Kishen Shastry on the Political Economy of Hindu Monastic Institutions (Mutts) in Karnataka

SHRUTI RAJAGOPALAN: Welcome to Ideas of India, a podcast where we examine academic ideas that can propel India forward. My name is Shruti Rajagopalan, and this is the 2026 job market series, where I speak with young scholars entering the academic job market about their latest research on India. 

I spoke with Kishen Shastry who is a Postdoctoral Research Associate at the Cambridge Centre for Alternative Finance, Judge Business School. He received his Ph.D. in economics in 2026 at the University of Cambridge where his research focus was economics of religion and informal institutions. We discussed his job market paper titled, “The Political Economy of Informal Institutions in India.”

We talked about the economics of religion, the role of long standing monastic institutions in India, how they are regulated by the state, and their impact on education, structural transformation, political competition in India, 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, Kishen. Welcome to the show. It’s so great to have you here.

KISHEN SHASTRY: Hi, Shruti. Thank you so much for having me today.

Hindu Mutts and Their Relationship to Structural Transformation

RAJAGOPALAN: Yes, your paper is super interesting because the only context in which I ever hear about Hindu mutts is when my parents or grandparents wanted me to visit one [chuckles] in the southern parts of India. So, reading your paper is making my lived experience and my childhood come alive and clash with what I do for a living, which is economics and structural transformation.

In this paper, you ask whether Hindu mutts, which are longstanding monastic institutions—in your instance, you’re looking at this in Karnataka, and you ask whether Hindu mutts helped shape the structural transformation and political competition in Karnataka. More specifically, you’re studying whether the 1991 reforms had a greater impact in those areas that had greater exposure to these Hindu mutts or greater proximity, in some sense, to these Hindu mutts.

And when we’re talking about reforms, you’re looking at two sets of major reforms that happened in India in the ’90s. One is, of course, the 1991 reforms, which was delicensing, trade liberalization, getting out of License-Permit Raj. The second part of it is the political decentralization, which is the 73rd and 74th Amendment combined with the Panchayati Raj Act in the states. In your instance, in Karnataka in 1993.

Now within this world of before and after these reforms, you managed to construct a phenomenally new dataset of 253 such Hindu mutts. You look at, across Karnataka at a district level, what is the exposure and proximity to these mutts, and what is the impact. You look at the impact on very specific instances.

One is, of course, what that does to the structural transformation. One way of thinking about that is nonfarm jobs versus farm jobs and how that changes anything, if at all. The second part of it is whether there is a greater uptake in secular education, not just in religious education, and what you find there. And the third is on the political side, if there’s greater political competition.

If I were to just give away the headline results, what you find is that mutt exposure is associated with 10 additional nonfarm jobs per 1,000 residents, roughly 15 percent to 20 percent relative to the 1990 baseline. Most of these gains are concentrated not in agriculture, but in manufacturing and services. So, that’s the structural transformation throughline. More exposed places, according to you, also see about a 9 percent increase in secondary school or senior secondary and college education per capita. And on the political side, you find that when it comes to assembly elections, the margin of victory is narrowing by about 0.7 percentage points from 1994 onwards, because the Panchayati Raj Act kicks in in ’93.

This is the overall picture. But these two things normally seem seemingly unconnected, that is exposure to these kinds of monastic institutions, which we associate with a very historical, long-run sticky institutions, and what is happening with structural transformation today, right? 

So first of all, did I get the paper right? Did I capture what you were trying to say? And second, what made you write this paper, and what we’re really trying to look at?

SHASTRY: Thank you so much, Shruti. I think that was quite an accurate description of the entire work. Generally, I find mutts to be quite a fascinating institution. It has a long history, as you’ve mentioned. Some of the earlier mutts, they go back to 8th century, 9th century.

Then there has been consistent revolution in terms of how people think about religion, and mutts became somewhat of a manifestation of this understanding of religious education overall.

I grew up in Bangalore, which is in Karnataka, and this is the main state that I take for studying. And as you know, economists are always interested in the tails of any kinds of distribution. In this case, it so happens that because these Hindu mutts specifically exist at the scale and size not found elsewhere outside of Karnataka, it was a natural place for me to study, Karnataka.

While I was growing up, of course, there’s a lot of journalistic coverage on these Hindu mutts, predominantly of two types. One would mostly focus on the kind of political lobbying or liaising that Hindu mutts would do.

And this is just to flag to everybody, this is across party lines. It doesn’t have to be just for a political party. This gets an enormous amount of attention. But the understated bit was how influenced and embedded they are within the society, particularly through their societal, or rather philanthropic, work that they conduct. Again, the scale and institutions that they generate is quite understated.

So, I was very keen when I got into economics to try and put these two things together: the political aspect and perhaps the kind of developmental impact they may have together. I think this paper largely attempts to do the same. And it is a bit surprising to me that not many people have done this kind of work in the past. In fact, the only two large, major works that exist on Hindu mutts in Karnataka are both either ethnographic accounts or by anthropologists. We have to thank—at least I have to thank them a lot for giving us fantastic primary resources to rely on, define what a mutt is, figure out what the kind of model they think about, and then bring that to economics.

As you put it together, I think getting the dataset here is not the easiest thing to do. And as I said, the interesting bit about Hindu mutts is perhaps they’re the only institutions of Hinduism in the entire country that are fully autonomous. This legal quandary in a way generates interesting incentives for us to study them.

Understanding Mutts

RAJAGOPALAN: Yes. So, before we get into the legal incentives and how mutts specifically were not treated the same as Hindu temples and so on, can you just walk us, for those who are unfamiliar with the institution, how has this longstanding Hindu institution functioned historically? What role did it play in the past? What is the role it plays today? And can you tell us something about its closest analogues from other religions, for those who are unfamiliar with Hinduism?

SHASTRY: Absolutely. I think oftentimes these mutts get translated as monastic institutions, and that’s what I’ve done, but they don’t fully do justice to what they exactly perform as their duties. A description in the nature of “monastic” gives a feeling that they are removed far from society. If anything, mutts are opposite to that. They’re deeply embedded, very active across different areas of the society.

But just to define them, the idea of a Hindu mutt has got to do with a monastery-like institution, often led by a guru. The way the guru is chosen is not through hereditary, as so happens mostly in temples, but they’re chosen to do this. And this guru is the prime renouncer of materialistic life. But if you think about why this becomes important, each Hindu mutt is generally associated with a caste, specifically in Karnataka.

So, if you have a pool of resources that a caste tries to either put to some use or collects together, they would rather do it to the renouncer because it’s a commitment device, in terms of an economics language. So, this guru becomes the central aspect to these mutts’ organization. Of course, he has tremendous autonomy to choose his team and run these organizations. This is what makes the institution quite different from the temple.

And in terms of how it is comparable, honestly, it’s quite similar to how a Catholic missionary would work, in that the order has a strict hierarchy. Again, the celibate life of the gurus there, equivalent guru system there. And they run a large network, not for the state, but for the community. So, in a way, the design is very similar, but often the structure, in a way, predates perhaps some of the other institutions before even they come to India.

The modern-day pioneering of the mutt organization was initially laid in the 12th and 13th century by Basava. He’s the founder of the unique way of thinking of Lingayats. It’s very similar to, in my opinion, how Martin Luther comes up with, in a way, opposition to Rome’s power, versus in this context, similar to opposing the Brahmin context. And there, they were opposing Latin. Here, in this case, it used to be the Sanskrit— 

RAJAGOPALAN: Sanskrit, yes.

SHASTRY: —controlling of the institutions. So, there are striking similarities. And of course, the economics of religion has done quite a bit of work on testing all of these hypotheses in the Christian world quite a bit. I try to do that for the Hindu world here.

RAJAGOPALAN: Yes. Since you talk about the economics of religion, we’ve had a few economists like Sriya Iyer, who’s also from Cambridge like you, and Rohit Ticku, and so on, but this is not an area too many people are familiar with. 

The way I think about economics of religion is it has a very strong literature on religious institutions providing club goods and solving very specific collective-action problems. Is this the way we should think about the Hindu mutt in its modern-day context? Not just about standing up to a particular kind of existing clergy or existing norms and institutions, not just about representing different castes or Vaishnavites versus Shaivites, which is also a major distinction in the different mutts. But what is a good way to think about how a mutt operates today? What does it produce, who produces it, and who are the chief consumers? Or whose incentives should we account for?

SHASTRY: Absolutely. We have a long list of people to thank in this subfield of economics of religion. Sriya Iyer, who happens to be my supervisor, is one of them. But the tradition goes slightly longer. In fact, I think even Adam Smith began some of his work trying to explain the incentives of the church and why some churches don’t do well. So, it begins with the regulation of religious markets. Larry Iannaccone comes up with this superb set of papers in the 1990s trying to explain—

RAJAGOPALAN: At George Mason University [chuckles].

SHASTRY: Exactly. Absolutely. I think it’s been a fantastic home, the university. Larry Iannaccone comes up with this mechanism to explain why religion can be considered a club good. The basic framework is that these clubs produce goods that are useful for all the members, but it has to have certain kind of strictness or a cost to enter this club. How do we do that? Typically, his paper was wonderfully titled, “Sacrifice and Stigma.” That opens up the world of economics of religion, particularly looking at this as a club good.

In terms of mutt, I would say it is similar to the club goods, but more so I would also pair this with the work that has been done in the economics of identity, particularly started again by many doyens of the field, including Akerlof and Kranton’s 2000 paper on explaining the incentives of creation of identity. 

Another fantastic scholar, Jean-Paul Carvalho, who’s at Oxford, he wrote a superb paper called “Identity-Based Organizations.” The idea here is, there is identity of any organization, or the people who are part of that organization, is a co-creation process. In fact, in this paper, he takes up a leader of this group who ends up creating an identity along with the followers. So, everybody has to put in their work. The identity could be used to solve many problems such as collective-action problems, coordination issues, and so on.

I feel mutts could be located in the theoretical framework of how identities are produced. Think about caste-based mutts, right?

There are fantastic accounts of a Japanese anthropologist who came to Karnataka, learned Kannada. And when she speaks to the aew age mutts—just to give a brief history, the Lingayats are the ones who pioneered this model of mutts, but many castes eventually start imitating this model because inherently it’s quite useful to have a mutt like this. And all of them told her that the goal to do so is to unify people, to create an identity of oneness.

So, this is embedded in the process of mutts: to create an identity for caste groupings, which can eventually be leveraged to meet the caste groupings’ ends of some kind. And what I find interesting is it brings in some aspects of club goods. For instance, if you have to participate in this identity creation, there are necessarily some individual preferences that you give up. Perhaps I vote the way my group thinks or asks me to vote. I participate in certain kinds of rallies, or I go to these. Caste groupings end up giving some of their donations to the mutts. There is an individual cost to pay to be part of this group, but ultimately this identity helps them reach a lot more goals that they have created for themselves.

In the modern day, I think I would set apart two different timeframes. This is of interest for my paper as well. Some of the dominant castes have had a long history of having mutts. Particularly Brahmin mutts tend to be more associated with a temple and, let’s say, Vedic learning and Sanskrit become a part of that. But Lingayat mutts somewhat are supposed to be opposite that. Their mutts genuinely do not have temples in most cases because that was the original philosophy of Lingayat mutts. And Vokkaligas—that’s the other major caste group in Karnataka—they tend to have mutts as well because of how successful Lingayat mutts become in running, for instance, free hostels, schools, and so on.

So, later on, after the 1990s, a lot of other castes which now want to start bargaining with the modern state of Karnataka or any kind of representative—it could be bureaucracy, it could be the political parties, they start imitating this institutional setup. Virtually all major caste groupings in Karnataka have their own mutt or collection of mutts. 

So, the modern-day setup is quite interesting. There is a market of entire mutts, in a way, to speak in terms of economics. And all of them have different levels of strictness, of commitment towards religion, but more so they use it as a fantastic enforcer of collective action.

Mutts in the Education Marketplace

RAJAGOPALAN: Yes. Since it’s a marketplace, there’s also competition, right, between them, which is brokered by the philanthropic and charitable services that they provide and the kinds of donations they can attract. There are multiple different markets that they are trying to operate in.

SHASTRY: Absolutely. I think this is where I would also like to reference my supervisor’s work, Sriya’s work, on religious organizations competing on nonreligious services.

RAJAGOPALAN: Yes, yes.

SHASTRY: So, how do we differentiate in a typical Hotelling model from other organizations? I start getting into giving them, beyond just satisfying the religious beliefs and religious services, I perhaps give them free meals or provision of education, free hostels, hospitals in some cases. As far as even enforcing contracts in some cases, right?

All of these services offer a bouquet of things for people who follow these mutts. It’s an interesting mix of creation of identity along with offering all of these services to their constituents.

RAJAGOPALAN: Yes, Sriya Iyer was actually on the podcast. She was a very early guest, and she talked about exactly this paper of hers on the podcast. We’ll throw in a link for that. 

So, now that we have a pretty good setup, I want to come specifically to education because these are also set up originally as somewhat educational institutions coming from the monastic tradition. And you are also specifically studying educational institutions in Karnataka as a big part of your paper.

Now, when I go back and think about the economics of religion and how it treats education, there is an interesting tension there depending on which religion we’re talking about, and the effect can go either way. One possibility is that these missionary or mutt-like institutions, they are complementing secular education, oftentimes providing secular education in addition to a religious education, right? You see this most with missionaries and now, as you point out, in the case of some mutts.

The economics of religion scholars who have studied Muslim equivalents of missionaries or mutts, which is basically the madrasas or some of these Sufi-tradition institutions, they often find the opposite effect. That is, it ends up being a substitute for a secular education through the religious education that is provided. 

So, how do Hindu mutts, specifically in Karnataka, fit into this literature? Because I think you’re the first person to study the Hindu institutions when it comes to this very large literature on the economics of religion and provision of education.

SHASTRY: Absolutely well put. I think the direction runs in either ways. It’s so interesting that it cannot even be attributed to a religion for good. For instance, in some cases, Catholic missionaries, say in France—this is a fantastic paper by Mara Squicciarini. During the Second Industrial Revolution, she finds that presence of Catholic schools and education actually impact downstream literacy rates to go down.

In many other cases where Jesuits and Catholic missionaries go out, even including in India, we notice a large uptick of literacy rates and even downstream economic development. And as you’ve correctly noted, even in case of Pakistan, great work by Mirza and Malik, they show how shrines in Pakistan essentially substitute for secular education. They use a large shock to show that over time, high exposure to the shrines actually reduced the literacy rates in Pakistan.

Hindu mutts, interestingly—I would like to quote an example when I was going through these papers. I came across this fantastic ethnographer from Chicago called Aditya Sood.

RAJAGOPALAN: Yes.

SHASTRY: He was a student of Sheldon Pollock.

RAJAGOPALAN: Yes.

SHASTRY: He came to study mutts in Karnataka as a Sanskritist. He came to study mutts because he thought he could learn more of how Sanskrit education is delivered in Hindu mutts. And much to his surprise, he came to do this work, and he realized almost all education in these Hindu mutts tend to be either in Kannada, English, and in what are the vernacular languages, and the education delivered mostly was secular education.

RAJAGOPALAN: Yes.

SHASTRY: And this is what makes mutts, in a way, stand out quite a bit. There is a very interesting historical aspect to this. Also, the role of, let’s say, the princely state of Mysore was involved in this, mostly because most of the government jobs at one point were largely dominated by Brahmins. Early 20th century, setting up of Lingayat mutts wanted to also come into this space. They start getting some patronage by the princely state of Mysore, so much so that you have, for the first few chief ministers who would come to rule Karnataka, some of them actually came from these mutt-run institutions.

One notable example is B.D. Jatti, who became actually the acting president of India, came from one of these mutt-run hostels. So, they have a long history of actually dispensing education through secular means, secular education. And this has been a standout feature, and hence the results in my paper show that they complement and improve the delivery of more secondary education schools, senior secondary education schools, and sometimes even colleges.

And again, anecdotally, some of these mutts, just to name the Siddaganga mutt in Tumkur, they run over 150 education institutions. Most people probably would not even be aware of the size and scale through which they function. This includes multiple universities, medical schools, engineering colleges. The key feature, from my point of view, is that where it becomes interesting is the autonomy to run these mutts was given fully to the organization itself.

That allows the manager or the guru to dispense with the assets that they have collected for whatever they think is the best use. It could be a school hostel in the 1920s, or it could be an engineering college in the 2000s. They would have the incentives to respond to the changing scenario. And that, I think, was one of the dominant reasons why, in my paper at least, I find that education institutions go up, and that, in a way, pushes the structural transformation of the economy.

Mutts and the Regulatory Regime

RAJAGOPALAN: This is really hard to study. You’ve given us the simple version of the story of how these mutts have actually operated. But the mutt as an institution itself, it’s religious, but it’s also educational. It’s also charitable. It plays a political function. And then there is, of course, the modern law and regulation which interfaces with this, right? And I’m talking about both the modern law and regulation that governs how these mutt organizations are run, and also modern law and regulation on how educational institutions in the secular side are run.

So, how would you conceptualize the mutt as an institution from the point of view of modern legal institution? You’ve hinted at it a couple of times, that they were allowed to run them autonomously, but that’s quite different from the historical setting. So, can you walk us through, in the instance of Karnataka, what is this regulatory regime that they are really under, and then we can parse out the effects?

SHASTRY: Perfect. I think the way to understand the regulation of temples and religious institutions generally, especially for Hindu institutions, where they were left to the devices of state governments to take [the] call on this. So, most of the states in southern India, where most of these temples and associated Hindu religious organizations work at scale, inherit the Tamil Nadu’s laws on the HRC, or the Hindu Religious Institutions and Charitable Endowment Act. Notably, what happens in Karnataka is they borrow the same, but they explicitly mention saying the mutts are kept outside of this. They say so in the law.

And interestingly, in 2013, the Karnataka High Court strikes down the law, saying, “We don’t understand why this particular institution is kept out of purview of the regulation.” But the Karnataka Legislative Assembly—again, irrespective of political parties, everybody supports this, just to show how important they are politically—repromulgates the Act. They bring back the Act, again, keeping the mutts outside of this. It’s a legislatively conscious decision to keep them out. In terms of how they actually get set up, the—

RAJAGOPALAN: Sorry, in terms of context, this is very different because in India, different religious institutions are actually—they fall under almost different regulatory regimes, right? And, historically, the view has been that there is a lot more state interference in how a Hindu temple is run versus how other minority institutions are run. And this is a very interesting example where you’re talking about a majority religion, but because of a certain kind of statutory or regulatory structure, you find that there is far less state interference overall, which is the second part of what you’re saying, which allows them to adapt.

But the allowing them to adapt goes in both ways. Sometimes mutts can do a lot of good. Sometimes they can also go the other way. We hear lots of stories, not just in mutts, but in various places, about embezzlement of money or mishandling of public funds, and abusing certain religious norms, and so on. And largely, these mutts have been kept out of that state supervision. You find that has some incredibly positive effects, at least in the state of Karnataka. Sorry, I just wanted to set that up because I think most people don’t know to what extent we regulate religious institutions in India.

SHASTRY: That’s absolutely well put. And I think the personal conviction that I’ve had when I’m doing this kind of work is that free organizations are better than not free. This is, in a way, almost unanimous in the literature and the economics of regulation of religious organizations. But yes, I think temples continue to be regulated. The main idea being cross-subsidy.

RAJAGOPALAN: Yes.

SHASTRY: For instance, in Karnataka, we have about 35,000 temples. All of them come under the state-run regime. And the issue with that is, in my opinion, there is very little incentive for these temples and the people who run the temples to do anything innovative.

That, in my opinion, causes a big problem. For instance, how do you respond to modern challenges like mental health, teenage issues with drug abuse? Specifically, given that we have literature that shows religion plays an important role. Including papers by Sriya and Daniel Hungerman show this deaths of despair. People who are severely abusing drugs and alcohol have the best successes when they go to a church program.

In that way, I think religion, temples in India are severely constrained. And mutts, specifically Hindu mutts in Karnataka, offer a welcome change. That’s probably the only institution we can study at scale in India to produce any kind of evidence. So, if somebody tomorrow wants to lay down a basis for whether any institutions need to be free or not, I would recommend we need to have some kind of impact study like this. It so happens that I had, in my case, an interesting example of mutts, which has longstanding history, and it helped quite a bit.

Stitching Together the Data

RAJAGOPALAN: Yes.So, now, coming back to my original question, it’s really hard to measure the different things that you’re looking at. There’s a religious impact, an educational impact, a political impact, and so on. How do you actually lay that out in Karnataka? What kind of data are you even able to look at? I imagine you have to stitch together lots of different data in addition to the data that you put together on mutts?

SHASTRY: Absolutely, and because of the nature of mutts, or generally temples and Hindu religion, it’s very hard to find one point registry for finding the list of all of these mutts. I know this because I tried. I tried the first possible solution to get this going, but eventually I realized that exercise is futile and I couldn’t do it. What I essentially did is—let me put this out in three different tranches of work. The first was actually to have a dataset on mutts, and specifically for my interest in this case, these had to be existing before the 1990s, because that’s when the shocks end up hitting the economy.

What I did was to use Google Search API to go to every village and figure out mutts or different variants of mutts that are actually located on Google. In many cases, it won’t work because people may end up missing very small mutts. In my case, I don’t have a problem with it. I patently want large and influential mutts. So, I built a large dataset, and there’s a big problem with lots of mutts happening, coming over, but I de-duplicate it quite a bit and use a lot of filters so that I retain genuinely ones that I can manually verify existed before 1990s. In most cases, it’s a few decades preceding the national shock that happens in 1990s. In many cases, many centuries, because they have a long tradition of having these mutts.

Now once the geotagged locations of all of these mutts are there, I need to combine them with an administrative dataset to claim if they have any impact either on economic outcomes or on public amenities. And here, a wonderful gift for the entire research community in development economics is the dataset called SHRUG.

It is the Socioeconomic High-Resolution Rural-Urban Geographic dataset. The idea here is lots of boundaries keep changing in the Indian census over multiple rounds. They fix these boundaries and essentially make this comparable over time.

So, I build a panel dataset using this. For the census, I use the economic census, and this is not a survey. We have every economic nonfarm activity actually picked up in this.

RAJAGOPALAN: Yes.

SHASTRY: For that, I have four different years that I get, 1990, 1998, 2005, and 2013. Similarly, for population census where I get schooling data, I have 1991, 2001, and 2011. Combining all these three datasets, and of course, the fourth one being the political dataset, which is a bit richer because it goes from 1978 until 2013. I combine all of this to then see areas which have higher density of mutts versus areas which have lower density of mutts. Do they start changing after the national shock of economic liberalization arrives? Just to flag this, this national liberalization shock has got nothing to do with the location of mutts or anything to do with mutts. It’s a truly exogenous shock.

RAJAGOPALAN: Yes, it was a balance-of-payments crisis. We’ve documented this pretty well. I don’t think you—

SHASTRY: Exactly. Absolutely.

RAJAGOPALAN: —have that kind of endogeneity problem.

SHASTRY: I agree. So, when that happens, crucially what happens is the returns to nonfarm activity improve quite a bit.

RAJAGOPALAN: Yes.

SHASTRY: That’s where I feel mutts start to kick in. They’ve created these assets. They have large amounts of land, education institutions, and they also liaison with the government. It’s a routine activity for these heads of these mutts to submit their wish list to the Karnataka government to provision more schools.

RAJAGOPALAN: Yes.

SHASTRY: I find that over time schooling improves after 1991 in areas where mutts are more densely located than not. The first thing that I find in the economic census is that quite a bit of manufacturing goes up.

Interestingly, it happens with the lag. If anything, in 1998, which is after the reforms have happened, my control and treatment, or rather, the high-density mutts actually have lower nonfarm employment than areas with lesser mutt density. But over time, the areas which are exposed to higher mutt density really improve quite a bit. My sense is that it comes from the education institutions going up. It, in fact, doubles over time per capita.

This mechanism, I think, has some bite also because this is how some of the other people who have documented Christian missionaries’ work also notice the same. Also, very similar gap as well in terms of—it takes about 10 years for these schools to come up, and then you notice the improvement in the economic census outcomes. So, that’s the larger structural transformation, which I think the mutts end up having some kind of impact.

Mutt Autonomy and Adjustment to Market Demands

RAJAGOPALAN: The mechanism that you find is twofold. One mechanism is just they have been placed under charitable institutions and trust as opposed to religious institutions. So they have greater autonomy, which means they’re able to deploy their capital much better, both their human capital and their physical capital and assets.

A second layer of that is very specific to education. They decide to invest in education services. This is not just charitable schools for poor kids. This is our typical understanding of what some of these religious institutions run. They’ll feed the poor, they will provide cataract operations for poor people, and they’ll also start a primary or a secondary school for young girls from poor families or something. These are actually colleges that then turn into proper engineering colleges. People are getting computer science degrees at these places, right?

So, is that the appropriate mechanism, that because they are able to be more autonomous in their deployment of capital, they can actually adjust to the demands? And being that close to Bangalore and post-1991, given that the biggest demand is for engineers in that area, that’s the direction in which they go. Or is this story just too simple and too convenient?

SHASTRY: I would like to agree with you largely. I just do want to add a couple of clarifications. In the analysis, I leave out urban areas and specifically Bangalore, because it’s so easy to get confounded with the presence of large cities. And as you know, service sector tends to agglomerate quite a bit around Bangalore.

RAJAGOPALAN: Yes.

SHASTRY: When I break down the results that I have in the economic census, a lot of this actually turns up in manufacturing. So, you’re absolutely right; it’s not just schools, but also, people who go to colleges, senior, secondary schools, and so on. So, they end up enrolling themselves in manufacturing sector.

What happens, interestingly, just to rule this out, is there is no uptake or severe difference in the government jobs. So, it’s not a simple story of these governments just want to appease mutts and they end up provisioning more government jobs. That doesn’t happen.

It so happens that it shows up in manufacturing, which makes me think—unfortunately, my data can’t fully prove this—is that there are other things beyond also just the education story.

It’s also that these mutts become, in a way, a congregation of information networks. There is anecdotal evidence of credit lending happening through these organizations.

Some of the largest nonbank financial organizations are actually run by these mutts. And it’s not surprising, because information about credibility becomes common knowledge through these organizations because people know each other quite well. And the other aspect is, of course, that in parts of these places, they actually end up dispute-adjudicating. A crucial question of contract enforcement, for instance. That may become simpler because mutts end up coordinating this kind of action.

Again, I find this very interesting, and I think they will add more complexity to the work. It’s much harder to say that with the data that I have right now. I suppose a follow-up work of some kind of survey of where people are getting this kind of credit will actually add a lot more depth to this. 

But yes, I agree with you. The headline is that education institutions improve. Of course, we see an uptick in manufacturing, but I do think, just based on anecdotal evidence and possible economic explanation, there are credit networks and information networks also playing a role.

RAJAGOPALAN: Yes, so, if I were to tell the simpler version of that same story, it’s again, whether you tell it through credit networks or through educational institutions, structural transformation requires resources to go to their highest-valued use, and that needs to happen relatively swiftly in a fast-growing economy. Basically, these mutts are able to maneuver that, or at least catalyze that if not directly maneuver it. Is that the better version of the underlying mechanism?

SHASTRY: Absolutely. I would completely agree. And again, just corroborating this thinking, we do have evidence from other parts of India where caste networks really improve the occupational choices. Kaivan Munshi and others, they work quite a bit on it. So, I think this is what’s happening in this case as well.

The Propagation Effect of Mutts

RAJAGOPALAN: OK.One follow-up question: Karnataka is not just the hub in India of mutt-led educational institutions, but it’s also, overall, the hub of educational institutions, right? There’s a long history to this. You could start with South Canara and everything that happened there with their credit networks, T.M.A. Pai and all the educational institutions. But that comes through a slightly different constitutional mechanism, which is linguistic minorities.

So, in India, we like to interfere with all sorts of things. The government can interfere with lots of institutions, but then we carve out these interesting exceptions, and educational institutions that are run by minorities overall, and particularly linguistic minorities, are protected under Article 30 of the Indian Constitution. 

So, you have a lot of these Tulu-speaking—which is technically a minority language—that set up these educational institutions, which incidentally don’t teach Tulu. They are actually teaching engineering or medicine or something else. So, how much of the overall effect of education uptick is coming from that effect versus from the mutts providing education institutions?

SHASTRY: No doubt. I think, specifically, the areas that you have named have been largely driven by probably Konkanis and Tulus. They’ve been pioneers in setting up of education institutions, banks. We have four or five nationalized banks coming from a small district in Karnataka, so it’s quite exceptional. I do think, though, we do have such minorities elsewhere in India.

RAJAGOPALAN: Yes.

SHASTRY: For instance, let’s say Parsis in—

RAJAGOPALAN: Of course.

SHASTRY: —Maharashtra or Gujarat. But the true success, in my opinion, seems to me that the organizational setup, or the blueprint, that the mutt gives helps propagate this effect elsewhere. In fact, even probably the first property right—and you’re an expert on this, the Shirur Mutt case with—happens to be a mutt again. And they have a fantastic history of eight mutts in Udupi, self-governed. Since 1522, they have a tradition of changing who governs the temple. There’s a lot of rule enforcement that happens as a collective rather than run by the state. But overall, this blueprint somehow transported across different parts of the Karnataka state and ensured that, on an average, I think the whole state has these institutions spread out quite a bit.

But I do agree. I think it’s a fascinating study to understand how the Tulus and Konkanis have had their own role. And also, the princely state of Mysore has given quite a bit of patronage to revolutionary thinking for their time period. It could be a study of its own to understand how the British colonial legacy, direct rule versus through these kinds of princely states ends up generating these heterogeneous outcomes. But long story, my claim, I think, is that the organizational design and subsequent autonomy granted to these mutts made it more imitable, and it transports across the state.

Parsing Effects

RAJAGOPALAN: So, the economist version of my question would be, are you able to isolate districts which have greater proximity or exposure to these, say, Konkani- or Tulu-set-up educational institutions, versus mutt-set-up educational institutions and see any difference? Or are they all just clustered together in a way that it’s very difficult to parse out?

SHASTRY: Yes. A great question. I think I’m a bit handicapped by the census data that we have from Indian government. In 2011, the best distinction we can make is whether an educational institution was private or if it was government-run. Going back 2001 or 1991, we can’t even say as much. So, to set these institutions apart, utmost we can do is geographical separation, but within a subdistrict or a village, it’s very hard to say what, actually, that is because we need a panel of data before the shock and after.

It’s much harder to separate these things out. What I can say, though, is, on an average, this is what happens. Underneath, there could be many mechanisms, including what you said, working beneath the headline number.

RAJAGOPALAN: Yes. What is a good way of thinking about this nonfarm employment result? Should we think about it as just broader structural transformation after 1991, or should we think about education playing a very specific role? What I’m asking is, do we see evidence of schools that train mechanics or electricians or a particular kind of industrial work as opposed to schools that may provide the more general K–12 education and so on?

SHASTRY: Absolutely. I think, again, having some kind of data on trade schools might help quite a bit. At this point, I have to rely on anecdotal evidence to say that this could have, or evidence given by ethnographers and anthropologists. I would very much encourage people who are listening, young researchers, also to do more on this because it’s just a handful of people who work on this space.

Right now, because of the census data that we have, it’s hard to say whether it’s driven by trade schools or not, just an aggregate number of what the education per capita overall looks like.

Largely, I think one extension that I am thinking about is trying to combine with other datasets that we have to see specifically what types of industry actually are driving this beyond just manufacturing, services, or government jobs. But it’s a nightmare of its own in terms of combining all of this data together to make it comparable.

Mutts' Impact on Political Competition

RAJAGOPALAN: Yes, and also, going back 40 years, I think, is the even harder part. I think more recently, you have slightly better data.

Now, in the end, we’ve talked a fair bit about the structural transformation and the education piece of the puzzle. You look at one more major area within this paper, which is looking at what greater exposure to these mutt institutions does to political competition. Does it increase political competition or decrease? And I imagine that your fundamental mechanism is that they have very, very strong identity formation in these areas. And a lot of the caste and religious and linguistic identity also plays into political competition. They might be able to broker, in some sense, what happens in that political marketplace. So, what do you find and how do you actually study this kind of question to figure out if there’s greater or lesser political competition?

SHASTRY: Yes.I think I was fascinated by this question myself because this usually tends to dominate the news cycle around mutts, specifically the political involvement that they end up taking. What I found out actually did surprise me. The mutts’ areas don’t really end up supporting just one party. It depends quite a bit on election cycle to election cycle. They are pretty much on the swing margins, which makes them even more important to be appeased.

I think they’re well aware of this fact, and hence they use it to the fullest. They have a fantastic bargaining chip in terms of their role. In the paper that I show, it’s less than a percentage point, but if you notice, Indian assembly elections have very small margins.

RAJAGOPALAN: That’s a lot. Yes, yes. [chuckles] That’s a nontrivial number.

SHASTRY: Exactly. These mutts, it’s very precisely estimated as well. Mutts have a unique role in making the electoral politics more competitive. It could be to meet their own end goals of any kind that they have.But inevitably they end up making it more competitive. I think this is the key mechanism that one can say that the electoral democracy becomes more accountable in a way to the people through mutts because it’s, of course, much harder to talk to individual constituents, and mutts end up aggregating this kind of information and preferences over time. One, theoretically, could have gone either way, but what I find interesting is the post-’90s phenomena of expansion of mutts, and there is clear government patronage to some of these mutts.

For instance, many governments, including the BJP government, which first, Karnataka came in 2008, explicitly started making budgetary allocation, for instance, during flood recovery and so on. It’s a very complex political economy between the state and the mutt system entirely. And again, it goes beyond any particular party. All of them try to appease them, and their support sometimes makes a big difference in the electoral outcomes in this case.

The Intersection of Mutt's "Club Goods" and Political Influence

RAJAGOPALAN: But this is such a lovely result in the sense that what are our priors when we just watch the news or read the news? It is that if there is a wave of majoritarianism, or in this case Hindu nationalism, you would imagine these mutts brokering that particular side, or there’s going to be a wave, and you would imagine there’s going to be less political competition in these areas because they are basically byproducts of the majority religion. The more you have a political Hindu nationalist wave, the less competition we should see.

What you find is, in fact, exactly the opposite, right? Even more specifically within that result, the resources are scarce, and political allocation of resources is not going directly to the mutt, but it’s going towards announcing a new factory or flood relief or setting up a new college or putting in a new port or something like that. Which means that the original point where we started, which is they broker a particular kind of club good or philanthropic, collective-action good, is also now interfacing with the politics of Karnataka.

SHASTRY: Absolutely, and I completely agree with the characterization that you’ve made. Again, the key factor— 

RAJAGOPALAN: Well, I would hope so because I got it from your paper. [chuckles]

SHASTRY: [laughs] An interesting counterfactual comes, or rather a comparable result comes from Indonesia, for instance, where you had extensive land reforms in the 1960s. But the waqf properties end up being exempt from that. So, large amount of land ends up getting deposited in those places, and mosques get built there, eventually explains the role of Islamism overall in Indonesia. It’s hard to have a unilateral understanding of mutts that way.

They tend to be a lot more dynamic, and perhaps this also speaks to the results where Timur Kuran says that ossification of any kind of assets under religious organizations end up making them less relevant. But mutts are, as you characterized, quite different.

They will probably go and do these things with whomsoever offers them a good deal for their constituents and where they’re located. It so happens, a great thing, that they care about education, they care about hospitals, they care about factories being located near where they are. The crucial start to understanding these Hindu institutions, at least in case of Hindu mutts, is they tend to want and care for secular education and secular goods rather than them being the substitute for secular goods overall.

RAJAGOPALAN: AndI think here you’re also quite clearly showing that the political representatives are, to some extent, in competition with mutt representatives. They are both in some sense competing for influence and patronage, and the mutts seem to be better at delivering what the local community wants, and that’s the direct point of competition. So, either the political patronage has to align with the mutt or take their chances and maybe have a close election or lose the election.

SHASTRY: Absolutely, and I think the only other work that I have seen in India in recent days—I think Kartikeya was also at your podcast. Avantika and Kartikeya put together this dataset on deras in Punjab.

RAJAGOPALAN: Yes.

SHASTRY: They show similar outcomes where all politicians end up taking the blessings of these dera gurus and people who head deras, and despite deras being a much smaller entity compared to a Karnataka mutt, and they had significant—

RAJAGOPALAN: And also newer—

SHASTRY: And newer as well.

RAJAGOPALAN: —less sticky norms. Karnataka mutts have 400-year traditions, in some cases, of how they’re going to appoint the next person and so on. Some of the deras play a little fast and loose with their institutional accountability.

SHASTRY: Exactly. Compared to that, I think, in this case, it happens to be even more credible, even more dynamic, and a lot more politically embedded than deras. You’re right, I think a lot of political leaders do end up taking the blessings of these mutts, and it’s often the kind of news you end up seeing in the news cycle is certain leaders of these mutts, they end up lobbying for a candidate to get a ministerial berth. That’s the kind of news that gets dominated, but behind it there is a large “state-ness” of the whole mutt system complex, and it’s quite hard to, in a way, veto their preferences because you never know, the next elections can end up being counterproductive to the party.

RAJAGOPALAN: Yes. You know, my favorite anecdote on this is when the Ayodhya temple was built and the consecration and the puja happened, there were actually a lot of Hindu mutt leaders—and not the small ones, all the major ones which have had a thousand-year history—who were quite unhappy because they said that what is basically a religious ceremony and a sacred ceremony has been completely politicized. And there’s going to be a political winner, so to speak, who is actually going to preside over this—what should be a religious ceremony. So, you can see that sense of competition between political influence and religious influence even at that highest level when you would, as a prior, imagine that their incentives are largely aligned.

SHASTRY: Absolutely. One key difference that I have noticed, and I could be wrong in this, is the mutt leaders in Karnataka seldom themselves get into politics as opposed to, say, the larger North Indian politics where you often see religious leaders becoming themselves political candidates.

RAJAGOPALAN: Yes, like chief minister of Uttar Pradesh.

SHASTRY: Yes, indeed. I think that is quite important for them to reinforce their credibility over time, and the more worldly somebody seems, the harder it is for them to run this large institution.That’s been one of the exceptional results that I, at least anecdotally, noticed in Karnataka.

RAJAGOPALAN: Yes. No, this was such a pleasure to read your paper. It was absolutely fascinating. I feel like I don’t get to read about economics of religion in India, which is such a shame because it’s the most fantastic petri dish to study pretty much every kind of religion, religious sect, religious practice, and so on. So, it was fascinating to read this. Thank you so much for coming on the show and sharing this with us.

SHASTRY: Thank you so much, Shruti. It’s a tough field to be in, specifically working on economics of religion, but I think we’ll be getting there step by step. Thanks a lot for having me.

RAJAGOPALAN: Ideas of India is produced by the Mercatus Center at George Mason University. If you enjoy this podcast, please subscribe on Apple Podcasts, Spotify, or your favorite podcast app. Help us grow by giving us a rating and leaving a review. Follow us on X @srajagopalan and @Ideasofindia. Also, check out our initiative commemorating 30 years or India’s market reforms at the1991project.com.