going to be an issue, because that is the political context we're in, where there will be a melding together of caste beyond SC/ST and the next delimitation.

RAJAGOPALAN: Yes and no, because in absolute numbers, the largest number of SC/STs are in these states, which are net receivers of revenue and also in the states where the BJP is relatively more powerful right now. In some sense, attaching delimitation to the caste census is the opposition's way of perhaps shooting itself in the foot, which I trust they do very well all the time. Maybe they haven't thought this through quite as well. I'm not a fan of the caste census at all. This is a very different reason from delimitation. I think the goal was to annihilate caste, and the caste census basically constitutionalizes caste.

It's like, “Oh, we're going to have this political and caste identity we can never shake off as long as the republic looks the way that it looks.” I'm against it for many other reasons. This would be the craziest reason in terms of combining the two. It would be crazy for the southern states to do that. The northeastern states, let's not forget, that's where most of the Scheduled Tribes are, and they are huge net receivers of fiscal transfers. Both on the Scheduled Caste issue, who are the most numerous in the poorer states where fertility dropped later, and in the northeast, which are net receivers, you're in a little bit of trouble if you're the opposition parties.

VAISHNAV: Yes. It's a good point. I think somebody has a question about whether we are making a mistake somehow by conflating things. In other words, delimitation is about political representation, what does it really have to do with fiscal federalism? Pranay Kotasthane, who's somebody that you and I both read and have had on our various podcasts, tweeted about this after the failure of the bills, linking to a piece he had done, I think in 2023 or so, where he basically said, “The Delimitation Commission is not the Finance Commission. It's not the Population Commission of India. It has a job, and it should do it, and then we have to figure out how to adjust.” I have some sympathy, I think, for that argument.

RAJAGOPALAN: Just to clarify, I don't think delimitation should split the fiscal pie at all. That would be a terrible and ridiculous thing to do. I think the fiscal pie needs to be split based on what we have right now with the Finance Commission. I think Parliament needs to actually discuss and vote on it. Except my simple point is, unless they devolve revenue more, there is no bargain to move towards lesser malapportionment. 

VAISHNAV: I think one of the things that Louise, Andy, and I have been working on in a separate piece that's going to be part of a book that Yamini and I are editing on federalism is thinking about delimitation, both in a theoretical sense and an empirical sense. One of the things that Louise has argued is that part of why we are where we are is because India is not what federalism scholars have called a “coming-together style federation.” It has operated for decades on the basis of an implicit federal bargain that has never really been explicitly negotiated.

The delimitation freeze, which obviously preserved representation for states that had, in fertility terms, grown at a slower rate, was a concession that helped to partially offset the Constitution's built-in majoritarian design. These were not constitutionally enumerated. They were norms that helped to stabilize this federal model. I do think that's one reason why you and I, in this conversation, are saying we have to think of all of these things holistically to a certain extent, because, like it or not, there were some norms of an implicit federal bargain that are at risk of unraveling for a whole host of reasons.

You have a single dominant party and massive growth divergence across states. India is one of those federal countries that has not seen convergence in economic outcomes. We've seen further divergence over time. I think that's also important to note.

RAJAGOPALAN: No, I completely agree. Do we have time for one more question?

VAISHNAV: I think we have four minutes.

RAJAGOPALAN: Oh, wow. Let's see if we can do one more question. 

VAISHNAV: Oh, I’m seeing one question here on fiscal devolution in terms of moving beyond poverty and population. Shruti, you didn't talk about this, but it was notable—and people did note it—that the most recent Finance Commission did include a criterion for GDP. I think that was seen as a fillip towards dynamic states such that part of the formula now is “working in their favor,” as opposed to the other way.

RAJAGOPALAN: It is. I'll have to run the numbers, but I don't know how this plays out, because given the population weightage and the poverty weightage, I still think this is heavily tilted towards the poorer states when it comes to horizontal devolution. I agree with you that it's not as lopsided as it was in the past. There is some reward for being economically more dynamic. I don't think it does enough to offset it.

VAISHNAV: Yes. One thing, Shruti, that somebody asked, and maybe we can end on this, is if states that have a lot of representation like UP and Bihar, have not actually become prosperous by having more MPs. What is the link between representation and development? To flip it, as this questioner does, “Do the southern states really have a lot to fear from losing seats?”

RAJAGOPALAN: Yes, because the money bill will be voted on in the Lok Sabha. There's a huge incentive for every state parliamentarian to send more money to their state. That's how you win elections. I think the BJP has shown that it's extraordinary at this. Of course, the fact that there was an Aadhaar plus UPI revolution and Jan Dhan account revolution in India really helped streamline transfers. In the time when the delimitation freeze was brought in, most of the welfare entitlements and transfers leaked, and they weren't that large to begin with.

Now, actually, the welfare entitlements and transfers reach the people they're intended to reach. They're a huge part of the voting politics in every state election and every Union-level election. If you don't have money to redistribute, you're in trouble. If the southern and richer states lose their voice in Parliament, more of the money will go to the poorer states. I would not be surprised if that happens at all.

VAISHNAV: Just one final thing on this. I think one thing that is true, whether we like it or not, is that members of Parliament and MLAs are not, first and foremost, lawmakers. They are people who perceive their jobs and who are perceived by their constituents as being intermediaries in the provision of public benefits and services. That is the de facto reality in India. I think having a constituency size of 2.7 million becomes somewhat unsustainable on all sides.

I'm not saying this is what I aspire for the link to be between MP and constituent. I think there have been great studies from the 1970s, '80s, and '90s, which have gone around and surveyed MPs and MLAs. What's interesting is they all say, “This is what we're there for.” Their constituents also say the same thing.

RAJAGOPALAN: I think this is a good time to end. I'm sure we'll have more of this conversation because delimitation and apportionment have not been resolved in any way. In fact, we just kicked the hornet's nest with this special session of Parliament. Thank you so much, Milan, for doing this. This was a lot of fun for me. Thank you to everyone who joined the webinar.

VAISHNAV: Thanks, Shruti, for inviting me. Thanks to Mercatus, who put all of this together and did all the heavy lifting. Look forward to continuing the conversation.