Nachiket Mor on Rethinking India’s Healthcare System

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

Today my guest is Nachiket Mor, a health economist whose work focuses on the design of national and regional health systems. He is a visiting scientist at the Banyan Academy of Leadership in Mental Health, a senior research fellow at the Centre for Information Technology and Public Policy at IIIT Bangalore, and a commissioner and author on the Lancet Citizens' Commission on Reimagining India's Health System, which published its final report in The Lancet in January 2026.

We talked about the different layers of the Indian healthcare system, the design and policy failures in both public and private sector healthcare, the role of community workers, the health insurance and regulation market, 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, Nachiket, welcome to Ideas of India. It is such a pleasure to have you here.

NACHIKET MOR: No, I’m very happy to be here. Thank you for inviting me to talk to you and discuss some of these ideas.

Policy Design Failure in India’s Healthcare System

RAJAGOPALAN: Yes, I have been reading a range of your papers. You have almost had like two careers: one in more traditional finance and now one in more traditional health policy. If I had to try to summarize your very wide-ranging work, which I encourage everyone to read and not just go by my summary, it would be that India’s health crisis is not just a funding failure but fundamentally a policy design failure. 

We’ve built a tertiary care infrastructure for a few people while neglecting primary care for the masses, for almost everyone. Our insurance model and financing models basically reward hospitalization over health. What ends up happening is people have massive out-of-pocket expenses. Healthcare does not just become a question of health policy. It is also a risk management issue at a household level. You’re skeptical that the current private sector system can deliver value the way it’s currently set up. You are also skeptical that the government can ever fund healthcare adequately, such that it is a universal healthcare system in India. 

Your argument is really that we need to redesign the system in a way that all these different pieces of the puzzle - which is insurance, financing, state-provided healthcare, private healthcare, primary, tertiary healthcare - need to be redesigned to work such that people get more value. That puts you a little bit at odds with the public health left who want more funding, tax finance, universal care, and also the market liberals who think competition and insurance ought to fix everything. Is that a good way to think about where you are in this particular literature or this conversation? 

MOR: I would say yes and no. To some degree, what you said is completely spot on as to how I’m thinking about it. In some degree, not quite. First, I have a paper in The Lancet Southeast Asia that came out in 2023, in which we looked at how much money is actually needed to deliver universal health coverage and what I like to call global health for all or good health for all, because coverage is not the same as good health, as we have discovered in some states.

What we were amazed by is that Indian costs and talent are such that it’s inappropriate to use the nominal exchange rate or even the PPP-IMF exchange rate, but a more appropriate exchange rate for rupee-dollar conversion is more like 5 rupees to the dollar. At that 5-rupee conversion, then the Indian expenditure of, say, 5,000 rupees, now suddenly $1,000 per capita, which puts us squarely in the middle-income country range of Thailand, Brazil, some of these places, and it’s more than enough money to deliver universal health coverage. 

I certainly believe that the public sector is adequately funded in many states. Most states by 2030 will be adequately funded from their government budgets. There are four states that still need extra money, but the center can provide that money relatively easily, which means that this idea that funding is a gap—many people talk about 3% of GDP, it’s not entirely obvious where this number came from, and certainly Indian costs are such that even with a number that is much lower - we are able to deliver. 

Then, of course, you might ask, which is the point you were making, is, okay, if the money is there, why don’t we have universal health coverage, say, in a state like Kerala, say, in a state like Goa, in the high-capacity states? This is where your point about it being a design problem, not a funding problem, comes up. In some ways, what we have done is created a system in which 1920s Russia features in our public sector, and 1960s America features in our private sector, so ideas from the Kenneth Arrow era, Medicaid, Medicare, when they were signed into law, is what reflects in our current private sector design. 

A lot of the work that happened in the ’90s with Mrs. Thatcher, with Thailand, with Turkey, with what happened in the European markets for private healthcare, that entire set of reforms, we seem to have missed. It’s not even a conversation we are having here.  

Layers of Indian Healthcare

RAJAGOPALAN: For those who are uninitiated with how Indian healthcare works, except they might have themselves gotten sick and had some kind of interface with it, how would you describe the different layers of how the healthcare system actually works?

MOR: The Indian public sector, one of the reasons why I continue to be a votary of the public sector, is it’s the only complete health system we have, which means it has, if I use check boxes, I can check off community health worker, I can check off what we call a subhealth center, which is a primary care center inside a village, and then we have what we call a primary care center, which is at a slightly larger location. Then we have hospitals, both at the secondary level and the tertiary level. This becomes a comprehensive machinery. 

Some states like Tamil Nadu, for example, you can have an additional layer of a public health cadre, so there is a group of health professionals that are only thinking about cholera and water and that kind of stuff. It’s a complete health system from end to end. That does not mean it’s well integrated. If you go to a primary healthcare center just outside a government hospital and you were treated for something and you went to the government hospital because now you needed a hospital, they wouldn’t know anything about the previous encounter. 

In some ways, there is fragmentation built into the public sector, even though it has all the links in the chain. Outside the government, we are in a little bit of a no-man’s land. The private sector is very distributed. More than 95% of providers have fewer than five employees. Often, it’s just one person doing something. We do have large tertiary care health systems, but they account for actually only 3% or 4% of the total healthcare spend in the country. If you do have small hospitals, they are what I would call the mom-and-pop nursing homes that you see. 

Somewhere near 60%, 70% of all healthcare is provided by this fragmented machinery. Other than delivering babies - which 70% of the babies are delivered by the public sector in India - all other forms of care, the dominant provider is the private sector and not the public sector. Even though the public sector could provide those services, but it’s actually not well set up to do so. 

RAJAGOPALAN: You have two parallelly running critiques of the public sector, which indicate both a design problem and a state capacity and therefore a delivery problem. One argument you’ve made is there just isn’t enough depth and enough capacity for primary healthcare systems. That is, we’re not nipping the problem in the bud, at least within the public healthcare system. 

People maybe end up going to the private healthcare providers or end up going to quacks, or end up going to alternative medicine, but there isn’t a general physician or a point of contact where people go, their records are maintained, their health is tracked in some way. Even though we have this end-to-end public sector health system, that first point seems to be the point where the maximum fragmentation or the damage happens. Is that a good way to think about your public sector critique?

MOR: I think that again varies state to state. If you went to Kerala, for example, you would find all the functionaries that you mentioned are very much present there. But you might say, “Well, then how come they have some of the highest diabetes burden? How come they have some of the highest hypertension burden? If indeed the primary care was so good, why do we not have the outcomes that we need?” 

That’s where I think one of the design flaws comes in, which I think American health systems have, Indian health systems have, because these were designed with the belief that people are the best custodians of their health. Even the idea of universal health coverage implicitly is an idea that says people know what they want, all we need to do is be there. 

I have a paper on the evolution of the community health worker, in which I look at the various stages of where the community health worker has gone, they argue, and if you see now the evidence from around the world, that argument is finding a lot of empirical support, that if you just build it, they won’t come. 

You have to go out there, build a cohort of high-risk people, and go to them to ensure that they take care of their health. Physicians think of prescription as the end of their role. What we now know is, in a primary care setting, actually diagnosis and prescription is becoming easier and easier. The big problem is compliance. Are the hypertensives taking their medicines? Are they aware they’re hypertensive? Are the diabetics taking their medicine? Are they aware that they are diabetics? Who’s responsible for making sure this happens? 

If you see the best examples around the world, in the US, the Alaskan health system has this character, for example. In the developing world, Iran has this character. Costa Rica has this character. To some degree, Brazil has this character with their family health teams. Turkey has this character. Thailand has this character. We have built passive health systems. Again, it’s a product of an earlier thinking that people know best, and all we need to do is to simply be there. 

In fact, that’s one of the reasons I worry about even the phrase universal health coverage. We, for example, in our Lancet report, find that Kerala is nearly 100% in the universal health coverage index, but it has one of the highest burdens of primary care modifiable diseases.

ASHA Workers

RAJAGOPALAN: In India, there is at least one subsection where we do have workers who go to the patients, and this is basically ASHA workers, and we also have maternal healthcare and midwife programs and so on. Do you see that in those areas India tends to do better? I believe not really. Even in those sectors where we do have something in addition to the overall infrastructure of the public health system, where people are actually going door to door and going directly to the patient, we don’t seem to deliver lower rates of maternal mortality and so on. 

What would be the gap, or am I just thinking about this the wrong way? It just needs to be more holistic, and you can’t just think of one tiny sector of people who are demanding healthcare maybe once or twice in their life when they’re pregnant.

MOR: I think the challenge for us has been, and this may be something we inherited from an earlier design elsewhere, we are not sufficiently ambitious about what our health workers can do. In my paper on the evolution of the health worker, I classify them as stage one, being the messenger. This is what the ASHA is. Whereas if I look at the Ethiopian health worker, she’s able to do an emergency obstetric surgery. If I look at the health worker in a place called Gadchiroli in Maharashtra, she’s able to inject gentamicin in the hands of a newborn. 

We have not built that aspiration into our thinking over the health worker. We have built what I would call a low-low equilibrium. They are part-time, we pay them contractually, and they do what they are paid to do. Even if I look at maternal and child health, just take that narrow statistic, 2015, ’16 data—I don’t have more recent data that I have analyzed—shows that out of every 100 women that were pregnant that was supposed to be served by an ASHA, only 3.5% were actually served by them. 

We have now layered program after program on this poor woman’s shoulder without paying her, without training her, without equipping her. The health worker remains key to this task. As you point out, we do have a tradition of this person going to homes. The argument my evolution paper makes is can we evolve her into a stage-four worker who looks much more like the Iranian behvarz or the Alaskan community health aide, who behaves like a mini physician and is able to do many, many things under the guidance of a physician. 

She’s not running loose doing whatever she feels like, but she’s trained to do a lot more work than the ASHA is trained to do. I think that is a key gap because the ASHA is almost everywhere across the country. That model is universal. Some states have done better, some states haven’t done so well. I don’t know if you’ve seen this movie, The Matrix, in which there is the character Trinity who needs to learn how to fly the helicopter, and the control downloads the program onto her. It’s a good example of what we will call just-in-time learning.

Now, if we think of the ASHA worker as Trinity, and well-equipped with ultrasound, with multiple devices, and she’s going out there, pulling out her various tools, we can do better than the West is doing. What we can do with an expensive GP can’t do is I can sit at somebody’s home and say, “Sir, you were supposed to have taken this medicine at 9:00 in the morning. It’s 10 o’clock now. You have not taken it. Can I please watch you take it?” If you’ve absconded and gone to a restaurant because you knew I was coming, well, I will show up at the restaurant. That kind of follow-up, combined with technology, I think we can deliver better outcomes than other health systems, even developed health systems can. 

RAJAGOPALAN: Just to stay with the ASHA worker and the maternal health example, you’ve also made other critiques in a different context, which is not about community health, but like the incentives of hospitals and the private sector more generally, and also the proliferation of nursing homes. We have an ASHA worker. We treat them sort of as support and almost like part-time staff. We don’t think of them as serious expertise, even though they’re the first point of contact, which means they are not equipped to do anything when there is a sudden emergency, and a birth situation is one of those situations, which means now the nudge and the push for everyone is to get C-sections. India has one of the highest rates of C-sections. 

I had recorded this episode with Janhavi Nilekani, whose work you also know well, about how the C-section rates are just off the charts and completely unnecessary, and actually exposing the women to more risk both at the time and later in life. Is there a better way of integrating the ASHA worker? You mentioned ultrasound machines, but we don’t allow them to have ultrasound machines, which is for a completely different central planning reason of not being able to detect the gender of the fetus and so on and so forth. If you can just give us the example of this microcosm, how would you rethink that so that all these weird incentives start lining up properly? 

MOR: I’ve spent some time researching this question a little bit. A lot of my papers are in review, so they’re not out yet. Remember, we have two distinct health systems. One is the public sector, one is the private sector. As far as babies is concerned, the good news is other than in the state of Kerala, where that number is 30%, nationwide, 70% of babies are born in the public sector. 

If we worry about C-sections, we would want to worry much more about it in the public sector because small changes in C-section rates there can have a far bigger impact than what the private sector is doing. The private sector, in any case, it is an urban, upper-income family that is largely seeking out care. Now, the same question that you raised comes up. Why is it that, for example, the state of Telangana, some of the districts have 60% C-section rate in the public sector? 

They don’t have the incentives of additional income. Then there’s another motive called the convenience motive in the sense I don’t want to be woken up in the middle of the night to deliver a baby, but that’s not how the public sector works. It’s a shift basis. I’m done with my eight o’clock shift. I go home. The next person comes on and they do whatever needs to be done. I’m not going to be woken up. Why are they doing such a high number of C-sections? My research is showing that what we have done is somewhere taken the whole accountability narrative too far. 

I have a post that I just put out in which what I say is that in order to change incentives of the public sector, we have to use trust a lot more as a tool than fear. Radhika Gore, she’s an academic at New York University. She has a nice series of papers in which she looked at the behavior of the public sector provider in the city of Pune. It is an observational study. She sat for weeks and months inside these hospitals just observing what was going on, and after a while, people became their natural selves in front of her. 

What she found is that because of this heightened, the BDI as it were looking at the public sector, the public sector employee has withdrawn into a shell. She has a lovely phrase called ensuring the ordinary. Why are they doing so many C-sections? Because others are doing so many C-sections. What they say is why should I stand out? I know this is wrong. I know I should do less. I did a survey of obstetricians asking them, “Do you need more training? Do you need more expertise in something?” The overwhelming group is telling me that when something goes wrong, we need to be protected from our internal management because they throw us on the street to fend for ourselves if something goes wrong. Then why should I use my medical judgment? 

That’s clearly an incentive. It’s an awkward incentive. If you see our Lancet commission report, what we have suggested is this is where the Thai reforms, the Turkish reforms, the NHS reforms become much more interesting because what they do is they go out there and say to the provider, “I’m not going to micromanage what specifically you’re going to do. I’m going to let your local team do that. I’m going to hold all of you accountable for longer-term outcomes in the community. 

You figure out what is the right balance of C-sections and this, that, and the other that makes sense for you. If something goes wrong, we will look at systemic issues rather than hold you individually accountable for what is going on.” This has been the experience of people who have built a new purchasing character. Clearly, this is something that needs to be experimented with and understood in the Indian context to see if this will work or not. I think that is where the public sector C-section rates are coming from. 

State Capacity

RAJAGOPALAN: If I had to zoom out from this specific example, and maybe it’s an awkward example to zoom out from, your position is always state capacity is not infinite. We need to think about using that state capacity well. If India is running an end-to-end public sector system, what you’re saying is what the government can really do best at the end of the day is also set standards and protocols in place, and actually make sure those standards and protocols are ensured. 

Is there a fundamental incentive alignment problem built in there that if they’re running everything end-to-end, then those standards and protocols are going to get a little bit messed up? There’s only that much capacity overall in the healthcare sector, they can’t do everything. Should they then be focusing on the standards and protocols, and not so much maybe on running certain types of services? 

MOR: The good thing about India being somewhat behind the world is there are many countries we can learn from. What has been the universal solution that has now become the consensus in the world is what they call the internal market solution. Where the government purchases from government on an arm’s length basis. India already has something called the National Health Authority and the State Health Authority. These are autonomous, independent bodies. Currently, they are being used to purchase care from the private sector, from various things. 

If you look at Thailand, they created the National Health Security Office. They asked the health department to choose: Do you want to be a provider or a policymaker? If you wish to be a provider, then we will take away purchasing from you. Now you’ll get paid only if you do the work that was assigned to you. Now, of course, there are downsides of approaches like this. For example, if you look at the NHS, if you look at it from the outside, people like me are quite pleased with the results we see, but if you talk to insiders, they will complain about the long queues, how much time it takes to get a knee replacement done. This is an organized healthcare system. 

Unfortunately, they have compounded the problem by somewhat underfunding. It’s not that they have the underfunding problem that the US has. They need to change funding by 1% or 2%, not a big change. For a variety of reasons, they have not done that, which is why the queues tend to get much longer than people are prepared to bear. In some ways, they are running a system now in which if you go to a public sector facility, you get very good care. There isn’t really the concern. This is what we find in Thailand, this is what you find in Brazil, this is what you find in Turkey. 

The Exit to the Private Sector

RAJAGOPALAN: In India, this is a concern because outside of, as you mentioned, childbirths, there has been a relatively large exit from the public sector to the private sector. What is a good way of modeling that? Is it just there are parts of India where we don’t have good public sector care? Is it that because that was the initial point problem, there has been an exit from that? It’s now an adverse selection problem, both on the supply and the demand side. 

That is, people who want healthcare will just exit to maybe not a great private nursing home. Also, people who are supplying healthcare, doctors are not going to go work in those hospitals because they’re like, “Yes, that’s a remote place. No one comes to that hospital anyway. What would I learn or gain from serving in that place?” That seems to be what it appears to be when I just look at it from the outside. What is it like on the inside? Again, I’ll rely on you to tell us the variation between states if there is much variation on this point. 

MOR: The reality on the ground is that government hospitals, every state are flooded with patients. There are no empty government hospitals. They are full. The poorer the state, the more the dependence on the government because if the government can’t make it work somewhere, the private sector is certainly not going to come there. The private sector is present in areas where the government is already successful. 

In some ways, if I take a state like Chhattisgarh, if I take a state like Bihar or Uttar Pradesh, C-section rates there, just to give you a sense of contrast, 0.2%, 1%, 1.5% because neither the public sector nor the private sector are present. You expect to see a reasonable rate of say 25-odd percent to say, some people get a choice if they really want a C-section, but by and large, most people get a normal delivery. Clearly there, I’m not expecting for a long time the private sector to try and make an entry. That will need the public sector to step up, and it’s not as if it can’t be done. 

If I look at the state of Andhra Pradesh, now we have two states, Telangana and Andhra Pradesh. Earlier, they were all one state. Telangana is where Hyderabad city is richer, more urban. Andhra Pradesh is actually not so rich. It is a poorer state. Yet it has very good healthcare in the public sector. The designers of that health system realized that if you take the health resources of a state and spread it like a thin layer of jam across the state, you will get no critical mass anywhere. As you correctly pointed out, if I’m an anesthetist in a hospital where there’s no surgeon, what exactly am I going to do? 

What they did is they built hospitals, staffed them fully, and districts are reasonably large. It’s not as if I’m sending somebody to the boondocks. These are cities, mini cities with reasonable infrastructure. If you staff it well, you have the resources to give housing, give other things, and then you move to the next hospital once you’ve completed the job here. 

In a northern state, you might find 80% vacancy of specialists. In Andhra, it’s 3%. They have built out a health system with patients, with design. There are people like Dr. P. V. Ramesh, who was a doctor himself. He then became health secretary, then finance secretary, and then chief secretary. I’m sure there were many others, I’m just giving you one name that I’m familiar with. 

RAJAGOPALAN: He’s the person who really was the steward of a lot of these mechanisms.

MOR: Indeed, and that lesson is available for free to anybody. In fact, Dr. Ramesh is happy to travel and talk about it to other people. You have to have a sense of, “I’m going to design this to make it work.” You can’t say, “I will just be mechanical about it.” Then you will not end up with anything. This is a part of the problem in many of our poorer states where we have spread this thing like icing on a cake without the depth that is needed to make each of these systems actually work. 

The other thing, I’ll give you an example. I went to one of the large hospitals in Bombay, and I met a-- When she just got the Padma Bhushan, Dr. Fernandez, Armida Fernandez, internationally educated, highly talented, suave, well spoken. She was the dean of this government hospital. I must say I was a little taken aback because I expected somebody like her to be a senior person in a private facility. She said to me, “As a professional, if I want to see the range of diseases that I trained for, this is the place. I get the additional satisfaction of helping people that are poor and this and that, but don’t forget, as a professional, I gain a great deal.” 

She set up the first human milk bank in Asia. For example, she recognized that the aunties that show up in India, which many hospitals consider a nuisance because there’s no place for them to sit, she converted them through a short training program into neonatal nurses so that they would look after the baby, the child. There was a premature child. She brought premature mortality from 95% to 5% using these techniques. These are opportunities she would not have been afforded in a private facility. 

RAJAGOPALAN: I would be the last to defend government facilities in, say, states like Bihar. Isn’t one fundamental difference between, say, certain districts of Bihar and certain districts of Andhra Pradesh, is that certain districts of Bihar are at GDP per capita lower than Sierra Leone at this point? That affects everything across the board. It affects nutrition. It also affects infrastructure, which means you’re more likely to have a fall and maybe break an ankle. You’re more likely to have a snake bite. The range of problems increases. The primary healthcare capacity is lower for the same reason that it is poor. 

Now, having one big district hospital is just going to overload the system. You’re never going to get exactly what you get in Andhra Pradesh, which is many times richer, has the capacity; even at rural levels, people are rich enough to go to a private physician, even at the very early stages of intervention, to a primary healthcare provider who’s private. Isn’t that somehow just the bigger problem in states like Bihar and UP and Madhya Pradesh, Rajasthan, which are our big four problem states for healthcare in some sense? 

Getting Ambitious with ASHA Workers

MOR: Yes and no. Depends on your design. If your design is drawn from the British and you want a GP and that’s what you think is primary care, you’re quite right. If you say, “No, that’s not the design I have in mind. I have the Alaskan design. I have the Iranian design in mind, in which I have a highly trained health worker, not the ASHA, but multiple levels above the ASHA.” 

RAJAGOPALAN: One step above.

MOR: She’s a local. Such people are available aplenty. There is a project that I’m associated with as an adviser in rural Satara, remote part, difficult part of Maharashtra. Not quite as poor as Bihar, but still quite poor. Local women are available aplenty, because they need work and they can be trained, but we have to get more ambitious about them.

RAJAGOPALAN: Karthik Muralidharan writes exactly about this, how we are not tapping the ASHA workers, selecting from within the ASHA workers to then promote them to the next layer of healthcare providers. I guess you are hinting at the nurse practitioner model that we also have in other parts of the world. Before the GP, you can have that. 

MOR: I think in India, the word nurse is a legally loaded word. There’s a nursing council, et cetera. The nurse practitioner, in some ways, is still a clinical worker. Whereas what I have in mind is somebody that is a field-based public health [worker]. Maybe a public health nurse is a closer proxy for what I’m talking about. Now, these people, and then the other thing is, because they have low GDP per capita, they are hungrier for work than if I started to try and do this in Bangalore, where somebody might say, “Unless you pay me something else, I will not show up.” There’s an opportunity there. 

Second, it is quite feasible for me to imagine a situation—because the government already is sitting on over $100 billion of cash—that there is no road, maybe. But actually it turns out to just another interesting piece of data. If you look at the rural road network of India, a per square kilometer basis, even in a state like Bihar— 

RAJAGOPALAN: It’s improved.

MOR: —it’s better than the US. No, it’s not just improved. The only country ahead of us is France. We are better than China because Prime Minister Vajpayee launched something that resulted in this situation. We have good roads. We have enough money to put optic fiber all the way into every single village. Now you have the beginnings of a good primary care system that is local. One is income. There are many areas of India where there are hostile situations, where there is violence, there is other problems going on, but see, a local has no choice but to be local. 

If you can provide her the backend support, this is what the Alaskans do. Alaska’s community health aide sits in a remote village, is trained centrally, but goes back to her village, lives there, and when there’s no connectivity, there is sheet ice, aircrafts can’t take off, internet has collapsed, she continues to operate. 

RAJAGOPALAN: Okay. You’ve actually written very explicitly about this, about how just putting more family physicians in a rural area and having one doctor, one veterinarian, that old colonial model you were talking about, is not the appropriate blueprint. Now you’ve talked about community health workers so far quite a bit in this conversation, but you’ve also written about how we can do pharmacy-linked primary care. Normally, the stack you have is like an urban system where you have a really nice, swanky pharmacy where you can also get your flu shot and other things, but you mean it a little bit differently as the first point of contact. Then you’re also talking about layering that with telemedicine.

Stacking Healthcare

What would that stack look like in some sense? If you can walk us through, imagine maybe the poorest districts of India, maybe not conflict districts, but places that don’t have this kind of primary care support, how would you think about the community worker, the community public health worker, and private health worker layered with pharmacy primary care and then layered with telemedicine? 

MOR: There are many, fortunately, working models. The inspiration for this work, I would say, in a developed country context is France. France was one of the earliest to realize that they have one pharmacy for 2,000 population. India has four times the number of pharmacies. My understanding from research, it’s not my own research, other people’s research from Odisha and other places, is that almost every human being, there will be some pockets where this is not true, is not more than an hour away from a reasonably well-equipped pharmacy. 

Now, the pharmacy means what? You have a refrigerator, you have injections, you have a range of medicines, you have the supply chain that is figuring out and working very well for you. Now, can this person become a channel for diabetes education? Government has a diabetes educator program. Can he or she become that person? Can he or she, using her mobile phone or his mobile phone or his computer, become a telemedicine link? 

There is a company in India—and I have a paper on it that I wrote in the Stanford Social Innovation Review— called Jiyyo, J-I-Y-Y-O, not Reliance JIO, but J-I-Y-Y-O—what they discovered, something quite interesting and something quite different from what I had imagined earlier, that the pharmacy, because they were also a semi-doctor, as it were, the routine coughs and colds, they were doing whatever they felt like. 

The right way to deal with them is not to say, “Oh my God, this is illegal, what are you doing?” It’s to just make sure you keep training them, that whatever they are doing is at least somewhat evidence-based. What they discovered was that whenever a complication emerges, the pharmacy is desperate to get advice. One of the largest areas of work is mental illness, for example, rheumatoid arthritis, gastric problems, cancer. This is where the telemedicine secondary consult, a specialist consult becomes quite important because now the pharmacy acts as a channel.

It’s fascinating to see the conversation because the pharmacist knows the medicines, knows the terms, the patient does not. The two of them are together conversing with the specialist sitting somewhere in Lucknow and figuring out what to do. Now, I’m certainly not going to tell you every pharmacy is appropriate for this kind of environment. 

Can we imagine building a chronic care, NCD care infrastructure like the French have done? If you are in France, you get your vaccine at the pharmacy. You get your blood pressure checked at the pharmacy. You give your blood at the pharmacy to figure out do you have diabetes. I think this is quite replicable in India. I have a paper that looked at multiple models. Mexico has tried this. Bangladesh has tried this multiple times. Forty countries formally allow their pharmacies to be used as primary care locations. This includes South Africa and Nigeria. It includes France. It includes Indonesia. 

It’s a constructive engagement. You go to the pharmacist, and you tell him that, sir, I’m happy to engage with you, but you have to play by the rules now. You can’t be a free agent doing whatever you feel like. If he sees enough income boost. What on-ground interviews show is that the second-generation pharmacists are keen to see themselves as professionals. They are no longer interested in making money in any way. If you give them that opportunity of training, certification, they are quite willing to come on board. 

RAJAGOPALAN: Here, when I read your work, there is the immediate caveat that you add in about unintended consequences. I guess this is the way all economists think. You’ve written separately about how we have a huge problem with overprescribing antibiotics. This is partly overprescribing, but this is partly also pharmacist-driven. People just show up at the pharmacist and they say that I have a sore throat or whatever it is, and they want to get over the problem quickly. Amoxicillin and these things are just being popped like candy at this point in India. There’s no immediate testing. 

Is the model the way you envision it, one that if you become a licensed pharmacist who’s also a primary care center, then we’re going to have certain guardrails and you have to follow them? That’s one model. The other model is we just train you to do a strep throat test then and there, or a flu test then and there. Then you can prescribe accordingly because right now you’re giving out antibiotics like candy because you really don’t have the information. I have this conversation with people in India all the time. When I get sick, I actually go to the doctor, and I don’t pop any pills until they tell me it’s a viral infection or a bacterial infection.

That logic is alien to most people in India. Even people in urban areas, well-educated, they don’t wait for a quick nasal swab or a throat swab before they start popping pills. What is a good way to think about those kinds of unintended consequences, because they can snowball into very large public health problems?

MOR: I think there are two separate issues here, just to make sure we are clear. One is there are two practices of dealing with a sore throat. One is a European practice, one is an American practice. The American practice is strep test. The European practice, which is the Indian practice, wait for three days. It’s an empirical approach. If the fever pops to a certain level, 101 and above, then you come back. Otherwise, stay at home, take rest, drink a lot of water. If you need paracetamol or Tylenol, take that, but no medicines. 

If at the end of three days you’re not better, the strep test is not the Indian recommended protocol, but it could be. I’m just giving you a narrow example here. The issue of overprescription is not necessarily just a pharmacist’s problem. This is a generic issue across the system. There is work by Dr. Jishnu Das, in which he sends standardized patients to trained physicians. Somebody has gone in with a complaint that said, “I woke up in the night, I was sweating heavily, my chest hurt.” He comes out with an antibiotic from a trained physician. 

There is a deeper problem we have of where is evidence-based medicine being practiced? Who is holding anybody accountable? In fact, I think it may be easier, which is why I like the community health worker model over the family GP model. When we did primary care in Thanjavur, for example, what we found is we were using nurses, nurse practitioners in this case. The good thing about nurses is that they knew they did not know. They were much more willing to follow protocol because they said, “We don’t know what we are doing.” 

Whereas somebody who has had the training feels quite comfortable doing whatever they feel like doing. India, one of the consequences of COVID-19 in India was mucormycosis. Why did this arise? Because overprescription by not just pharmacists, by doctors of steroids that were not needed. It’s a deeper, more generic problem as to what are we to do about this stuff. It’s a public health crisis. I’m 100% with you, it’s a public health crisis. 

Now, how would we go about dealing with it? I think it needs a deeper discussion with the medical profession overall, not just pharmacists, as to how are we going to go about addressing this large problem, because we have two problems. In some pockets, like cities, you get overconsumption, but on average, India is still using too few antibiotics. There are too many people dying of infectious disease. They’re not getting the antibiotics they need. We’ve got to deal with both problems. It’s like the C-section problem. Some districts are showing 1%, some showing 60%. 

RAJAGOPALAN: In the case of antibiotics, it’s going to get worse because the people who don’t get antibiotics they’re going to die in larger numbers because the people who are overusing antibiotics are leading the system toward antibiotic-resistant bacteria. You’ll only see this divergence or this polarization get worse in some sense, in terms of numbers. These numbers, very small degree changes can actually lead to relatively large effects. 

Now, when I think about how you write about the private sector, and this is not just primary care, this is all the nursing homes and then moving on to tertiary care, you think about the incentives of the private sector, but not just in terms of market competition the way we would normally think about any other private sector design. It is inbuilt with the financing and the insurance model. What is a good way to approach this now? Should we talk about just the private sector and its incentives and then move on to financing and insurance regulation, or should we club all of that together as we look at India’s private sector? 

MOR: One thing that people like Enthoven at Stanford pointed out, and the Europeans have gone in this direction, Israel has gone in this direction, the Americans are trying to, but have not yet been able to move fully in that direction because of entrenched habits, is that finance and healthcare, insurance and healthcare, are not self-contained products. It is like if I sold you the front of a car and I sold you the back of a car and I told you put it together, it is not a complete product. 

As you go into theory and look at the first welfare theorem, and you say, “First welfare theorem suggests that open market competition produces competitive markets, which means there is a price, and Pareto optimality, which means there is good use of resources.” In healthcare, you get competitiveness, you get price, but you don’t get efficiency, which is why in the private market you see excessive consumption of hospital care, underconsumption of primary care. It’s an incomplete product.

If I were to bring the two together, like Kaiser has tried to do in the US, for example, for a long time—but in the US, there are the Medicare, Medicaid is the big driver of the opposite force. If I go to Netherlands, for example, if I go to Israel, for example, and look at those models, or Colombia, in a developing country context, what you’ve done now is you’ve gone to customers and said, “Insurers and providers have now merged with each other. Now you don’t have a choice of buying separately insurance, buying separately healthcare. You now have five providers who can all sell you plans. You can buy a Geisinger plan, or you can buy a MassGen plan. You can’t use the plan of one to seek care, or you could, but then that will cost you a lot more money.” It’s a much narrower network. 

Now, the benefit of that is that the hospitals have no revenue source other than the insurance premium that was paid. Which means now, if they wish to give you a C-section, or if they give you more stents that you need in your bypass surgery, they make no more extra money. They have deep incentive to keep you well. Of course, they have incentive to deny you care, which regulation therefore has a role there. 

The argument is that the first welfare theorem now will apply much better because you and I, as regular customers, are better able to assess who’s keeping me and the people that are subscribing to this plan well and treating us well. I don’t know, should I get this cancer surgery or that cancer surgery? Should it cost this much or should it cost that much? We can argue till the cows come home about transparency, but I don’t know enough to be able to judge whether this is better for me or that is better for me. I don’t have the medical training. 

Whereas what I’m doing now is I’m completing the market. In finance, you have this phrase called is price a sufficient statistic? The insurance price and the healthcare price are not sufficient statistics. The integrated plan is a sufficient statistic. Now you can get competitive markets because now these markets are competing on the basis of a wellness promise, not a promise of I will be the cheapest C-section provider, but I might be more expensive on cancer. We’re not getting into that conversation at all. 

India’s Private Sector Healthcare 

RAJAGOPALAN: Before we get into the combined product, I think most people who are listening to this podcast won’t be very aware of what happens in the rural areas, or anything other than the most elite tertiary care, frankly. Unless they’ve worked in one of these places. How does India’s private sector actually work? The sense I get from your work and just generally looking around is, it is competitive on price. There’s a fair bit of price discovery, whether someone’s broken bones or it’s a cancer treatment or something else. 

People do actually go to different nursing homes. They figure out what the prices are. The nursing homes are fairly transparent about all the pricing. It’s literally like brochures are given out, both in terms of how much maybe staying at a hospital costs and every additional thing that is priced out. There’s a fair bit of pricing transparency, but your argument is there is still asymmetry of information on a whole range of other things when it comes to healthcare provision. 

MOR: Even on the pricing, I have decided because I got a dream that I need a bypass surgery. Now I’m looking for pricing, but then, do I need a bypass surgery? Who gave me that advice? Yes, I can compare C-section prices, but do I need a C-section? That’s not a conversation anybody’s having with clean hands because they all make a lot more money if you do C-sections. 

RAJAGOPALAN: Then let me back up. The problem is now multiple levels. There’s a competitive market. The market is competitive on price, conditional upon someone actually requiring the service, but do they require the service or not itself is subject to asymmetric information, and because we don’t have good primary care, it’s not clear that they always need that particular intervention. That’s your base layer argument. 

MOR: No, also in the primary care. Even in the primary care front, as we discussed earlier, why does Kerala have so many diabetics? Because they are not seeking the care that they need. Somebody needs to go to their homes, track the cohort using the Starfield idea, measure everybody’s blood sugar. If you are above 40, I must draw blood. I’m not going to wait for you to figure out that you need to draw blood. That kind of proactive care. Paternalistic care. 

If I’m responsible for your health and you start to get a foot amputation, I’m going to have to pay for it as a plan. I’m going to go out there and say, “No, no, I’ll figure this out.” I have a new paper on schizophrenia insurance. Why does schizophrenia insurance work? Because hospitalization costs a lot of money. I know that if I proactively reach out to you, I can keep you well so that the number of hospitalizations you need falls. I’m interested in that because I’ve already taken the money from you for the plan. I make no more money. In fact, I lose money every time you show up at the hospital. 

RAJAGOPALAN: Just to back up for a second, let’s assume that we’re still in a system where we don’t have primary healthcare workers. The state system is a little bit broken, so they’re heavily relying on the private sector. Just for the sake of this example, if you can stick in that world. Now, the competitive market doesn’t necessarily produce the best healthcare outcomes. You are saying the price discovery information is correct, but the price is not the only information which is currently at stake. One thing that the private sector does do in these cases, it does save lives. 

In an emergency where the alternative is either not getting care at all, or you end up waiting endlessly in a government hospital, which is very crowded in most of these districts, you get some care. Maybe you were on the margin, you could have gone a few more years without a bypass, and taken some medication, and gotten your health in order. They did the bypass, and maybe you’re just better off at this time, and you’re alive. That’s the way I think about a lot of the very expensive private healthcare intervention. 

Now my question is, is there a different way of thinking about this kind of intervention? What are the situations where they are literally peddling things that no one recommended? That would be the ultimate worry. My worry on the margin is not, in one case, if someone’s overweight and clearly not going to do much about it, they did a bypass which may have been avoided had we restructured our entire healthcare design. It’s really, are we killing people? Are we selling terrible healthcare products just to make money? Where does India lie on that? 

MOR: No, clearly any hospital, private or public, is better than no hospital. You do need emergency care. Unfortunately, the challenge is that the urban context is an excess supply of these things. If I’m talking to you and I get a heart attack, my wife has to figure out which of the six numbers she should call. It is a surfeit of choices. If I was in Mon district in Nagaland, I have no choice. There’s nothing. There’s a district hospital, and they might not even have all the staff that I need. That is, in fact, the bulk of India. Population-wise, somewhere near a billion people are in that no man’s land. 

If you were to ask me, let’s back up and say, why would the private sector be interested in any of this? The reality today is that 40,000 hospitals exist in this system in primary care. These are all mom-and-pop small setups, and some are maybe a little bit larger. They need revenue certainty. How are they going to get revenue certainty? The only way to do it is they launch their own plan, build their primary care, and make sure that they have 500,000 patients that are now bonded to them. They have certainty of revenue, so that now they can start to feel more secure.

If you look at Geisinger, for example, in the US, it’s a local hospital system. There are 40,000 such systems in India. In rural India, that’s the other thing. You might say, “Well, okay, I can see why it makes sense for the hospital. Why would a rural consumer who’s used to going shop to shop pay you a plan premium?” The reality is, actually, that they are well aware that if they have a problem, they have to go very far to find a solution. If somebody says to them, “Well, I’m going to show up every month at your place, you pay me 500 rupees a month, I’m going to look after you.” 

I’m involved in a project in rural Satara, as I mentioned to you, people are signing up like there’s no tomorrow because they’re saying, “Oh, my word, really? You’re going to come, and the medicine is going to show up, and you’ll do the blood tests here and, I don’t have to go three hours to Pune to figure this out.” No, if you have an emergency, I will make sure you get to the right place. In fact, the problem is more in the urban areas where in urban areas, the people have an illusion of choice. 

If anything, the managed care will work much better in rural and semi-urban areas. It’s in the urban areas where what you point out is going to be a bigger problem because somebody like me says, “Why should I trust you? There are six other providers.” Here, there are larger hospital systems that can take that step. I’m on the board of a hospital system called Narayana. That’s what we are doing. We are rolling out an urban model in a city like Bangalore to build a full Kaiser-style, end-to-end service system where insurance and healthcare have been brought together. 

RAJAGOPALAN: Now, there are two aspects to this. One is being in the American system, where a lot of this is stacked and vertically integrated sometimes down to the pharmacy. One problem is now prices disappear in an effective way. I have received insurance bills where, for surgery or a minor treatment or an X-ray, they’re just making up numbers as they go along. There’s no way it costs that much. It’s impossible to know how much a service costs unless you specifically ask someone how much would it cost if I just did not have insurance and I paid out of pocket. Oftentimes, those services cost, all cumulatively, whatever I’ve used through the year, they cost less than what my insurance premium was. One tradeoff is the transparency disappears. It’s all vertically integrated, end-to-end in one sense. 

The second problem is once a particular individual locks into one system, you have the same asymmetric information problem. For certain rare diseases or congenital diseases, insurance may not cover. For preexisting diseases, insurance may not cover. It’s very difficult to know that ex ante when you’re signing up for the premium because there is an incentive to sign up more and more people. There is a lack of transparency on two margins, now on the price margin and also on the service margin. What you’re saying is, conditional upon good regulation, which will intervene at the right time and make sure that people don’t get locked into one system and locked out of all access at the absolute high-risk moment—   

MOR: Or start getting low-quality care because now that I’ve locked in, I don’t have to worry about you because you already paid the premium. There is definitely regulation involved. The issue is if you bought a car, let’s go back to the car example. It is possible that the boot space you needed, this car bundle does not have, that car bundle has. The overall experience of the car is what you’ve bought. Yes, if you had to move a lot of luggage, you now have to go to U-Haul and rent another car, another vehicle in order to move your stuff. 

Now, even then, despite all these exceptions, by and large, the car market works because I can judge the car and the bundle much more accurately. It’s the same idea here. Now, obviously, if there are rare diseases, if there are conditions that need to be addressed, regulation has to provide for these opportunities. Where, if the hospital system doesn’t have the ability, they are required, by law, to make sure you get access to something. Those would be the corners. 99% of the patient base, a large hospital system like a Kaiser or a Johns Hopkins, or even a Geisinger, would have internal capacity. 

RAJAGOPALAN: I’ll tell you where I deviate from your car example. There isn’t much of a market for just front-end and back-end. I know that was just a random example that you were throwing out there, but the price system—

MOR: I’m saying if he had constructed is a market like that.

RAJAGOPALAN: Fair enough. I’m saying the price system works very well for discovery for the combined product. Even if I need a U-Haul on a random day, the price system doesn’t work very well for the combined market here. It’s a fundamental problem. 

MOR: No, that’s the point. That’s the issue of debate. Does it work better than a market in which I’m pricing C-sections and cancers and all of that separately? I’m arguing this is a more complete product and therefore price is closer to being a separation. Is it perfect? It’s not even perfect for toothpaste. It is closer to being a market that you and I can understand. Right now, what am I doing? I’m talking to people, not about your cancer experience and C-section experience. When I’m shopping, what I’m saying to people is, overall, when you needed care, did they provide it? Was primary care good? Are you healthier?

If you look at Israel, Israel is a good example of a system like that. There’s a lot of data that comes out about the overall performance of the system. And we now know, are people in this system more well than people in that system? Because that’s the true test, right? Are they living longer? What does their life expectancy look like? I certainly don’t want to sound like somebody that says this is the perfect system, that is the flawed system. It’s a matter of degree. 

The understanding I have is why does the system in Netherlands, system in Israel, system in Colombia work better than, for example, in Germany? It’s because now somebody is responsible for your health. Now your health is no longer just your lookout. Somebody is watching what’s happening to you. They are much better placed to do that, and they have commercial incentive to make sure you stay well. 

Government Insurance Instruments

RAJAGOPALAN: Now, when we think about some of the government instruments which have come in in the insurance market, so now you have the prime minister’s insurance scheme. There are many things that are being rolled out. This is not universal health coverage, but it’s also not universal insurance. It’s kind of a mix between multiple things. Just like a tertiary market is private and public, we have private insurance, and we have public insurance. How would that fit into this? 

Are you thinking about it in terms of almost like a healthcare choice model where customers can vote with their feet and say, “I have the PM Yojana, and if you want me to come to your system with my preexisting Yojana, the government’s going to pay you out every time for whatever services, and then you get me as a customer, effectively.” Ensuring some kind of a revenue stream in future. 

MOR: No, no, it wouldn’t quite work like that. If you look at Israel, what happens? It’s a tax-funded system. It’s like PMJAY. The patient chooses, “I want the Clalit system.” The Clalit system for the whole bundle of services, not procedure by procedure, charges per member per month X dollars. The government pays the remaining dollars.

RAJAGOPALAN: You’re saying these would just be separate. They don’t interact with each other.

MOR: I mean, the government in this particular case, even in PMJAY, it’s just a payout. Today it’s paying procedure by procedure. Unfortunately, as I said to you earlier, our designs are ancient designs. This design that we just spoke about, what does it look like? Medicaid, Medicare of the 1960s. That is where we got it from. That is our insurance market. What does the public sector look like? 1920s Russia. All of these systems have been reinvented in the ’90s and the 2000s.

Which is, if you look at the Lancet commission, that’s the recommendation we are making. What we are saying is, let us update our thinking about what has worked because a lot of these models we already know. Medicare, Medicaid, my understanding—I have not studied the US system carefully—is one of the biggest drivers of the overall health system inflation in the US economy because every year prices keep going up and the private sector is just a multiple of the Medicaid, Medicare pricing. Now, if we seriously implement PMJAY at the level at which we aspire to, we will build the same system, very high costs and very poor outcomes. 

RAJAGOPALAN: My understanding of PMJAY is not that they are paying too much right now. My understanding was actually that in certain states, private sector doesn’t want to deal with it at all because it costs too much.

MOR: No, that’s true at a cross-sectional level. That is at cross-sectional level. Dynamically, PMJAY prices have been revised upwards, I think, six times already. They will go on getting revised. In 1963, Kenneth Arrow published a paper. Mark Pauly, I am told, took five years to write a rejoinder because he was Kenneth Arrow. Mark Pauly was just a student. He wrote in that letter to the editor, it’s a famous, famous letter to the editor, in which he says, “If you make the good zero price at the point of consumption, you will get overconsumption.” It is not about people being unethical. It’s simple consumer economics. 

If you make sure the supply curve doesn’t bend, you can supply as much as you want and price, there will be no impact. You will build a health system that will be excess supply of services that people demand, which is hospital care, and excess pricing, which is exactly where the US has ended up. This is where we are headed with our current, not just PMJAY, even the private sector insurance models, are all indemnity insurance. 

This was the case in the US in the ’80s and ’90s, even. Now you see the US, it’s still a journey, but 90%, 95% of plans have some kind of managed care overlaid  over it. They can’t quite reverse gear completely because the incentives are too locked in, but the Europeans have gone in a very different direction.

RAJAGOPALAN: Yes. In some sense, this goes back to the head of our conversation. There are multiple things different people are solving for. One thing that many people in government are solving for is the financial risk. That is, if there is an adverse health event, then there’s a huge out-of-pocket expense, and we need to backstop that. There’s a second kind of risk which a lot of us worry about, which is people shouldn’t be needlessly dying. It’s the healthcare risk. Are we actually providing healthcare? 

Then there is a third level layer which you are talking about, which is, can we just nip this in the bud? Are we actually providing a value which is end-to-end, that a lot of this is not required in the first place? In some sense, it’s like three parallel conversations happening, each one of them with a different solution, and they naturally can’t interact with each other.

MOR: Reform is a hard journey. It’s like an aircraft. You can’t stop it. It has to keep flying. You have to modify what is going on to the aircraft while it is flying. It has been done. I did a paper that looked at outliers in health systems in 2022, in which I looked at 200 countries, and I put Indian states as countries in that calculation. 

People have tried many, many experiments. For example, France and Germany. Why does France spend 5% less and have 5% better outcomes than Germany? The one difference I found was France does not permit patients to show up at hospitals without a primary care consult first. Germany allows you to bypass primary care. That one difference seems to produce the bulk of this result. Now, a lot of this we can learn. We can start to gradually bring it in. There is a whole discussion about how do you layer in the reform, but I think we have to start with what is our end state? What do we want? That’s what we are discussing now. 

RAJAGOPALAN: Yes. Your argument is we want a more complex system. We want it to be more adaptive, but we also want behavioral change. The behavioral change has to happen at the patient level, at the community healthcare level, at the nurse level, doctor level, all of it.

MOR: The patient-level behavior change is hard. Patient-level behavior change is hard. Doctor behavior change is hard. 

RAJAGOPALAN: Everything in the middle is doable.

MOR: Hospital behavior change is hard. Health workers, you can do a lot. Hospitals will respond to financial incentives. If you change their incentive structure, they will turn on a dime. If you told them, “I’m going to give you money only once a year. Not per procedure,” immediately their mind will shift to say, “Oh, now it’s a very different environment.” The benefit of commercial people is they can pivot, pivot, pivot as quickly as you need. It’s a matter of thinking it through. I’m not going to suggest to you that this is some trivial task. It’s a huge, complex country, and there are many moving pieces here. My sense of what we are discussing now is what are some of the endpoints that look desirable, and what are some of the pathways to get there?

Insurance Regulation in India

RAJAGOPALAN: I have a couple of questions on the financing and the insurance regulation part. When you think about insurance regulator in India overall—and we have a regulator not just for health insurance, it’s across the board for many different kinds of insurance—how does our insurance regulation and overall regulator need to change to adapt to what kind of healthcare system we have at the moment, and also what we’re building as we go along? 

MOR: See, you’re quite right. We have to start with what we have. Now we cannot raise a magic wand and say we need a new—maybe we’ll get there eventually. Today, there is a deep asymmetry that exists. Hospitals can own insurers, insurers can’t own hospitals. 

RAJAGOPALAN: Yes, which is exactly backwards of what you would like.

MOR: One should remove that asymmetry. Today, the law requires you to discriminate patients on the basis of their prior health condition. This is illegal in most countries. That’s the one thing you can’t do. We have to shift out of this older regulation to a more modern view, which says that you have to take anybody who applies. You can charge a differential premium, ideally charge the same premium, but then we need a more deeper market reform, which Germany has done, many countries have done, which is what they call risk equalization. 

Now, what happens is that your portfolio is assessed, and the reinsurance machinery takes extra money from the low-risk portfolio and pays the high-risk portfolio. At the point of acquisition of the customer, you’re not worse off. There are a number of issues. Now the other issue is entry barrier. Today in India, it costs 100 crores—I don’t know what that would be in million dollars, but a fairly large amount of money—to start an insurance company. 

France requires one-sixth of that. Why do we have such a large number? Now we need to think through these issues and start to imagine an insurance market in which a small hospital can get into the insurance business and build a plan for their town. Currently, this is impossible to do except for very large hospital systems, because they don’t have the kind of resources that are needed.

RAJAGOPALAN: Here, I want you to also wear your experience-from-your-banking-days hat. We seem to have a fundamental problem trusting smaller players to deliver on financial instruments. India’s experience is either chit funds or huge state banks. We don’t think of small solutions. When there are excellent working solutions like, say, Canara Bank in the ’50s and ’60s, we end up nationalizing them. We generally don’t trust private sector players to actually deliver financial instruments to the lowest decile of income earners in India. Is this a mindset problem? Is there some risk pooling issue that I’m missing? Why is it that India’s bad on this across the board? 

MOR: This is no longer true in banking. In banking, now you can have a small finance bank. We’ve experimented with local area banks. We have cooperative banks. Now, for example, the regulator requires every bank to be fully automated. What we found is some banks didn’t have the resources. They were subsidized to automate. 

At the press of a button, the regulator knows your balance sheet. That doesn’t mean everything goes away, all the problems go away, but the size is no longer the issue. What is different about insurance from banking is that insurance has got reinsurance. The reinsurer is also a quasi-regulator because when they reinsure you, they spend time with you. I’m not talking about millions of insurance companies. At most, we’ll talk about 7,000, 5,000, 10,000 insurance companies. 

RAJAGOPALAN: That’s not so crazy in a market like India.

MOR: It’s not so crazy. With automation, with all the digital stacks that we have, we can create backbones to tell everybody, you have three choices, this, this, or this. This is what we did in the cooperative banks. We gave them four choices and said, one of the four, you have to pick if you want our guidance. Otherwise, you find something else, but then you have to manage it yourself. The good thing in India is we know these businesses extremely well. We have good actuaries. We have good analytical capabilities. We have some of the smartest brains in statistics. 

It’s not impossible for us to imagine a regulatory environment in which there are thousands of insurance companies that we are managing because most insurance companies that are small will reinsure bulk of their risk. I don’t think the barrier is coming from the belief that we don’t know how to manage them. 

RAJAGOPALAN: You’re saying the barrier is just that we’ve kept some old thresholds of very high capital requirements, basically, and that can very easily be revised.

MOR: I think it’s an incumbency bias. I think the incumbents are not comfortable, particularly in the public sector, in allowing new players to come in because they feel they will lose market share. I don’t understand where the fear is coming from. Insurance penetration in India overall is 3%. 

RAJAGOPALAN: Yes. There’s a huge market to tap.

MOR: The voluntary insurance market is 5%. It has not grown in the last 10 years. There is an enormous opportunity out there. Reserve Bank saw this, which is why Reserve Bank has gone out and licensed so many different types of banks. I would hope that the insurance regulator is inspired by that and says, “Well, those guys did it, and they are doing a good job in a much more complex business,” because a bank cannot share its risk. A bank is an aggregator of risk. 

RAJAGOPALAN: Yes. I mean, the Reserve Bank is not great, according to me, at supervising both public sector or private sector banks. I mean, we’ve had some serious problems in public sector banks. If we do think of an insurance regulator where we already have some very large incumbents, which are public sector insurance companies, we’ll need to rethink how we model that, right? 

MOR: I was on the board of Reserve Bank for many years, and I chaired the eastern board of the Reserve Bank. We’ve gotten better and better at it. It is true that we were not perfect. Even today, we are not perfect, but we are better and better and better at it. Today, for example, with the modern technologies that we have, the understanding of risk science that we have, we are running a relatively stable banking system, which is relatively large without too many crises. You are right, it could get better. 

RAJAGOPALAN: Crises are one thing we’ve been relatively good at arresting. We’ve also been relatively good at arresting terrible runs on banks, things like that. We’ve had bank failures. Mostly, we hide them by merging them with SBI or merging them with each other or something like that.

MOR: Oh, no, we have deposit insurance as well. Cooperative banks fail a lot.

RAJAGOPALAN: We have deposit insurance. Yes, but the bigger problem in the public sector is a lot of times it used both the public sector banks and also the private sector parasitically to say that we’ll have loan waivers for farmers and things like that. Do you worry at all about things like that in the health sector? Let’s say there’s a public health situation happening—COVID was an extreme case, but you could have local public health problems in a given district or something like that. Are you worried about politics intervening and saying, “We’re going to make this is free for everyone, the local insurance companies have to serve everyone”? Is that a concern at all if we start going toward this kind of an insurance stack?

MOR: Not in a universal health coverage market, because there, by design, you want it to be free. You want it to be tax-funded and free.

MOR: There are challenges. I’m not saying there are no challenges. Like the tobacco lobby might make it hard for you to tax tobacco. There are issues, but at the health system level and the separation that we’re talking about, what we call the purchaser-provider split in the public sector is, in a way, a shield against interference by the political system on the day-to-day functioning. Why does ONGC work so well? Why does Indian Oil work so well? Because the corporate structure preserves the autonomy of the institution, even though the policy machinery, the oil ministry, is responsible for all of them. It cannot overstep the corporate boundary easily. 

Whereas today, when it’s all integrated, you can have an individual doctor say that, “I don’t want to be transferred to Purnia, you rescind my transfer.” A politician could come in and say, “No, no, don’t do that.” In a structure where you split it, you now are accountable for outcomes. You get paid for outcomes. The good thing is, it’s not as if India is the only political system in the world. Public sector implies the government. I wouldn’t say while we have our own share of problems, if I compare ourselves with Turkey or Thailand or other countries, they have as many political problems as we have. If not, in some cases, maybe even more. 

RAJAGOPALAN: If I go to the top layer of how we finance the entire healthcare system, you’ve again written about this, the two broad models are use taxation, to then have redistribution of healthcare services in some sense. It can work with various market versus state providers. That’s a matter of delivery. The second option is have some kind of a contributory system, which is private contributions. Ideally, not this crazy out-of-pocket expense we have in a situation of emergency. 

What you’ve written is that the standard prescription that works for most low and middle-income countries, which is just use taxation and then redistribute because there are too many people who are poor and who either don’t have the knowledge or the access or the funds to contribute, is maybe not such a great idea because there are a whole bunch of other political economy problems that also are present in low and middle-income countries, right?

Their taxation structures tend to be very regressive, tend to rely on consumption taxes, and they have all kinds of state capacity problems when it comes to redistribution. You still recommend a reasonable role, at least that’s how I read it, for the contributory system, even in a country as poor as India. First of all, am I reading what your position is correctly, and how would you think of that when you think of everything else you are describing in the Indian market? 

MOR: No, my view is that—and you see this in our commission report as well—our country, while it has a very small formal sector share, and you might argue how can they pay for the whole country, the reality is 80% of income and assets belong to them. It’s 7%, 10% of the people, but they control a very large proportion of income generation of the entire economy. It’s quite fair for them to pay taxes. You’re right, a lot of our taxes are consumption taxes, they’re regressive, but be that as it may, they already exist. 

The question is, how do we use the taxation resources to build out a universal health coverage system? My argument is, in many states, we’ve already got enough money now. The issue of do I have enough money, can I tax more people, is for many states is off the table. Now I have to get into a design conversation, starting with a state like Goa or Kerala or Delhi, to say, why can you not build a universal health system for everybody, which is completely free from primary care to tertiary care? 

It may not have the most expansive benefits package. If you want gender alteration surgery, it may not be covered in this package that we have, but everything short of those extremes is taken care of. The argument I’m making is we have enough tax resources to afford this. Now there will be a group of people, which is this 10% or 5% or 15% of the population, that will say, “I want an expanded benefits package.” 

RAJAGOPALAN: And they can buy that.

MOR: Even for them, we need what we discussed earlier, for two reasons. One is that if you don’t give them the system that we spoke about, where the payer and provider are integrated, they will not get good health outcomes. Who are the people, for example, in Kerala getting the highest and the least required C-sections in Kerala? The rich. In fact, the poor are getting better healthcare in Kerala because they are poor. Because they cannot afford the C-section. They are getting a reasonable amount of C-sections. We have to protect that population. Make sure that they are getting good healthcare, even if they feel that they know the best. For the second, there’s a deeper problem. If you now build a system in which they go out and do currently fragmented purchasing, the prices of specialists are going to rise. 

RAJAGOPALAN: Exactly.

MOR: Now you will not, in the public sector, be able to hire a cardiologist. Why? Because cardiologists are getting paid astronomical sums by the private sector doing things that they should not necessarily be doing. We have to bring price control. If you do it Japan-style, in which you regulate all the prices through a blue book, what the Indians have done, what the Japanese have done also, is they’ve increased supply. Now you said a stent price is fixed, but I’m going to do more stents now. An MRI price is fixed, I will do more MRIs. 

The Japanese consume, I think, six times more MRIs per capita than the British. I have a headache, get an MRI. Which is why we have to think a little bit about how do you complete that market. Partly for the interest of the consumer that will buy those products and partly for the interest of everybody else because now I can’t get specialists because I’m priced out of them, which is what has happened in the US. Oncologists are paid million-, multimillion-dollar salaries. If you talk to the Japanese, they complain that they are paid like everybody else. 

RAJAGOPALAN: Here there are two parts. You’ve written about how we need to think about pharmaceutical prices because that’s a very large proportion of the out-of-pocket expense. There is a regulatory problem in the pharmaceutical market. The government ends up being a big purchaser, but individuals at the private level are also spending a lot on this. What is a good way to think about the pharmaceutical market in India, especially given that we have so many generics and we have supreme court rulings on waiving international patent requirements for lifesaving drugs? 

We have a whole set of exceptions which are built in for India. Maybe you don’t get the latest cancer drug in India the moment it’s available in other parts of the world, but most drugs are available in India relatively cheaply. Not just PPP comparison. Just standalone, they’re available relatively cheaply, but they still end up being a huge out-of-pocket expense. What is going on in that market in terms of both pricing and regulation? 

MOR: There is excess consumption. It’s not the price. Indian drugs are very cheap. It’s just that people are consuming too many medicines relative to what they need, because they go to the pharmacy as the first point of contact. The de facto primary care provider in India is the pharmacy. Not the Ayurvedic doctor. Not the MBBS doctor. It’s the pharmacy. I think 55% or 60% of all first conversations happen with pharmacies. Now, pharmacies are of all kinds, and they are doling out medicines at a very high rate relative to what people need. Not just for antibiotics. If I look at, for example, contraception, they’re giving the emergency pill. Now emergency pill is not meant for birth control in a marital setting. You need a longer-acting reversible or nonreversible method.

The pharmacy knows that that’s not a market that can work because the drug is expensive. The direction of travel for us has to be drugs. This is the same problem in diagnostics. Drugs and diagnostics have to become free. They have to come as a part of the overall delivery system and not be a separate out-of-pocket piece of expenditure.

RAJAGOPALAN: This contradicts a little bit what you said before, because you were talking about, at the point of choice, if something is priced at zero, then people are going to overconsume. Won’t it make that problem worse? 

MOR: No, which is why the payer, that’s the issue. That’s why you don’t separate it. 

RAJAGOPALAN: Okay, so you integrate it. 

MOR: Yes, because now I will not get a prescription for a medicine I don’t need. Because who’s paying? Somebody’s paying. Now the hospital is paying. It’s not you that is deciding I need this medicine or that medicine. It’s the hospital that is saying that I will give you a prescription for this medicine, and then the second thing is, we build formal partnerships, even in the current system. Let’s assume that the Holy Grail we don’t find, but in the current system, can we start to build partnerships? 

Then don’t forget, there is a whole online revolution that has come. Now there are many online pharmacies that can deliver medicines to 99% of Indian PIN codes. They are a very controlled system because they are corporate agencies that cannot afford to play fast and lose with the law. Because, one lawsuit against them, billion dollars of business gets shut down. They are very careful. You call up PharmEasy, you call up 1mg, order a medicine. If you have a prescription, that’s good. If you don’t have a prescription, some doctor will call you. 

I’m saying that either we do a formalization of the pharmacy, one way or the other, even in the current system. What I was saying to you earlier is, the interactions that I have observed and literature I have read about pharmacies in India, they want to go in that direction. They no longer wish to be outside the pale of the law because they want to be a part of the chronic disease machine. They make a lot more money if they have steady clients. There’s excess supply in that market. 

RAJAGOPALAN: People want to go to them, which is the most important thing. In healthcare, you want to be where the people are already going because that behavioral change is also incredibly hard. They trust the local pharmacist, so presumably that’s the intervention that you need. One last question on the supply side of overall healthcare. India is always chronically understaffed for all sorts of things. Whether you’re talking about number of judges per million or number of police officers, number of community workers, everywhere India is running low on supply.

On doctors, it’s a little bit crazy because everyone’s dream in India is to be a doctor. There are kids who are studying and overcompensating in their teenage years just to become a doctor. There is a demand for it from all sorts of consumers. Then there is this ridiculous artificial clamp down or artificial scarcity because we’ve not been very good at regulating medical colleges or setting up new colleges and so on. When you think of a stack, like you do, which is end-to-end, and you think about the government regulating pharmacists or the government regulating insurance providers, what is a good way of thinking about how we produce more healthcare professionals? 

MOR: At the primary care level, if you buy the idea that we have to get more ambitious about community health workers, we don’t have a supply problem.

RAJAGOPALAN: Yes. No, no, I’m talking at the doctor level. 

MOR: There are literally millions of workers who desperately need a job. You need to make them equipped well. That’s why I keep going back to the Alaskan model. This is exactly what they have done.

RAJAGOPALAN: They don’t need five years of medical school and two years of residency and sort.

MOR: No, on the contrary. Technology makes it possible. It’s a near-zero training approach because it’s the tool plus some training that makes you that person. It also avoids the problem of you thinking of yourself as a doctor. You’re not a doctor. You’re part of a system. 

RAJAGOPALAN: If you get protocols on the spot at that moment in time when you need to deliver the healthcare, there’s a system of doing that. Relatively, no low training except to follow protocol.

MOR: Follow protocol, use the instruments. How do I use an otoscope? How do I use an ophthalmoscope? I am trained somewhat. How do I draw blood? Not for five years. If you see the Alaskan program, the base training is one month. If at all you have an opportunity, visit with them and take a look at what they are doing. They’ve been running for 50 years. Clearly, it’s doable. You need some certification. You need some diploma. You need career paths. 

What Thailand did is they had the same problem. They started to certify these nurses on the ground, community health workers, but they built a career path. Many of the health workers that they started out in 1995 are today oncology nurses because they gave them a path which we have not done. We have ossified them into the ASHA. We ossified them into the ANM. There’s no next step. So we have to think that through , but it’s a doable idea. The good thing is, despite all these constraints that you mentioned, the undergraduate medical education seats have expanded dramatically. 

RAJAGOPALAN: Have they still expanded enough is my question.

MOR: I think they have expanded perhaps more than enough. Despite that, because don’t forget, there’s a global demand for Indian doctors. 

RAJAGOPALAN: Exactly.

MOR: The world is aging.

RAJAGOPALAN: A lot of the best leave eventually.

MOR: The thing is, if I don’t need the doctor at the primary care level, I can staff my subdistrict hospital, district hospitals much better. I did a project recently for Himachal Pradesh. Himachal has 20% excess supply of MBBS doctors.

RAJAGOPALAN: You’re basically saying there needs to be some reallocation mechanism.

MOR: Now you’re saying what about the specialist? We are overusing the specialist today, which is the point I was trying to make to you about the private sector. Maybe there are some disciplines, like psychiatry, where we may absolutely have a problem. In many disciplines, if we did the cascading and the step care properly, while the number of seats in the specialist category have not expanded as quickly—and then don’t forget India has this unique thing that began with the British. If at all you are in Bombay the next time, there is a college right in midtown Bombay called the College of Physicians & Surgeons. This was set up by the British in 1911. What it does, it takes an MBBS undergraduate doctor, in six months, gives him or her a specialist diploma in orthopedics, in obstetrics, in pediatrics, in doing anesthesia, in radiology. India has adopted that called the DNB system. It’s a diploma. It’s a short diploma, one-year, two-year diploma. 

RAJAGOPALAN: We can expand supply quickly for areas where we really need it.

MOR: Very quickly.

RAJAGOPALAN: We need endocrinologists desperately, we need psychiatrists desperately, and that can scale fast.

MOR: Then there’s a third possibility, which is what now we have rolled out, say, in the northeast, is the tele-ICU services. Now it’s not telemedicine, but the central hub has the top pulmonologist, top oncologist, and he is able to manage. We now have 300 hospitals, government hospitals, that are able to provide ICU services, even though they don’t have any specialists. Because the MBBS doctor knows enough that they can listen to an instruction from a specialist and do it. 

Like intubation, that’s a skill you have to learn. It’s not a lot of work. It’s a week of training, a lot of practice, but you can get good at it. There are multiple tools that India has to address, but we have to design the system well. If we send everybody to tertiary care, we have a COVID-like situation. 

RAJAGOPALAN: Then everything gets overloaded.

MOR: Then everything gets overloaded. Nobody has any time. The curse of OPD in large government hospitals has to disappear. Why are these people standing in line for days at a time to have a 30-second conversation with a doctor? Thailand have fixed this. They have made sure that you cannot show up at a hospital without a primary care recommendation. They have made sure that primary care is working well. 

RAJAGOPALAN: I think that’s the part we got to fix, really. Everything you’re talking about eventually comes down to the first point of contact.

MOR: We can. Now we’ve built out these health and wellness centers. I was involved in the design of these ideas. They came from some work I did, some work others did in primary care. Now we have 250,000 of them. Now you might say, many people said initially it was just a coat of paint and nothing’s going to change. That’s not the evidence. The evidence is showing that actually these centers are starting to deliver on more comprehensive care than the previous subcenters, et cetera, were able to do. It’s more than just a coat of paint. They’ve added new staff. They’ve added new people. Now we have to build on it. We have to make sure that all— 

RAJAGOPALAN: Capacity building takes time.

MOR: —the other pieces of the puzzle fall together. I think it’s very doable as a direction of travel.

RAJAGOPALAN: That’s very comforting to hear. It’s an optimistic note to end on. Thank you so much, Nachiket, for this. I don’t work in health economics and health policy, so a lot of it was revelatory to me and, as an economist, super interesting to read. Thank you so much again. 

MOR: No, thank you for having me. I enjoyed talking to you. Sometimes when you talk, ideas get clearer. That was also very helpful to me. I learnt a lot from listening to you about other examples and other issues that you raised. 

RAJAGOPALAN: Thank you so much.