#
How should we think about health?
#
Well, here's how I think of my personal health.
#
I don't sleep enough, I don't exercise enough, I eat the wrong food.
#
In other words, just because I feel okay now, I act as if I don't need to do anything to
#
make sure I feel okay later.
#
But one day, I'll fall ill and then I'll be at my doctor's clinic saying kindly give
#
medicines, I will be disciplined in taking them, I care about my health.
#
This is the wrong approach.
#
My thinking about my personal health care is flawed and I will only realize it when
#
I fall ill and then if I get better, I will forget it again.
#
And that's our nation's attitude towards health care, especially if we are among the
#
urbanized English speaking elites.
#
We feel okay most of the time and when we don't, we have access to fancy doctors and
#
We don't spend time worrying about our health care system.
#
And yet, we are now in the middle of a crisis where we can't ignore it anymore.
#
You could say as a nation, as a democracy, we are suffering the impact of a chronic disease
#
that we had earlier normalized and it is now making us suffer.
#
COVID-19 has made it clear that our health care system is broken.
#
As much as that, it has also revealed that our dysfunctional state is a greater crisis
#
In fact, our dysfunctional state is a chronic crisis that we have been suffering from for
#
Now, maybe fixing the state is beyond us, but we must ask ourselves, how can we fix
#
our health care system?
#
We need to do it and we need to start now before COVID-19 recedes and we slip back into
#
And before we fix it, we need to understand it.
#
Welcome to The Seen and the Unseen, our weekly podcast on economics, politics and behavioral
#
Please welcome your host, Amit Verma.
#
Welcome to The Seen and the Unseen.
#
My guest today is a brilliant economist, Kartik Muralidharan.
#
Kartik had last appeared on episode 185 of this show, which as of now is the third most
#
downloaded episode of mine.
#
It was a comprehensive discussion on our education system and we had promised each other that
#
we would one day have a similar long chat about our health care system, another subject
#
that Kartik has studied deeply.
#
Well, that day has come when the subject feels so immediate and visceral and not just something
#
that academics and policy monks should care about.
#
Now, just as that earlier episode was titled, Fixing Indian Education, I thought of calling
#
this one, Fixing Indian Health Care.
#
But we both agreed that it seems almost arrogant to say that we know how to fix this.
#
This is such a deep problem that we first need to try and understand it.
#
Thus, this title, Understanding Indian Health Care.
#
Now, it so happens that Kartik is writing a book about fixing the Indian state and he
#
sent me the early draft of his chapter on health care.
#
I can't share a chapter with you, but the show notes contain many, many of the papers
#
that are mentioned there, which give a foundational understanding of our health care system.
#
I encourage you, don't just listen to this episode, also go through those papers, we
#
all have a stake in this thing.
#
Kartik and I have also agreed in a few weeks from now to do another episode on federalism
#
Now, one thing I found delightful about this particular conversation is that it felt like
#
it was part of a larger ongoing conversation spanning many episodes of mine.
#
Kartik wrote to me and offered to have this conversation after listening to my episode
#
on vaccines with Ajay Shah.
#
And during this conversation, he kept quoting previous guests such as Ashwin Mahesh and
#
Anirban Maapatra, adding to their insights and carrying their arguments forward.
#
This is the best thing about the seen and the unseen, as regular listeners would have
#
Kartik and I disagreed on quite a few things in this episode, such as the high petrol prices
#
because of how much the government taxes, which I feel is a bad thing, but Kartik feels
#
And that actually illustrates another fundamental disagreement between us.
#
He has faith that the dysfunctional state can be reformed.
#
And while I would love all efforts towards that to continue, I feel that we must stop
#
depending on the state and try to figure out how society can solve its own problems.
#
But that's a larger ongoing conversation, and what we both agreed on is that we need
#
to understand this problem first before we think about how to solve it.
#
And some of Kartik's insights in this episode are just stunning.
#
Yes, my friends, we have a little Antakshari session at the end of this episode, and Kartik
#
sings a couple of songs for us.
#
Music is the best medicine.
#
But before we get there, let's take a quick commercial break.
#
Do you want to read more?
#
I've put in a lot of work in recent years in building a reading habit.
#
This means that I read more books, but I also read more long form articles and essays.
#
There's a world of knowledge available through the internet.
#
But the problem we all face is how do we navigate this knowledge?
#
How do we know what to read?
#
How do we put the right incentives in place for us?
#
Well, I discovered one way.
#
A couple of friends of mine run this awesome company called CTQ Compounds at CTQCompounds.com,
#
which aims to help people up level themselves by reading more.
#
A few months ago, I signed up for one of their programs called The Daily Reader.
#
Every day for six months, they sent me a long form article to read.
#
The subjects covered went from machine learning to mythology to mental models to even marmalade.
#
This helped me build a habit of reading.
#
At the end of every day, I understood the world a little better than I did before.
#
So if you want to build your reading habit, head on over to CTQCompounds and check out
#
The next batch starts on Saturday, May 23rd.
#
And if you're listening to this after that date, they keep having regular batches.
#
They also have another great program called Future Stack, which helps you stay up to date
#
with ideas, skills, and mental models that will help you stay relevant in the future.
#
Future Stack batches start every Saturday.
#
Also check out the Social Capital Compound, which helps you master social media.
#
What's more, on all of these, you get a discount of a whopping 2,500 rupees, rupees 2,500,
#
if you use the discount code Unseen.
#
So head on over to CTQCompounds at CTQCompounds.com and use the code Unseen.
#
Karthik, welcome back to The Scene and The Unseen.
#
It's a pleasure to be back.
#
You know, when we did our episode on education, episode 185, which was so incredibly popular,
#
we had sort of left a teaser for the listeners that one day we will talk about health care
#
and hey, that day has come in rather difficult times where health care has become not just
#
some abstract issue, but it's really touching everybody's lives here in intimate and sometimes
#
Tell me, how have the last few months been for you?
#
Because you know, we have, of course, faced this global pandemic.
#
And you are, you know, given what's happening in India, you are very close to it because
#
so many people close to you must be here and getting affected.
#
At the same time, you're far away.
#
At another intellectual level, you're closer to it than most people because you understand
#
all the fault lines in our health care system, which we'll talk about as we go along and
#
all the fault lines in the Indian state as we'll go along.
#
So just at a personal level, how has it been for you, you know, emotionally and just in
#
terms of your intellectual pursuits during this time, you've also been writing this book,
#
which I can't wait to read on, you know, fixing the Indian state.
#
So just tell me a little bit about how it's been for you over the last few months.
#
You know, I think the last month has been particularly difficult.
#
The most kind of difficult part in some ways has been this is the longest 15 months I have
#
I mean, normally I come every two to three months and I had gotten my vaccines in early
#
March and I was really looking forward to making a trip.
#
In fact, in late April, early May, and that is clearly not feasible right now.
#
But more than that, just seeing how this situation has exploded in the past month, I think it's
#
yeah, it is very surreal to be kind of physically safe where I am in the US, but kind of mentally
#
essentially facing as much if not all like, you know, I mean of the stresses and turmoil
#
of parents and friends and people who are affected.
#
No, it's been very, very difficult.
#
I think there are obviously things you can do from here in terms of donations and stuff.
#
But I'm also trying to apply my mind to the few policy things that could be done.
#
And you know, we'll talk more about that going forward.
#
So yeah, you know, I think just applying first principles to sensible design of health care
#
systems and how do you kind of use this moment of kind of reckoning almost about the weakness
#
of our health care systems to make the right investments going forward.
#
So you know, just like we say about the 91 reforms, the crisis by itself was not enough
#
You also needed the blueprint to be ready of what you're going to do.
#
And so my hope is that some of the analytical principles we'll talk about today, you know,
#
can be useful in kind of the nation's reactions to this moment of crisis in terms of the kind
#
of investments we'll make going forward.
#
So there are a couple of thoughts I have.
#
I mean, obviously, we want to talk about health care in general and really maybe talk a little
#
bit about COVID at the end.
#
And I've had a few episodes on COVID already.
#
So I don't want to give my listeners too much more of that depressing subject right now.
#
But this kind of comes up as a pertinent question to ask you, which is that there are two aspects
#
of the pandemic that I've noted.
#
One the smaller one, almost a trivial one that I chatted about with Anup Malani in the
#
episode I did with him is that it inverted some of our values, like some of the things
#
which could otherwise be considered good, suddenly worked against us, like population
#
density, which we know is good, which is why people go to cities, you've got larger economic
#
networks, more people together, more opportunities, all of that.
#
But in the context of this pandemic, that suddenly, you know, became a bit of a con.
#
But that's a trivial minor thing.
#
The larger thing that I noticed, and I wrote a column about it last year in April when
#
this first began, stating that our greater crisis is not just COVID, which will pass,
#
but the dysfunctional Indian state, which has been in crisis for, you know, the last
#
I think in that column, I used a stat, I don't really remember it, but something like 3000
#
children in India die of starvation every day, something like that.
#
I'm sure you have kind of better stats, which to me is indicative of an ongoing disaster,
#
which we have completely normalized.
#
And if there's one thing that COVID has done, it has shed light on a lot of things which
#
The apathy of politicians, the dysfunctional state more than anything else.
#
The fact that the state simply doesn't work.
#
I mean, I've always said that where the state is most required, most justified, it is absent.
#
There is effectively no rule of law in India.
#
And today we see that in the state of our healthcare, where it was most required, it
#
was absent, and we are kind of seeing the consequences of that.
#
So do you think that there is also, therefore, a teachable moment in this, that you can look
#
at this moment and say that, listen, these things have been wrong forever, right?
#
We haven't noticed them right now.
#
You are all looking at them because they are stuck, they are in relief.
#
Now, this is a time where we have to therefore sit down and at an intellectual level, analyze
#
the problem because it's clear we can see it and there is a will to do that.
#
And then think about what to do.
#
Is this something that you've thought about and as someone who studied the inner workings
#
of both the state and healthcare so deeply for so many years, you know, is that how you
#
That these are things that were there and now these fault lines are becoming stuck?
#
Yeah, no, I think I've been rewriting many parts of the introduction of the book.
#
I started the first chapters two years ago, right?
#
Well, before the pandemic.
#
Yeah, I think those of us who have been studying the state have kind of known about these weaknesses
#
and known that there was kind of a gradual atrophy, right?
#
I think Ajay also talked a little bit about why state capacity is weaker now than it was
#
like, you know, 30, 40, 50 years ago.
#
And frankly, you know, I think as with all of these things, it's multi-causal, right?
#
And we can get into that.
#
But I think in terms of the book, it almost feels like in a macabre way, right?
#
Mean that it is kind of, you know, it is hopefully rises to the moment, right?
#
Because the book is called Fixing the Indian State, but, you know, hopefully the one silver
#
lining from all of this disaster is that people realize that this is no longer something that
#
we can allow to kind of rest.
#
So coming to the different reasons for why I think we've under invested the state, one
#
of the downsides of liberalization, I think it was positive in almost every way, right?
#
But is it has also led to a complete seceding of the elite, right?
#
Mean from being recipients of public services, right?
#
So the Indian elites and policymaking classes, middle, any taxpayer, it pretty much avails
#
of private health care, private education, private security, you know, and private whatever
#
I mean, and so one way to characterize the current moment, it's almost like the elites
#
and middle classes have been flying first in business class in the plane that is India,
#
Mean, and kind of surrounding them with creature comforts.
#
But now is the moment when you realize when the plane crashes, it takes you all down,
#
like, you know, I mean, regardless of what class you're flying in, I mean, there is obviously
#
the hyper rich or the private jet types who are leaving the country, right?
#
Like, you know, but except for that type, I mean, I think the rest of us are kind of
#
in this plane together and sometimes physically not, but emotionally we are, at least that's
#
Hopefully, this is one of those wake up calls as we, you know, we would have wanted to approach
#
the 75th year of independence in a moment of kind of, you know, feeling great pride
#
at what we've achieved.
#
But, you know, sometimes you need these kind of rude wake up calls to really look at how
#
hollow the state is inside and build and start working on the systematic project of building
#
up, you know, as academics and intellectuals, you know, we can contribute to this public
#
discourse with a certain amount of conceptual clarity and analysis.
#
And then in terms of taking it forward, it's obviously the larger community of state citizens
#
So you know, in that way, I hope, yeah, so timing wise, I think I'm about 95% done with
#
the first complete draft of the book, so it's almost done.
#
But the problem of course, is that as an academic, the default first drafts tend to be very defensive
#
because academic writing, you write every sentence, assuming a peer reviewer is going
#
to come pounce on you and saying, how did you say this, how did you say that?
#
And so, you know, so in that sense, there's still substantial editing needed.
#
I think the first draft you've seen in the chapter I've sent you has a lot of content,
#
but it's still very terse.
#
And so I, it'll probably another six, eight months of like, serious editing, but I'm hoping
#
the book will be able to hit the stand sometime next year, you know, and coincidentally leading
#
into the 75th year of independence and hopefully like, you know, can be part of the national
#
conversation at that time about what we need to do.
#
And hopefully we'll remain part of the national conversation because I'm really looking
#
I think it's an important book.
#
I'll come back to the book in a moment because I do want to talk a bit more about the book,
#
the process of writing it, editing it, almost workshopping it with your students as you
#
kind of discussed before we started.
#
But another sort of question, which is something that I've been thinking about over the last
#
few days, but before that, a little story I'd like to share with you and my listeners
#
and this concerns a dear friend of mine.
#
Both his parents got COVID and were in different ICUs in different hospitals in Delhi.
#
And his dad was in a hospital called Jaipur Golden.
#
There were 21 beds in the surgery ICU there.
#
And he was in that particular ICU and one night oxygen ran out, 20 people died.
#
So only his dad survived.
#
The other 20 died and his dad was devastated because the next day everybody else who was
#
a patient in the ward with him was taken away in a body bag.
#
And his dad is still struggling and fighting it out.
#
And my friend's mom died a few days later, though she was getting better.
#
He thought that she'll get better.
#
He was worried about his dad.
#
And the failure of the state, the failure of everything, our politics, the state, everything
#
And he called me and he said that we have to do something.
#
That everything is broken.
#
You know, it's up to us to fix it.
#
We have to do something.
#
And his first thought in his deeply emotional state was that we do something in politics.
#
And me being the cynic that I am, I said that no, nothing will happen because politics
#
is the supply end of the political marketplace.
#
And also, what will happen if you become a successful politician, you'll take charge
#
The state is dysfunctional.
#
The state will remain dysfunctional.
#
That's my negative view, which I've always held, which you're kind of aware of.
#
So I said, you can't have hopes from the state.
#
What you have to do if these problems are to be solved, they have to be solved by private
#
You've got to figure out ways to solve it.
#
And like you've pointed out, the rich and the elite, and I loved your metaphor of the
#
plane called India and these people are flying first in business, but it's going to crash.
#
But the thing is, how the rich and the elite have been getting so far is that they've insulated
#
themselves from the state.
#
They've done their job.
#
And they've got everything worked out.
#
And now they realize, damn it, we're in the same plane together.
#
So just as they've insulated themselves, a lot of these problems, it seems to me, of
#
the state, I don't have much hope for much happening from the state.
#
At most, you make incremental marginal improvements.
#
Now, you had spoken to me in the episode where you give so much insight on education.
#
You said that at the level of the state, when you're talking of that kind of scale, even
#
a marginal improvement can touch hundreds of thousands of millions of people, in fact,
#
and therefore it's worth it.
#
But you know, it kind of strikes me in this dark moment that we can't have hope from the
#
If things get better with the state, that's fine.
#
It will happen over time.
#
But private individuals have to kind of step in.
#
And therefore, the larger question, which perhaps I should have asked after we discussed
#
all our health care problems and so on, but it seems like an opposite time to ask it,
#
is that when we talk of the problems in the Indian state, in our health care, for example,
#
you know, and we say that, of course, the state has to do it.
#
And I agree with you there, that, you know, the positive externalities if everyone is
#
healthy are so huge that, of course, the state has to do it, especially in a poor country
#
But if the problem is not getting solved from there, then private people have to step in
#
and find ways to do it.
#
What are your thoughts on this?
#
I'm sorry if I rambled a bit.
#
The question is absolutely spot on.
#
And see, the book, for the most part, is addressed to senior policymakers, right?
#
Because that's kind of the audience that has the capacity and the locus standi to actually
#
But my final chapter on making it happen, right, is essentially about state, citizen
#
and civil society, that's kind of speaking to some of these points that you're talking
#
There's a chapter that's not yet written, I have an outline, but precisely because I
#
think it's evolving so much as we see, right?
#
So India does have an incredible tradition of civil society.
#
Yeah, you know, there's no shortage of very, very well minded, well motivated, sincere
#
organizations trying to do a lot of good.
#
And I do think there's a lot to be a lot to be said that now I think the challenge is
#
what I've seen in the space is that there is a lot of distributed subscale initiatives,
#
That you have to do in different places and thinking about how do you leverage that general
#
angst that citizens have that they want to shape their surroundings for the common good?
#
How do you take that and channel that enthusiasm and desire in some in a constructive way?
#
So one of the big things that's happened between the last podcast and this is like I have gone
#
back and heard so many more of your episodes and the lockdown has helped me in the sense
#
that the only exercise I have is long walks and and you know, so just as an academic always
#
makes cross references.
#
The first time I came on the show, I hadn't really heard many of the past ones now I have
#
heard said make connections to a lot more of your previous episodes here, right?
#
But I think Ashwin had multiple episodes has said this incredibly I think powerfully, right?
#
So start local, but instead of saying that I am going to go and create some organization
#
that does something if you go sit with your local municipal council, if you go sit like
#
you know, with your local public health guys, the guys are doing the suites and then understand
#
how can you leverage their existence as opposed to try to do something in parallel, right?
#
Because the thing when you do something in parallel is you have so I mean, this is a
#
complete digression, but I guess we've done a one hour digression in the past, but you
#
know, and one of the things I'll talk about in that last chapter is the there is a range
#
of philanthropic civil society actions, right?
#
There are organizations like say the Red Cross and Doctors Without Borders and you know organizations
#
and you had a show on feeding the hungry right mean during the pandemic, right?
#
So the the organizations that do what I call immediate humanitarian relief right mean alleviating
#
Okay, and that's incredibly important in the moment because if you can't survive that moment
#
in the long run, you're dead.
#
Okay, so but that's very, very important.
#
That's one kind of civil society, right?
#
Then there are civil society actions that say let me come and build a school or build
#
a hospital or build something where I want to build a certain center of excellence and
#
I will cross subsidize access to the poor and you know, maybe I will create something
#
that the rich are willing to pay for and there are different models of that and the reason
#
people do that is because you're well minded.
#
You want to do something institutional as opposed to just give immediate relief, but
#
you know that if you were to just fund the state, then you have it goes into the morass
#
of the bureaucracy and so you want to preserve a certain amount of operational autonomy on
#
your school or your hospital and that you can run into high standards.
#
Okay, like you know, so that's another model of civil society, right?
#
Then there are models of civil society action that focus a lot more on citizen awareness
#
Okay, and which I think is again very, very important and this has to the extent that
#
a big part of the dysfunctional state is that the people that is meant to serve are often
#
not empowered enough to hold it accountable, right?
#
I mean that there are, you know, a lot to be said for that, okay, but I think here is
#
Again, the problem with this book right now is that each chapter could well be a book
#
I mean, and I have this longish academic discussion of why has democracy mean per se not delivered
#
for India in the way that we would think it should, right?
#
And the argument is quite subtle, right, because it's basically democracy has helped in terms
#
of expanding the ability of people to make claims on the state, okay?
#
But what it has not done is expand the capacity of the state to deliver on those claims, right?
#
I mean, so when you go in in activist mode and saying I'm going to kind of teach people
#
their rights and go and ask them to kind of, you know, mobilize better, that's useful,
#
but it very quickly hits the capacity constraint of the state itself, right?
#
Because essentially the state has so limited resources in terms of budget and bandwidth
#
that you're just kind of moving the attention from the place that's shouting less to the
#
place that's shouting more, okay, like, so the real impact will come when you can expand
#
the supply and kind of enhance the capacity of the state to deliver and the state is a
#
metaphor really for all of us, right?
#
So and again, I'll elaborate on this a lot more in terms of both what individuals and
#
But my broad thinking right now is that ways in which we can leverage the existing state
#
and help it work better are going to give you the highest ROI and not everybody is going
#
to go sit at the state level and start changing policy, but in your local, in your city council,
#
you know, to the extent that that itself is a collective good, right, I mean, that could
#
be a very, very good place to start.
#
So again, you know, this was unplanned.
#
I thought we were going to talk about health, but we always have this habit of doing digressions,
#
but hopefully some of us.
#
I mean, my whole modus operandi of learning about the world is thinking aloud and having
#
people either, you know, fill in the gaps in my knowledge on or add to it and so on.
#
So I'll take another digression from this digressive digression, which is that you mentioned
#
that our democracy has worked in terms of making people feel empowered enough to make
#
demands of the state, but it hasn't worked in the other way of the state having capacity
#
to actually deliver on those demands.
#
So my question is that the reason the state hasn't developed that capacity is also a question
#
of incentives because our government is not local enough, that if our government was incredibly
#
local, that if, you know, I had done an episode on, I think a very old episode with Shruti
#
Rajgopalan on local governance and all of that, and she pointed out that if you just
#
look at Maharashtra, right, the person that you vote for, your municipal councillor or
#
whoever has practically no power to do anything to improve your life.
#
And the person who does have all that power, the state government and all that, they don't
#
Their vote bank is something else.
#
So there is a disconnect between power and accountability.
#
And the more local that government gets, the more accountability there is.
#
And therefore, if they were accountable to you, they would build the capacity to serve
#
So what I'm going to do is I'm going to punt on this to our future episode on federalism.
#
And because I think last time we said, you know, we put teasers of multiple future episodes.
#
So we'll do health today and then we'll, but I will come back to federalism some other
#
But before we get to health, let's kind of get to your book.
#
Tell me a little bit about your book, the process of writing it, the process of kind
#
And also the interesting, the other interesting aspect, which is that does teaching teach
#
you in the sense that if you think about there being like, I think in an earlier episode
#
with me, Krisha Shouk had spoken about this pyramid of learning where, you know, your
#
learning is more shallow.
#
If you just listen to something, it's a little deeper if you read it and so on down the way
#
and it's deepest if you teach it, I think.
#
So in your case, you mentioned that one of the ways you workshopped the book was you
#
also taught bits of it, did the process of teaching it, having to pair everything down
#
to a systematic structure, explain it in simple language, answer questions on it, did that
#
deepen your understanding and in general does, you know, through your career, you've obviously
#
been an incredible researcher.
#
That's a matter of public record, but you've also taught.
#
So what's the relationship there between teaching and learning?
#
There's a reason why top research universities have this positive feedback loop between research
#
So there's no question about it that the act of putting together a lecture and teaching
#
forces you to distill both what you want your audience to understand and how do we know
#
I'm forgetting his name now because his handle is Bhalo Manoj, right?
#
But like, you know, four episodes.
#
But what Anirban mentioned, right, like, you know, that when he got to the US, his first
#
exam, he bombed it because all his life he'd been taught how to spit out the facts as opposed
#
to kind of ask where do those facts come from, right?
#
And I think one of the first things I say is like, you know, your typical rote learning
#
system teaches you to answer questions.
#
It doesn't teach you to question the answers, but good teaching is all about not just teaching
#
you something, but teaching you principles, teaching you how to think, right?
#
I remember I think in my first class, one of my freshman year classes, I think it was
#
a chemistry class, I think being taught by, I can't remember it was, yeah, it was either
#
a Nobel laureate or somebody of similar eminence where he kind of said, he said, listen, you
#
know, you've gotten here essentially by solving exams, but, you know, when you get into the
#
lab, right, I mean, there is no script, right?
#
There is no script because you're by construction facing a new problem, right?
#
I mean, and so what you're going to get is just a set of principles of how you're going
#
to approach and navigate when you're in uncharted territory.
#
So I think coming back to good teaching, what I did with this particular book teaching was,
#
to be honest, the first draft, which I taught, it was, I committed to teaching a seminar
#
class mainly as a commitment device to myself, to force myself to get that first draft done
#
because I needed to give the chapters to the students while teaching it.
#
So it was a seminar class of about 16 to 18 students and book has 18 chapters and the
#
So one intro, one conclusion, and then for each lecture, I would basically assign them
#
a chapter 72 hours in advance, and then students would read it and submit a two page response
#
paper before the class.
#
And I just wanted them to basically have three categories of notes.
#
First is the three to five things that they learned, the most important things they learned
#
The second is what are the things they're still confused about or unclear about.
#
And third is, you know, what are the, so ideas they have for strengthening the chapter, whether
#
it's with facts or some students had some ideas for cartoons, which, you know, I think
#
will be great when I get a chance or just, you know, other ways of making the points
#
So I required that before the class.
#
So that was a commitment device that the students had to do the reading.
#
So then coming to effective teaching, you know, you, the best teaching is not when you're
#
repeating the lecture, but if people have done the reading and then you come from the
#
beginning with a set of questions and you're engaging in the discussion.
#
So I think that was a lot of fun.
#
Now, like I said, the first draft of the book is still very, very, very terse because essentially,
#
you know, I'm just trying to synthesize a huge body of research and get it across, but
#
what's going to take a lot of time in the editing and where the student feedback will
#
help is to say, okay, now how do you simplify, simplify, simplify?
#
Because my hope is that this is a book that is both essentially readable for your, you
#
know, airport reader and is viable as a textbook in any class in kind of public policy.
#
So the challenge is, of course, there's lots and lots and lots of books in India written
#
kind of in a more experiential way.
#
And I think what hopefully makes this book different is just how much research there
#
So you kind of want to convey the depth of the research without necessarily getting
#
people bogged down in the details.
#
So the challenge is to then put those in footnotes, perhaps create like an online appendix to
#
the entire book that will then have the studies and some of the more detailed technical pieces
#
there so that that becomes an accompaniment to people who want to teach with it.
#
But that's kind of where I am right now.
#
So I think I'm about 95% of the way done.
#
But like we said last time, you know, the podcast is so much fun because I can jump
#
and make connections, right?
#
Because the way the book is structured is set up around a set of teams and then a set
#
And the hardest part is there is so many cross connections across everything in the book,
#
but the book by construction is linear.
#
So how do you make each chapter fully self-contained while at the same time not repeat what's in
#
So in writing, say an education chapter, a health chapter, right?
#
I mean, if I say, go back and look at this chapter and state in the market about how
#
do you regulate, go back and look at this chapter on measurement, go back and look at
#
this chapter on federalism, you know, then it's kind of, anyway, so that's, I think,
#
I think somebody transcribed that education podcast and came to about 40,000 words.
#
And so sometimes I really wonder why am I wasting three years writing a book when I
#
can do four podcasts with you and be done in like half a day.
#
No, it's not a waste because I've read one chapter of it and I cannot wait to read the
#
I think, you know, one of the things on a very serious note, I'm glad that you're writing
#
this book that Ajay and Vijay Kelkar wrote in service of the Republic.
#
I think, you know, these books together and hopefully there'll be many more like them
#
will form a body of work, which a student of public policy in India will, or anywhere
#
for that matter, will be able to look at to get different sort of foundational understandings,
#
which are really important.
#
And I'm just thinking aloud, you spoke about how, you know, the book is so interconnected
#
and how do you express those interconnections?
#
And I've been taking all my notes for my episodes recently in this app called Roam Research,
#
which has something called bi-directional linking where, you know, within the category
#
of health, I could be talking about, say, regulation, but regulation itself is a massive
#
category which has five subcategories.
#
So somewhere a subcategory becomes a super category and all of that shit happens.
#
And the software kind of allows for that, but you feel like that maybe there are even
#
more refined versions that are kind of happening in the future and a book is really old technology.
#
I mean, I think what I like about reading books in Kindles these days is because you
#
can click to the references, right?
#
So I think because the academic part of me, when I read something, you know, half the
#
time I want to click the reference and kind of, you know, check that.
#
Yes, so having more digital books that are able to just hyperlink back, right?
#
And so in fact, sometimes I remember a piece of advice I got when I was in the academic
#
The faculty job market is when you go for a faculty job, you kind of get these half
#
an hour interviews where people ask you all kinds of questions and you have no idea what
#
they're going to ask you.
#
So one of the things was what you train your students in terms of preparing for a job interview
#
is to kind of think about their first three to five minute spiel as the front page of
#
And then, but that hyperlinks based on what people questions they're going to ask you,
#
And so you need to be able to drill one, two, three, four, five levels below.
#
So you know, if you're say an econometrics candidate and you're being interviewed by
#
faculty members, one is in macro, one is a micro, one is econometrics, one is in development.
#
The people outside your field will ask some interesting applied questions, but that might
#
be one level deep, right?
#
The people in your field will go five levels deep.
#
And your challenge is not just to answer, but once you finish the answer to come back
#
to the front page and kind of continue the narrative without kind of getting lost in
#
And when you have the experience each year, you read more and more and more that mental
#
filing system expands in a way that you make more and more and more of the connections,
#
which is why I think after 20 years of this work, I felt ready to write this book, right?
#
Because till now, most of the work as an academic is writing deep individual papers and but
#
those papers don't really make the connections across what's happening.
#
And so the book is a chance to make those connections.
#
Now, that being said, the linearity of the written word is still quite vexing, right?
#
I mean, but you know, it is, it is, it is what it is.
#
That's, that's absolutely kind of fascinating.
#
Now tell me about how healthcare, because, you know, if one goes on YouTube and searches
#
for Karthik Muralitharan, you know, besides, of course, our episode together, you'll also
#
find a lot of talks you've given on education.
#
So I was trying to kind of make sense of this, that, okay, he's an expert in education.
#
How has he also gotten into healthcare?
#
And then I remembered something that you told me in our last episode, where in the introduction
#
you spoke about how the first person you named as an intellectual influence on you was Amartya
#
And Amartya Sen, of course, build this capabilities framework about what individuals kind of need
#
to reach their full potential.
#
And that had two prongs to it.
#
And one was education and one was health.
#
So it was like, voila, okay, that's why.
#
So I hope I'm not being too flippant, but tell me a little bit more about how you got
#
into it, why you got into it, what your journey doing health economics was like, because one
#
of the things that I've kind of noticed is that the body of work that you have is exciting.
#
The body of work, all of these other economists who've done all of these papers when seen
#
as a cumulative whole is exciting.
#
But each individual paper seems to be, you know, like a lot of kind of stodgy legwork
#
and a lot of just gathering data, gathering data, gathering data, just to say that, you
#
know, so many people are absent, you are doing such a long study over such a period of time
#
when you are, it's common sense, they are absent only, you know, so how does that process
#
And in a sense, you just described your mental process where you do a lot of little concrete
#
micro things, and then they begin to form a bigger picture.
#
And it's to illuminate that bigger picture that, you know, you're writing the book.
#
But how has that process of being a health economist been like for you?
#
Why did you get into it?
#
Like, if you are interested in the big questions, and how do you figure out which are the small
#
things to dive into first, or is it happenstance, take me through a little bit of this, because
#
the world of academics, you know, both fascinates me and intimidates me, especially now when
#
you talk about at an interview, you have to go five levels deep and come back up.
#
And I'm like, what is this, you know, I'm just a person who you would have noticed I
#
just go deep, there's no front page.
#
So they tell me, yeah, and thanks for asking that question.
#
Because the truth is, I'm not a health economist, and I shouldn't pretend to be one, right.
#
But health, thinking about health permeates a lot of, you know, my larger thinking and
#
teaching and some of my research as well.
#
So I think one of the things I want to do for your listeners is partly because I see,
#
you know, I really liked Ajay's episode, and I see this almost as building off on that.
#
But he said, you know, one of the things we need is really building a community of young
#
researchers and scholars in health research community.
#
So I see part of my goal here is almost excite, say, college goers or master students to say,
#
what a wonderful, you know, amazing field health economics is and how much there is
#
So I think in terms of my own personal entry points into health, they come from three angles,
#
So the first is public finance, right.
#
So at heart, I'm a public finance economist, right.
#
So my core fields are public finance and development.
#
And in public finance, right, health ends up accounting for over 20, 25% of public spending
#
in developed countries, and which is, you know, where I've studied.
#
And so when you take public finance in the US, a big part of your coursework is in fact
#
health economics, it's almost becomes like a subset, because that's where the government's
#
They follow the money, right.
#
You know, so if, and in fact, one of my advisors, so Michael Kramer was one of them, but the
#
other one, I had three of them, but the other one was Marty Feldstein, who was one of the,
#
you know, the giants of US public finance and also one of the people who kind of almost
#
created the modern field of health economics, right.
#
Because I think just like Ajay was saying, there used to be a tradition where health
#
was left to doctors and education was left to educators, but policy is fundamentally
#
about allocating scarce resources.
#
And so you kind of need an economic lens in this, right.
#
So the public finance aspects of why healthcare is, you know, occupies a large share of my
#
mind space is A, that there are fundamental questions about how much public expenditure
#
I mean, what should we be spending this on, given that there are other competing claims
#
on public resources, you know, how do you increase and optimize the effectiveness of
#
your public expenditure?
#
So the unifying thread in all of my research is really improving effectiveness of public
#
And like I said last time, the sectors I work on are sectors where the government spends
#
over 10,000 crores a year, right.
#
Education, health, NREGA, PDS, now ICDS.
#
So essentially improving then the effectiveness of that by even 1% is going to give you a
#
huge bang for the buck, right.
#
So my first kind of entry point into health economics is to public finance.
#
And I'll come back and talk about what we've learned from that approach.
#
The second one is development economics, right.
#
So that is my core field.
#
I teach both undergrad and PhD classes in development economics.
#
And you know, health is a substantial part of what you teach.
#
And you know, going back to Amartya Sen, so both health and education have this unique
#
place because they are both what we call intrinsically important and instrumentally important, right.
#
So the traditional economist Chicago view of health and education is human capital,
#
It's human capital because these are investments that make you more productive.
#
So a healthy worker is more productive, a more educated worker is more productive, right.
#
Whereas the Senian inversion, so to speak, right, mean is that health and education are
#
not just about making you more productive, but these are the intrinsic goals of development,
#
Because people care about living healthy lives, regardless of whether you're productive.
#
It is good to be educated and be a more sophisticated consumer of the world around you, regardless
#
of whether it makes you more productive, right.
#
So in that sense, health and education occupy like a special place in development because
#
of both this instrumental and intrinsic importance.
#
So if you go look at, say, the human development index, you know, with Sen inspired, essentially
#
what it's doing is saying the traditional measures of development are just GDP per capita.
#
But the GDP, you can't eat GDP.
#
The whole point of GDP is to kind of live a better life.
#
And health and education are intrinsically so important that the HDI then has a one third
#
weight on health, a one third weight on education and a one third weight on GDP per capita.
#
So health becomes central in development economics that way, but also in two other very important
#
I think when you think about development economics, what is development economics, right?
#
I mean, how is development economics different from just being about reducing poverty, right.
#
And so the conceptual key point about development economics is the idea of poverty traps, right?
#
The idea that people can be in initial conditions that make them so vulnerable that they cannot
#
bootstrap their way out of that initial condition.
#
So one of the contrasts, I would say, between public finance and development is when you
#
do public finance in the US or in any developed country, the de facto understanding you have
#
is that there is almost always a trade off between equity and efficiency, that you want
#
to do some redistribution because you want to support the poor, but that redistribution
#
will almost always reduce efficiency in the economy.
#
It's because to redistribute, I need to raise taxes on the rich and that's going to reduce
#
I'm going to give out something for free, which may reduce incentives to work.
#
And then the most kind of, I think important, but underappreciated part is that if I'm targeting
#
something to the poor, okay, then by construction, I need to phase out that benefit as you get
#
richer, right, because I'm giving it to the poor.
#
So at some point, as you earn more, I have to phase it out, correct?
#
But that phase out often creates a very high marginal tax rate for the poor, right?
#
So often for every dollar I earn, I might be losing 50 cents in benefits, okay?
#
So suppose that $20,000 or less, you're getting free health insurance, okay, through Medicaid
#
and then you earn 30,000, you become ineligible.
#
Let's say this benefit is worth $5,000, okay?
#
So effectively when I'm going from 20,000 to 30,000, I have a marginal tax rate of 50%
#
because I'm losing a dollar of 50 cents of benefits every dollar I'm earning.
#
And then in a funny way, that can actually create a poverty trap by itself because now
#
you've reduced the incentives to work out of that place of poverty.
#
So the general thinking in a lot of public finance and public economics is that there
#
is this inevitable trade-off, but because we care about equity intrinsically, right,
#
Nina, as a society, you care about some redistribution.
#
But what is, I think, very nice in development is that we have a whole class of models where
#
you show that when people are close to subsistence poverty, that there is often no trade-off
#
between equity and efficiency, that a well-designed social welfare program, a well-designed intervention
#
can not only improve equity, but also improve efficiency, that you're making people more
#
productive and therefore able to pay back.
#
So for example, there's a brand new paper which I just released this week on the ICDS,
#
which is the Integrated Child Development Services, and at one level it's a very simple
#
It's a very simple paper because we're just looking at what is the impact of adding an
#
extra worker, okay, focused on preschool education.
#
And in fact, some people said, you really need an RCT to go study this.
#
The point is not does it have an impact.
#
The point is to quantify the impact and show that the public rate of return on this is
#
about 12 to 20 times the cost.
#
And therefore, if you would invest in this, you will increase productivity enough that
#
your future discounted tax revenue increases will more than pay for the program.
#
So I think the point is in a world of limited resources, when everybody wants funding for
#
everything, the holy grail for a policymaker or policy analyst is to think, can I find
#
a class of policies that improve both equity and efficiency so that I am both supporting
#
redistributive justice, but doing this in a way that is not just about distributing
#
a fixed pie, but doing this in a way that expands the size of the pie, right?
#
There are a class of policies that do that, but you need to do a lot of heavy kind of
#
lifting in terms of both thinking and empirics to kind of design and implement those programs,
#
See, but again, the point is that's why it plays such a big role in development economics,
#
thinking about health, because the idea of a nutrition-based poverty trap or health-based
#
poverty trap are some of the oldest ideas in development economics.
#
Now there is a third strand in development, which is very, very kind of exciting, which
#
is when you think about development economics, in general, you expect poor countries to be
#
able to grow faster than rich countries.
#
And there's basically two reasons for that.
#
One is you start with a much lower capital stock, okay?
#
And so the returns to marginal returns to capital are higher, okay?
#
So the U.S. Treasury may be giving you 1%, 2%, because this is a capital-rich society.
#
You don't have those same high-return opportunities, whereas the returns to capital investment
#
So automatically you would expect poor countries to grow faster.
#
But the other big reason why you would expect poor countries to grow faster is that it's
#
much easier to adopt an existing technology than to invent new technologies at the frontier,
#
So if you think about just this corona crisis, right, I mean, the reason we are even in this
#
place of talking about vaccinations is because you've had the entire research infrastructure
#
of the developed world, right?
#
Like, I mean, go after this kind of this virus and crack this in about a year, right?
#
So in that sense, for a development economist, in a way, the advantage of being away from
#
the frontier is that you just have to focus on adoption.
#
But then the big question is, why are there so many kind of free lunches left on the sidewalk?
#
Why do people not adopt existing technologies?
#
So if you look at health, you know, you can talk about how do I manage a healthcare system,
#
but the lowest hanging fruit is why are people not taking their vaccinations?
#
Why are people not taking their ORS?
#
These things that are incredibly inexpensive, incredibly like, you know, high returns, and
#
that's kind of free money that you're leaving behind.
#
And so that's kind of created a lot of kind of, you know, growing interest in behavioral
#
development economics and behavioral economics, and particularly, I think, health-seeking
#
behaviors has been a very, very active area for, you know, behavioral economics to come
#
So that sums up my own journey in health as the public finance piece, as the development
#
piece, and then there's the service delivery piece, right, which is kind of how my first
#
research project in this area was what I did in 2003 in this nationwide study of both teacher
#
and doctor absence, right, means it was a study on public service delivery.
#
And so I also got into health, partly thinking about how do you kind of design public delivery
#
systems across both health and education.
#
Now, so that's in terms of the pieces of the puzzle, which I have direct research experience
#
There are many other parts of the puzzle, right, where we can talk about, again, in
#
the interest of just exciting your listeners about this as an area of study, right, of
#
We can talk about health in kind of the, of labor economics, of economics of innovation,
#
economics of regulation, market failures, market structure, you know, industrial organization.
#
There's health everywhere.
#
So I can take a quick pause here, and then I'll come, but I do want to give you listeners
#
maybe one or two really exciting papers in each of these subfields to give you a sense
#
of, you know, just how much fun people are having in kind of research and how much good
#
stuff that is to be done.
#
So I'll ask you for those papers.
#
And by the way, the show notes will contain enormous amounts of masala.
#
So kindly do check out the show notes after the show, not during it, because just concentrate.
#
And you know, I loved your line, you can't eat GDP.
#
So if some listener of the show would like to make t-shirts out of that or whatever,
#
you know, feel free to do so, you know, maybe you can put Karthik's face on it.
#
Yeah, I think the show is, I think the meme from last show, I mean, if there's one thing
#
which maybe I would like a little bit of copyright too, is this line on consultants having confidence
#
and academics having confidence intervals.
#
I think enough people like that, like, you know, so maybe that's a remarkable line.
#
That's a classic t-shirt line.
#
And now you can't eat GDP is like an incredible line.
#
I'll just quickly go over a couple of the concepts, sorry, and coming back to popular
#
Like, you know, in Godfather, right?
#
It's like throw the gun, keep the cannoli, right?
#
Like, you know, so it's the food that you keep in a crisis.
#
So I'll quickly go over a couple of the concepts you mentioned, thinking aloud.
#
You can tell me if I'm stating it correctly or not.
#
First of all, you know, Karthik mentioned the term poverty trap and the classic example
#
of a health poverty trap is let's say that you are poor because you are poor, you are
#
malnourished and you are stunted, right?
#
And because you are malnourished and stunted and you're not getting enough nutrition, you
#
don't have the kind of cognitive development you would otherwise have.
#
You don't have the kind of energy you would otherwise have to do stuff.
#
And therefore you stay poor because you can't work, you can't get ahead and it becomes a
#
vicious cycle you can't get out of.
#
Now there is tremendous value to society in getting you out of this vicious cycle.
#
And that is because of, you know, my mind goes back to this term that the economist
#
Julian Simon once used.
#
He wrote a book called The Ultimate Resource where he spoke to the human being as the ultimate
#
resource and this again argues against this, you know, false Indian notion of our population
#
Simon's point was our population is our greatest strength.
#
All the things that we ascribe to overpopulation are actually issues of bad governance, that
#
every individual has a potential to give more to the world than she takes out.
#
But for that you have to cross a minimum level of subsistence where you can actually do that.
#
You know, if you're caught in a poverty trap of any kind, such as this health based vicious
#
circle where you just can't get out of it, then you can't reach your true potential.
#
And if you reach your true potential, it helps everybody.
#
It's like a positive externality.
#
So in that sense, you know, the trade off between redistribution and growth doesn't
#
come into the picture because this kind of redistribution actually aids in growth because
#
people are reaching their potential and helping everybody else also as they do that, because
#
it's a positive sum game.
#
We all sort of grow together.
#
Now we can get to the papers that you'd like to recommend.
#
And that's exactly right.
#
And in fact, you know, let me take a slight tangent.
#
So let me pose a quiz to you, okay?
#
So the definition of the poverty line in India, okay?
#
So the definition back in the day used to be, do you have the income needed to consume
#
a minimum amount of calories?
#
It was just the minimum caloric content of say 2000 calories a day.
#
Now, so the question I have for you, this is the puzzle is when you have a poverty rate
#
at independence of over 50% or 60% or even now of say 20%, that means by definition,
#
these are people who are earning less than the minimum amount of calories needed to survive.
#
So the question is how do people survive?
#
If the definition is the minimum amount of food needed to survive is then how do you
#
survive below the poverty line?
#
I mean, one could just argue that you're not surviving as a human should, that you're being
#
forced by circumstances to operate on some, on a level below what a human can operate
#
So you're starting to get warm, right?
#
Because you know, people will often say, okay, maybe they're getting transfers from others
#
The reason that doesn't fully work because the NSS measures based on consumption and
#
And it also includes home production.
#
So the food you produce, but the answer is broadly what you allured to, right?
#
Which is the caloric minimum is the caloric minimum for a person with a healthy BMI, right?
#
I mean, but what the body does when you don't have that nutritional inputs early on is you
#
just don't grow to potential, right?
#
And so by reducing the size of the body, you reduce kind of the caloric needs needed to
#
maintain it, but that comes at enormous long-term costs in terms of both the cognitive development
#
as well as, you know, other aspects of higher likelihood of physical morbidity, right?
#
And so, yeah, we'll talk about nutrition if we have time in a way, because that's likely
#
different from health care, but you know, in some ways I would say that's almost one
#
of the most important kind of health investments we need to make.
#
So yeah, I think, you know, just stepping back about health economics is a field more
#
broadly is not just in terms of papers, right?
#
But just to give your listeners a sense of, and again, building off on some of the things
#
Ajay talked about, about, you know, what makes health one of the most vexing areas of kind
#
of economics and of policy is it goes back to the fact that your basic market accountability
#
mechanisms often break down, right?
#
And Ken Arrow, you know, one of the most brilliant economists of the 20th century, I think has
#
this famous paper where he says, you know, the laissez-faire solution for health care
#
And that's essentially because you're in a situation of what's called a credence code
#
where you don't know what you need and the doctor has to both tell you what you need
#
So there's all of these asymmetries of information.
#
There is the fact that comparison shopping doesn't work when you had an accident and
#
like, you know, you're looking for a hospital, you don't go there and bargain like, you know,
#
and saying, you know, what the price of your life-saving treatment should be.
#
And so which means your standard market metaphors, you know, break down, okay?
#
So and then pricing, okay?
#
So pricing in any market is, as we know, it can sit anywhere between cost and value, right?
#
So there's this huge wedge.
#
But if you're in a moment of very inelastic demand, right, I mean, the ability to charge
#
essentially your last living penny, right, I mean, is very, very high, right?
#
So there are a whole bunch of reasons for why unbridled market kind of, you know, conditions
#
don't work, which again makes this a very, very, very active area of thinking about what
#
would be the right kind of policy frameworks to allow you to mitigate some of this because
#
there are problems with everything.
#
That's, I think, the really hard part about health, right?
#
So which is why I wouldn't even pretend to say we have solutions.
#
So in fact, I'm reminded of my first PhD lecture I attended in macroeconomics, right?
#
So this is by Greg Mankiw, who I think, you know, later went on to be CEO to the president
#
and he started his lecture by saying, you know, macro is a difficult subject.
#
He said, you are confused and I am confused, okay?
#
The only difference is you are naively confused and I'm profoundly confused.
#
So my goal, perhaps at the end of just the discussion on health is to make people profoundly
#
confused, okay, so that you're at least asking the right questions.
#
But coming back to, you know, other areas of what makes health economics both so interesting
#
and so vexing is let's think about the economics of innovation, right, like, you know, so,
#
and you're seeing this battle play out right now, which is how much, what kind of patenting
#
regime should you have, right, to both serve as an incentive for the people who are innovating.
#
But then conditional on having the innovation, it is massive deadweight loss to then charge
#
these high prices because the marginal cost is obviously very, very low, right?
#
I mean, so how do you kind of both set up structures to get the innovation and conditional
#
on the innovation make that access, you know, to everybody, right?
#
And I can give you an example of specific solutions, right, I mean, that are worth thinking
#
Then you can think about the economics of regulation.
#
So I have a student today, like, you know, who will be presenting in our lunch workshop
#
in about four hours on asking about is the FDA too conservative in its regulatory protocols,
#
because there's this trade off of type one and type two error, right?
#
And so because bureaucrats tend to be risk averse that one adverse event is much more
#
costly than the potential cost of not allowing a positive drug to enter the market because
#
So, you know, he's got some very nice evidence now in the context of a certain class of therapeutics
#
to then show that it looks like they may be getting too conservative.
#
And then he's got another paper looking at, you know, regulation versus litigation as
#
a way of enforcing good behavior.
#
So again, you know, these are just huge areas of research.
#
In fact, coming to economics of innovation, there's a very nice paper and this is these
#
are tangents, but I think it's fun to think about.
#
So one of the big misallocation of public expenditure in my view, okay, say in the US
#
is that how much more money we spend on end of life care in the last year of life, okay,
#
compared to how little we spend on say early childhood care, right?
#
When you have an entire life of 80 years to get the returns to that investment.
#
Now, for the longest time, I used to think that there's a very simple political economy
#
explanation for this, right?
#
Which is that children don't vote, okay, whereas your retirees are incredibly like a reliable
#
And so the American Association of retired persons is an incredibly powerful political
#
And so, you know, by construction, you've created political incentives, right?
#
You need to spend much more in the ending phase of life, like, you know, than for young
#
children, though, as a society, you would benefit much, much more from those early life
#
But it turns out that there's this very, very nice paper that is asking this question about
#
why do we have so much more medical innovation in end of life care, okay, than medical innovations
#
that say at the life of 20 or 30, that if you got this new medication, a new drug would
#
dramatically improve your life for 60 years, okay, but we still get much more innovation
#
So why do you think this might be happening?
#
Because the end of life people would be a more powerful lobby.
#
So that was my initial initial hypothesis.
#
And that is certainly part of the story.
#
But it turns out that there's a much more basic scientific explanation that has nothing
#
to do with the incentives in terms of the political incentives of lobbying.
#
Well, there is also the seen in the unseen in the sense that if you create, say, drugs
#
at 20, which make me, you know, live 10 years more at the end of my life, the effects are
#
Whereas if you create something which will make someone who's 79 go to 80, the effects
#
So yeah, so you're starting to get very, very warm, okay.
#
And so again, this I'm very impressed because I didn't give you any hint that I was going
#
to talk about this or this paper.
#
But yes, so the basic point is this, right, that if you look at any clinical trial, clinical
#
trials have to designate what they call an endpoint and saying, what is the outcome we
#
So for some of, so one of the reasons these different vaccine trials are not strictly
#
comparable is that the designated endpoint was different, right?
#
So for some of them that the outcome they were looking at was symptomatic cases, others
#
they were looking at hospitalizations.
#
But if you think about a pharmaceutical company that's coming up with a new drug, right, the
#
holy grail of efficacy of your drug is if you can show impact on mortality, okay.
#
If you can show that this had an impact on reducing, increasing life expectancy by even
#
three months, you're golden because the value of a statistical life is kind of several hundreds
#
of thousands of dollars.
#
And if you show that I extend this by three years, three months, then you have the ability
#
If I have an intervention at age 20, that's going to increase your life expectancy.
#
I have to wait 60 years like, you know, for the study to be completed because essentially
#
people you're not dying at the same rate, okay.
#
But like if you have a universe of 85 year olds in your study, you will reach your endpoint
#
in your study within five years.
#
So ironically, one of the reasons the vaccine trials got done much faster with Pfizer Moderna
#
was because the case rates were exploding.
#
So which meant that the endpoint was being reached faster in the control group relative
#
So an example of how subtle and this is a lovely paper by Eric Budish at Chicago and
#
Heidi Williams who used to be at MIT and is now at Stanford.
#
So, you know, and it kind of discusses these time horizon aspects of what determines incentives
#
But the reason I'm mentioning this is that, as you've said in previous podcasts, one of
#
the great miracles, despite the pains of this pandemic has been how science has kind of
#
stepped up to the field.
#
But the point is, what are the institutional incentive structures for science, right?
#
And what determines research, what determines what gets research, what determines what gets
#
So all of those are also questions where economists can come in and think about the incentives
#
because essentially economists, the two core concepts, I would say there's three core concepts.
#
One is scarce resources, like I mean, how do you kind of optimize and maximize value
#
Second is incentives, incentives, incentives, right?
#
Anytime something doesn't make sense, you unpeel the onion and saying, okay, why are
#
these incentives of the individual agents not aligned with the macro good?
#
And then the third is, you know, empirics and data and causal analysis and causal inference,
#
which is the problem, particularly in things like health is theoretically, I can argue
#
any issue both left and right.
#
So and which is why I think the empirical turn of economics in the past 20 years is
#
or 30 years has been such a good thing because we're now able to take a lot more of this
#
abstract theorizing and come down to brass tacks and saying, okay, now let's look at
#
the actual data and saying, which side do these trade-offs go off?
#
So this is the kind of why economics in any policy area, right?
#
I mean, in fact, Marty Feldstein, you know, in my last few years before he retired, like
#
was trying to create a whole new field of the economics of national security.
#
So an even bigger black box of public funding is what we spend on defense.
#
So one very, very, very simple metric of how little accountability there is or efficiency
#
there is in defense spending is if you look at the budget share of the army, navy and
#
air force in the U.S., okay, this is basically for the most part been unchanged since World
#
And it is impossible to believe that the marginal product of your offensive capacity has stayed
#
the same across these kind of land, air, but no, the budgets will go lockstep because it
#
has to be proportional to keep the balance between the powers and the forces, right?
#
So in education and health, we've economists have started doing a lot more in terms of
#
analyzing and understanding the outcomes for something like defense is obviously much,
#
much harder, but I would conjecture that there is such an incredible amount of waste and
#
inefficiency like I mean, and how we spend inside the black box of kind of defense ministries
#
that it is ripe for kind of, you know, more value for money type analysis.
#
But this is just a way of saying why economics is and should be everywhere.
#
Now, of course, you know, sometimes the field gets a little bit of a bad rep because it
#
is, it can be imperialistic, it can be a little hubristic, but I think these three core principles
#
of economics apply to almost any area of policy, which is the resources are scarce.
#
And so the opportunity costs and how do you maximize value for money?
#
Second is incentives matter.
#
And third, you know, let's look at the data, look at the empirics, look at causal inference
#
and then try to, you know, optimize given all of these constraints we have.
#
And this is also why economics is so fascinating that it's a lens through which you can look
#
at all of life and every little bit of the world and it's incredible.
#
Now you've said so many things and I have so many asides to make, which I have been
#
noting down that I will quickly go through the asides.
#
And now we will go back to that prehistoric time almost 20 minutes ago when you were talking
#
And my aside there about how the body adapts to the nutrition is getting is that I remember
#
I made an observation in the 1990s, which when I first made it, I thought, wait a minute,
#
it's just today, it can't be right.
#
And then I saw it again and then I was like, oh shit.
#
And the observation was when I came to Bombay in the nineties, mid nineties, I used to commute
#
And in local trains, I noticed that the people in first class were taller than the people
#
in second class, always without exception.
#
The first day I noticed like whatever, and then you notice it again and then you notice
#
And the difference at that time was actually stark.
#
And it's almost like the visibility of luck, you know, because it's basically your luck
#
where ever people have ended up and it's almost as if it's visible.
#
You don't need to theorize about it.
#
The other aspect which you pointed out and you know, you, you cited arrow saying, you
#
know, laissez faire doesn't work in healthcare for all those excellent reasons.
#
You gave a symmetry of information and all of that.
#
And at the same time, you know, it is such a vague subject that in one of your papers,
#
you actually demonstrate why it doesn't apply to India.
#
So we're going to talk more about that as we go along.
#
Then you spoke about your student who's giving a presentation on the FDA and I'm quickly
#
going to make a book recommendation for my readers, which is a book called Fortress and
#
Frontier in American Healthcare by a dude called Robert Gribois, who in fact won the
#
Basia prize in 2014, the year before I got it for the second time.
#
An excellent book, which kind of conceptually, you know, if you read the book and the concepts
#
of the fortress and the frontier as one way to think about it were sort of quite eye opening
#
And I would agree with whatever your student is about to say in his presentation in the
#
sense that I think the FDA is way too conservative and the lives that could have been saved had
#
they not been conservative or what is the unseen.
#
But that is a subject for another episode.
#
And then you spoke about, you know, the whole patent and intellectual property debate and
#
we won't get into that.
#
But what I will put in the show notes and recommend that my readers check out is Alex
#
Tabarrok, who's a frequent guest on the show, wrote a post recently about why patents are
#
not a constraint in this current situation.
#
And I absolutely agree.
#
They are not a constraint.
#
It's just in Alex's words, you know, virtue signaling for a certain sort of people.
#
They don't solve a problem.
#
But at the same time, what I'd like to also point my readers to is something called the
#
Tabarrok curve, which is about eight years ago.
#
Alex showed this beautiful curve, which demonstrated his point of that even though patents are
#
necessary, if you make them too strong, too stringent, they actually affect innovation
#
and innovation goes down.
#
And Alex's point is that they are too stringent and they are a problem.
#
And he demonstrates this with a beautiful graph.
#
And this is where I love the elegance of economics and math and all of that.
#
So I'll put up all of those links.
#
Let's go back now and talk about at the level of concept, I'm going to ask you to get a
#
And here I'm sure at some level, it mirrors your intellectual journey of understanding
#
And my broad question is this, that how should we think of health care?
#
Like in the sense, if you imagine India when it becomes independent and in one of the papers
#
you gave, there was a startling figure that our life expectancy in 1961 was something
#
like 41 years old, you know, which means we'd both be dead by now.
#
So we've clearly come a long way now in that poverty stricken state that we were in then
#
at that design stage, you know, how should we think about health care?
#
How did we think about health care?
#
You know, what did we get wrong?
#
What is a model to follow because everybody's poor, there's no market as such.
#
And so it's in a sense, it's the state's responsibility, everybody benefits.
#
How should one think about health care?
#
And in your own journey, what were the priors you started with that this is how health care
#
And how did those priors changed over time as you got deeper and deeper into the field?
#
The good news is, you know, like Sherlock Holmes famously said, right, it's a capital
#
mistake to theorize without data, right, you know, so I don't even think I came in with
#
very strong priors, right?
#
I mean, I started my journey in this field just by wanting to figure out what's going
#
So going back to that dirty shoes kind of analogy, right, just like I was doing the
#
work in health in education, I was also visiting PHCs around the country, like, you know, as
#
part of that original field work we did for that absence study and just kind of seeing
#
So I've had less priors in this.
#
So there's two separate questions.
#
Okay, one is let me talk a little bit about my journey in health care research in India
#
and the things that I know from my own work, as well as, you know, people like Jishnu in
#
particular and Jeff Hammer have done a lot more work, in fact, in this area.
#
So I'll draw a fair bit on their work.
#
And of course, a lot of others do.
#
But these are kind of, you know, the economists who I feel are thinking about the systemic
#
incentives, design challenges in the right way.
#
So taking a step back to the origins of the Indian health care system.
#
So there was this committee called the Bhore Committee, because the Bhore Committee was
#
set up in 1943, and this was a Maharashtrian gentleman who was in the ICS and, you know,
#
I think, had held multiple senior positions.
#
And the Bhore Committee took about two years to submit its report on the structure of health
#
1946 is when they submitted the report.
#
And you know, a lot of this follows what one I would call is a broad WHO template for the
#
building of health systems in developing countries.
#
And the general sense is that the market doesn't work for a variety of reasons.
#
And the market doesn't work not just in terms of the asymmetric information, but also simply
#
in terms of locating providers in remote areas or rural areas, right, mean that you just
#
people don't want to be in those places.
#
And so unless the government comes actively and builds out a network of clinics and staffs
#
them with public doctors and nurses, there is no way you're going to get health care
#
And this kind of structure follows a very well established template of saying that you
#
will have, you know, a sub center, so there's the district hospitals, community health centers,
#
primary health centers, sub centers, okay, so you're supposed to have a sub center for
#
about 5000 population, a primary health center for about 25,000.
#
So sub center is supposed to be staffed with a nurse, you know, one and one male multipurpose
#
worker and they are supposed to be like the first line of primary care in a village.
#
The primary health center serves about 20 to 30,000 and is supposed to have at least
#
one fully qualified doctor, I mean one MBBS doctor and, you know, has oversight over the
#
sub centers and then you're supposed to have the community health center that caters to
#
about 100 to 120,000 with a larger number of beds, you know, more specialty doctors
#
and then you've got the district hospital that's supposed to be like your primary point
#
of care in the district.
#
Okay, so, so on paper, this is a perfectly sensible structure, right?
#
I don't think it was different from kind of the structures envisaged for many countries,
#
But the problem over time has been one that, and I don't think this is unique to health,
#
I think the fundamental problem for India and this is again something I discussed in
#
the book in some detail because again, historical political economy is a massive field.
#
In fact, there's a new handbook coming out of like 52 chapters, okay?
#
So I'm trying to synthesize that, that entire literature in like half a chapter, one section
#
But, but I think if you, one important piece of perspective to have is that when you say
#
that people are poor, therefore the government has to do it, that's not obvious because
#
when the people are poor, the government is also poor, okay?
#
So because the government is nothing but the sum of the people, right?
#
I mean, and what is your tax base, okay?
#
So when people are poor, in fact, the government has less ability.
#
So coming back to state capacity and that's because the tax to GDP ratio usually increases
#
as you get richer, okay?
#
And that's because you, it's A, as people are super poor, like you can't even morally
#
Like I mean, because you need the basics you have just to survive.
#
And institutionally, it's much, much harder to tax informal workers.
#
And so, you know, the formalization of the workforce is then very strongly correlated
#
with the rising of a tax GDP ratio, okay?
#
So, so I think the first slight nuance in what you said is when the people are poor
#
doesn't automatically mean the government can do something because the government is
#
And in fact, relatively, the government is more poor when the people are poor, you see,
#
because the tax to GDP ratio is lower in a poorer country, okay?
#
So that's, I think, a first very important point, like, you know, for listeners to understand
#
because, you know, it is just so natural.
#
You go out to rural India, you just see kind of how, you know, abysmal the conditions of
#
say health facilities are, and it's just very, very normal and natural to come back and say,
#
oh, the government should do this, okay?
#
And, and I think the single most important problem, what we don't, if you want to make
#
sense of why the Indian state seems to flounder at everything, the simple answer is we're
#
trying to do too much, okay?
#
And, and the, and if you go look at the long arc of the state over a 500 year period, right?
#
So I characterize the evolution of the state into what three broad phases, okay?
#
That is what I call the security state, right?
#
Which is the primary goal of initial states was national security, right?
#
I mean, save your borders and do law and order.
#
And because the biggest existential threat was you were going to be run over and enslaved
#
So, and one simple way of seeing this is that in ancient states, even medieval Europe, you
#
were spending somewhere from 70 to 80% of the public budget on defense, 70 to 80%, okay?
#
Like, you know, of, of money was essentially going on defense, right?
#
So then it's in the 18th and 19th century that you kind of get the second phase of the
#
state of what I call the industrial state or the developmental state, where for the
#
first time now the state realizes that, Hey, let me take tax revenue and build public goods
#
and build public infrastructure, whether it's roads or railways or ports or public
#
sewage systems, and that these are things that will dramatically increase productivity.
#
And so the return on investment in these public investments is way, way, way higher than simply
#
going and fighting wars.
#
I still need a military so that I'm not taken over, but my ROI in terms of growing my productive
#
capacity is now going to be higher than going and taking over somebody else, okay?
#
And it's not like the older states didn't build roads, right?
#
But the industrial revolution dramatically increases the complementarity with the physical
#
infrastructure because now you have at scale production.
#
And so then the investments in the logistics and the transport and the storage and all
#
of that starts paying off, like, you know, once the scale of production starts going
#
So, and it is that period of the industrial state and massive investments in public goods
#
that is associated then with the escape from the Malthusian trap, right?
#
Where for the first time you start seeing this massive increase in GDP per capita, because
#
till then the 3,000 year history of humanity is essentially GDP per capita stayed mostly
#
So, because the productivity went up, but population also went up.
#
So coming back to your old point about our human beings, are you brains or are you stomachs,
#
I mean, I think is how you had classified it and the Malthusian view is that humans
#
are stomachs and therefore, like, you know, there is this trap and you need to reduce
#
Whereas the Romer view, which is what he won his Nobel Prize for, and it's also in Kramer's
#
famous 93 O-ring paper, right, is that people are brains.
#
And if the innovation is a function of the right tail of your distribution, then having
#
more people automatically means you'll be more innovative, but that requires a certain
#
minimum kind of capacity of the brain.
#
And so it is related because the demographic transition where you move from high fertility,
#
high mortality to low fertility, low mortality is very, very important in giving parents
#
the resources to invest in the human capital of the children, right?
#
So in that sense, it's a bit more subtle, right, that you can't just say I will expand
#
the population and get the innovation if that expansion is coming at the cost of being able
#
to invest in the human capital, right?
#
So, but anyway, but coming back to the state itself, so you get this massive increase in
#
GDP per capita, both because you're increasing the numerator from the Industrial Revolution
#
and because you're reducing the denominator, like, you know, through the demographic transition
#
and investing in the human capital, right, I mean, and so that's why that all those three
#
pieces go together, right, in terms of because GDP per capita is about making people more
#
productive and people get more productive and they have more human capital, right?
#
So that is this big transition that happens in kind of the 19th century, mainly in the
#
I mean, but I think the important point we need to understand is that the modern welfare
#
state, right, I mean, has only been in existence for less than 100 years, okay?
#
So if you look at the data on kind of the fraction of GDP that is spent on welfare programs,
#
okay, so whether it's food assistance, healthcare, unemployment insurance, old age insurance,
#
all of these major welfare programs have only showed up in the world in the last 70, 80
#
years, right, after the Great Depression and particularly after World War II, okay?
#
So which means that, again, I think informally, Arvind Subramaniam and I have talked about
#
this, others have talked about this, but we are just putting a lot of data in the book
#
and documenting this in a comparative way and then showing that, listen, a bulk of this
#
explosion in the welfare state that we kind of now expect that a state should do really
#
happens only after today's Western countries reach a middle-income status, right?
#
I mean, so a simple example of, say, food security is that the US launches food stamps
#
at a GDP per capita of $8,000 PPP, okay, like, you know, which is when you're trying to do
#
India does PDS at a GDP per capita of $800, okay, at one-tenth of that GDP per capita.
#
So now at one level, this is a great moral triumph, okay?
#
So and then the other important point to realize is that these three phases of the state are
#
also very strongly correlated with the amount of democracy you have, okay?
#
So the security state is essentially associated even today with non-democratic regimes, right?
#
You know, you go to Pyongyang, like, I mean, that's your modern security state or you,
#
you know, you go to, so in that sense, that serves the interest of just the pure ruling
#
Now, what is subtle and underappreciated about this phase of the industrial state, okay,
#
is that it is not just that, you know, people decided to do this.
#
It reflected a political arrangement of expanded but not universal franchise, right?
#
I mean, so the extent of the vote in that period is, you know, you're still restricting
#
the vote to the property elite, to the educated elite, right?
#
I mean, so yes, it's more than kings and nobles.
#
So maybe 20, 25% have the vote, but no women, no minorities, right?
#
And so what that means is that, so over time, infrastructure investments are going to benefit
#
It benefits the poor through integrating markets and reducing prices, okay?
#
But in the short term, the returns to those infrastructure investments are mainly going
#
to be capitalized by the property owners in the first place, right?
#
I mean, because the property price, so which means that that is the class that then has
#
an incentive to invest in those public goods, okay?
#
And it is not a coincidence that the welfare state is strongly correlated with expansion
#
of the franchise, right?
#
So when you start getting to universal adult democracy, you automatically then create the
#
political pressures for welfare, okay?
#
Now this is a wonderful thing, okay?
#
So we have study after study after study that just show, listen, because in a way it's easy
#
to look at this industrial state and saying like, okay, see, they managed to do all of
#
this because of limited democracy, but that's coming at a huge moral cost, right?
#
It's coming at the moral cost of being built in the backs of either slavery domestically
#
in the US or a colonial empire in the case of the British, right?
#
I mean, so, and which is why the democracy before development, which is what makes India
#
so uniquely different, is a great moral triumph, okay?
#
And at no point of this discourse should it be suggested that that was in any way responsible
#
But like I said, the moral triumph of democracy then increases our aspirations of the claims
#
we make on the state, but we are trying to satisfy those claims at a GDP per capita that's
#
10 times or 20 times lower than what today's West had when they tried to do their welfare
#
So essentially this is a long-winded way of coming back to your starting point of saying
#
that how should we think about healthcare in a world like, you know, when people are
#
It is that much as you may want to do healthcare, right, like, you know, you often simply don't
#
I mean, because you're starting at such a low base.
#
And so what it does is it raises enormously the premium on cost-effectiveness.
#
So it's not that the state should not be doing things, the state needs to do things, right?
#
But the premium on cost-effectiveness and hard-headed attention to evidence is even
#
higher when you're trying to do this, right, mean of such a low base, okay?
#
So that is more as a matter of kind of framing the context, but I haven't really started
#
answering your question in terms of how should you design healthcare in that setting.
#
But like I said, like, you know, it's just good to frame that historical context before
#
we start getting into what we should do, and that's relevant even today, right?
#
Because yes, our tax-to-GDP ratio is now at about 18%, but that's still half of what the
#
So, you know, every sector-specific activist, education wallahs want 6% of GDP, health wallahs
#
Everybody says like, you know, like, we just need more money, but boss, you're not allowed
#
to ask me for more money till you tell me what you're cutting, okay, like, you know,
#
so which is why a big part of the book is to say that given how much inefficiency we
#
have in public spending today, rather than spending, arguing about increasing the size
#
of the budget and fighting battles with other ministries, if every minister and secretary
#
would turn their attention inwards to say how do I improve the effectiveness of my existing
#
budget, we will be way, way better off as a country, right, like, I mean, than fighting
#
the political fights of who gets to control bigger budgets.
#
So before you get to answering my question, and I loved your aside, a long aside, a couple
#
of aside taking off from the aside, which is that, you know, first of all, for someone
#
who is so much into empirics and data, it is obviously the Senian influence that you
#
mentioned the word moral triumph and morally wrong and you know, you're using these phrases
#
while sort of chatting about all the things you've shattered about.
#
So I'll go to a couple of them.
#
First of all, you spoke about how democracy is a moral triumph and I agree, but except
#
that I'm not sure the democracy in India is actually a democracy in its fullest sense.
#
Like one of the things that we really did was that we took over the colonial state apparatus.
#
You know, the oppressive state remains the oppressive state.
#
You know, the tax on salt is more than what the British were charging when Gandhi went
#
on the Dundee salt march.
#
So whereas like the modern day Gandhi, and that leads me to your other use of, you know,
#
the word moral, when you said that it is morally wrong to tax the poorest, which of course
#
But of course we do tax them in terms of indirect taxes, you know, even the beggar at the traffic
#
signal will buy salt somewhere, will buy soap, will buy all of those things.
#
You look at the price of petrol today, where it has gone, you know, completely sort of
#
It's equal to theft and
#
So by the way, sorry, there I disagree, I think the petrol tax actually good, but I'll
#
come back in a moment and talk about why, like, you know, but with a few caveats and
#
But all else being equal, given that you need a certain amount of revenue, I still think
#
But I'll come back to that in a moment.
#
We'll come back to that.
#
I mean, I think because a rising price of petrol affects everything else.
#
And therefore, I think that is slightly problematic.
#
And also what we saw with, you know, the Modi government in 2014 is that global oil prices
#
crashed, but our oil prices didn't.
#
They weren't passed on to the consumer.
#
They were, you know, kept by the predatory state.
#
And apart from that, obviously, I couldn't agree with you more.
#
I've said it repeatedly so often that it's almost become a cliché of mine that we have
#
a state that, you know, doesn't do the things it should do, like a rule of law and like,
#
you know, what we'll be discussing today, but it does a lot of things that it should
#
And those are not just, you know, wastages like all the areas of rent seeking and building
#
statues and grand buildings for our rulers and all of that, but also in the many different
#
ways that it comes in the way of society solving its own problems.
#
Like, you know, one reason as I think Ajay and I discussed or Gautam and I discussed,
#
I forget who I discussed it with, but I think everyone would agree that one reason this
#
second wave came so hard is that we didn't have good enough data about how bad the first
#
And one of the reasons we didn't have good enough data is a government monopolized testing
#
in the early part of when COVID-19 struck and said only we will test and it got in the
#
way of private sector solving its problems.
#
And of course, as Ajay and I discussed in our, you know, vaccines episode, you know,
#
the government has to do what it has to do, go on a war footing and try to vaccinate people,
#
but also allow private people to get Pfizer, Moderna, J&J, whatever, from outside.
#
But not only does it not do the few things it should do properly, it gets in the way
#
of us solving our own problems.
#
But having done with my aside, I will let you talk about the petrol aspect of it.
#
And then we can, then you can get to answering the question, which in case you've forgotten,
#
I'll remind you, is how should we think at a meta level of our healthcare system?
#
Like what kind of, what were the design mistakes we made?
#
And is there someone who got it right in an equivalent condition?
#
So let me save the petrol tax question when I come back and talk about pollution and why
#
I think it's a good tax.
#
But the, so, but coming back to systems of healthcare, I think the, I would say the cardinal
#
I mean, and this is not just India, the cardinal mistake is prioritizing curative over preventive
#
So if you were in a world of scarce resources, if you have to say, I need highest ROI on my
#
limited spending, where am I going to spend that money?
#
It is, it is really is in water, right?
#
So water quality, air quality are, you know, just, these are the ways through which most
#
diseases kind of pass, right?
#
So and things like surveillance, right, mean, and prevention.
#
Now there are some things we've done well, vaccination is something we've done well.
#
And I'll come back and talk about the micro incentives and why it works well, right?
#
But this problem of prioritizing curative versus preventive is not unique to India,
#
I mean, it is the nature of politics.
#
It's the nature of the fact that people, it just responds to the fact that people viscerally,
#
you know, think about health only when you don't have it, right?
#
I mean, so when you're healthy, you don't think about it.
#
You don't do your exercise.
#
In my case, I don't sleep enough, you know, so we all have our own kind of ways in which
#
we under invest in our health in a preventive way, right?
#
But the moment something goes wrong is when it becomes salient, okay?
#
So that I think is, unfortunately, it is, I would say, a government reflecting what
#
people's revealed preference is now, ideally, this is where leadership matters, right?
#
Because a good politician gives his voters what they want a good leader, like, I mean,
#
takes his or her voters to where we should go, right, I mean, in the public interest.
#
And so, and this is why public health and health policy in particular requires serious
#
scientific expertise and serious economic expertise in terms of cost effectiveness and
#
making the public case and saying this is what we need to be doing, okay?
#
So I think in terms of the structure, you can't argue with the core of the system that
#
I think the reason is weakened so much over time, frankly, again, is it goes back to these
#
resource constraints, right?
#
I mean, that you're understaffed and under equipped, but also the incentive constraints,
#
So part of the problem that I was documenting in the absence work is that you have these
#
clinics with about 40% absence rates of doctors and nurses.
#
And again, you can't even necessarily blame the doctors and I'll come back to that in
#
But I think where the system fails is the system still primarily thinks of itself as
#
a logistics operation and doesn't think about, you know, the incentives and how this plays
#
So, so when we come talk about policy, because I think my broad hope is that, you know, given
#
our three hour rough track record is to use the first hour to just talk principles, the
#
second hour to talk research in India and what we've learned in the last hour to talk
#
But, you know, when I come back to policy, I'll talk about both kinds of investments
#
we can and should make in terms of public health and also in terms of, of the curative
#
Now, but let me give you an aside in a fun joke, you know, which you really like this
#
problem about prioritizing curative as opposed to fundamental things that give you deeper
#
So you see this in the US, right?
#
Schools of medicine are much, much better funded than schools of public health.
#
So medical school professors will have like tons of endowed professorships, tons of huge
#
research grants, so much money, that is the public health guys are always kind of scrounging
#
for resources because it's kind of seen as not as urgent, right?
#
I mean, so that problem is that even in the US.
#
So one of my apocryphal, I don't know if this is a legend, but I've definitely heard this,
#
but I've forgotten who I heard it from.
#
So I can't source this.
#
But there's this wonderful apocryphal story about the director of the National Science
#
Foundation and the director of the National Institute of Health.
#
So these are the two biggest funders of research in the US and, and they are both testifying
#
in front of the Senate, okay, making the case for their budget.
#
And so the NSF guy, you know, apparently shows up with like all of these files and like tens
#
of thousands of papers, right?
#
Think here is all the amazing research that's been produced.
#
Here's the citation count, here is how much, you know, this research is changing XYZ and
#
And the NIH guy has like a very, very thin folder, he has one thin folder with just hundred
#
And this guy has about 10,000.
#
But at the end of these hearings, the NIH guy walks out with a budget that's 10 times
#
Like, you know, so with his hundred page binder.
#
So the NSF guy apparently asked him, hey boss, like, you know, how did you do this?
#
Like I, here I have all of this evidence, all of this stuff, and you're walking away
#
with a 10 times bigger budget.
#
What is in that folder of yours?
#
So he says, aha, what I have in these hundred sheets of paper are the personal medical histories
#
of every one of the hundred senators, like, you know, in this room.
#
So you know, the point is that we are just much more viscerally connected to things that
#
we have experienced and ill health is, you know, the, the need for curative stuff is
#
something we see in the moment of having fallen sick, right?
#
But the need for preventive stuff that has much higher returns is the seen and the unseen,
#
Means so everything almost like my book should have a subtitle seen and unseen for like every
#
Like, you know, and I'll give you some copyright royalties on the title, but yeah, I mean,
#
it is, it's, it underpins everything.
#
But therefore I think it's easy to kind of blame the leadership, but in a way it reflects
#
what the population wants, right?
#
Our population wants tertiary care in a way more than primary care.
#
So part of the point of a moment like this is for scientific leadership and kind of economic
#
policy leadership to explain to the political leadership and teach them how to explain to
#
I mean, that this is what we need as a country and why we need to do this, right?
#
I mean, but yeah, so that's, that's what we need to do.
#
Firstly, I mean, the seen and the unseen, you should, you know, give the credit to Frederick
#
Bastia because obviously the title is taken from his famous essay, that which is seen
#
and that which is not seen, though that appears to me to be a clumsy English translation of
#
whatever the French original title was.
#
A couple of asides and one is an anecdote and it's almost a circular anecdote because
#
you sent me a paper by Das and Hammer from which I posted an anecdote on Twitter, which
#
I'll repeat later when we get to kind of that section of this episode.
#
But in response to that, there is this excellent doctor in Mumbai called Lancelot Pinto and
#
he shared this anecdote.
#
You know, he gets asthma patients all the time who come to him when they have an asthma
#
attack and they want medicine for it and all of that.
#
And each time he tells him that, hey, I gave you an inhaler for preventive use.
#
Why don't you use that?
#
But they don't, they don't care about preventive.
#
It's almost like there's an illusion of eternal good health and immortality until, you know,
#
disease strikes you or something strikes you, you don't care.
#
And then you want it to be cured.
#
And this sort of brings one to the question of incentives again, you know, in the sense
#
that look for a politician.
#
Why should he invest in preventive care or public health?
#
You know, when the results of that are not seen, you know, and will be seen like way
#
beyond whatever the election cycle is.
#
And one answer as to how those incentives would be different is, you know, I'm just
#
thinking aloud because you were mentioning pollution earlier.
#
And if you sort of think about the history of London, London is an excellent city to
#
live in today in terms of livability.
#
But at the turn of the 19th century, it was a hell hole.
#
It was way more polluted than it is today, right at the turn of the 20th century, rather
#
way more polluted than it is today.
#
But what typically happens and has happened in all the great Western cities is that once
#
people reach a certain level of wealth and empowerment, they demand better governance
#
And then the state has no option but to give it to them.
#
But I would imagine and I'm thinking aloud that the condition which, you know, makes
#
that possible is that there is empowered local government where it can actually happen.
#
And that's not really the case in India again.
#
I mean, I know these are all old laments and we will no doubt discuss this in the episode
#
I will punt into the episode on federalism.
#
Yeah, we'll talk about it in the episode to federalism.
#
So now is the time to kind of get into the research part of the episode and into the
#
Without further ado, let's take a quick commercial break.
#
Long before I was a podcaster, I was a writer.
#
In fact, chances are that many of you first heard of me because of my blog India Uncut,
#
which was active between 2003 and 2009 and became somewhat popular at the time.
#
I love the freedom the form gave me and I feel I was shaped by it in many ways.
#
I exercise my writing muscle every day and was forced to think about many different things
#
because I wrote about many different things.
#
Well, that phase in my life ended for various reasons.
#
And now it is time to revive it.
#
Only now I'm doing it through a newsletter.
#
I have started the India Uncut newsletter at indiancut.substack.com where I will write
#
regularly about whatever catches my fancy.
#
I'll write about some of the themes I cover in this podcast and about much else.
#
So please do head on over to indiancut.substack.com and subscribe.
#
Once you sign up, each new installment that I write will land up in your email inbox.
#
You don't need to go anywhere.
#
So subscribe now for free.
#
The India Uncut newsletter at indiancut.substack.com.
#
Welcome back to The Scene In The Unseen.
#
I'm chatting with Kartik Moolidharan on the state of Indian health care.
#
Like let's understand the problem before we begin thinking about solving it because right
#
now it seems almost unsolvable, but no, but let's I mean understanding the problem is
#
good enough and at least we'll have profound confusion instead of naive confusion.
#
As Greg Mankiw once told Kartik, so Kartik, let's get down to the nitty gritties.
#
One of the sort of delightful thing and you send me a bunch of papers which are linked
#
from the show notes and one of the delightful things about them was that one, they were
#
So they confirmed a lot of my intuitions and two, a lot of the data also, you know, felt
#
counterintuitive at times and, you know, gave me a more nuanced understanding just by virtue
#
of being what they are, just the figures themselves without any interpretation necessarily.
#
And I found it fascinating and in the chapter in your book, in fact, you know, you've structured
#
it very well where you begin talking about outcomes.
#
So I'll ask you about our health care system.
#
You've spoken about different metrics like infant mortality, life expectancy, stunting
#
and malnutrition and so on by which we can get a sense of how far we have come.
#
And the point is, it's a mixed picture.
#
We've come quite far, but we should have been much further along.
#
So tell me a little bit about sort of what these measures are and which ones of them
#
were sort of revelatory to you or spoke a lot or surprised you.
#
One of the points I wanted to come back to is, remember the thing I said, which you said
#
should be on t-shirt, you can't eat GDP.
#
Now that's not fully right because you can eat GDP, right?
#
You can use the GDP to buy food, to buy like better medicines, buy a bunch of things, right?
#
So I think and we'll see here that GDP really does matter, okay?
#
So that systematically countries and states with higher incomes will have better health
#
The point is more to say that GDP is not an end in itself and you know, it's what it allows
#
So with that little caveat.
#
Quick aside there, this is a modification on the t-shirt for fat people like me.
#
On the back, the t-shirt should say you can't eat GDP, but on the front where the ponches
#
it should say you can eat GDP, sorry, continue.
#
So I think, you know, the, yeah, so like I've done in the book with the draft of the book
#
chapter I've sent you is that if you want to assess the health system overall, you want
#
to kind of pick a few key indicators that speaks to the larger status of your population
#
So life expectancy is usually a good statement of population health overall, right?
#
And infant mortality is a good measure of your health system.
#
And the reason I say that is that infant mortality is something that for the most part we know
#
what to do about, okay?
#
For the most part, like, you know, the biggest causes, whether it's diarrhea or whether it's
#
you know, diseases that you can address the vaccinations or with very targeted, you know,
#
institutional deliveries, neonatal health.
#
So we kind of know what to do.
#
And so which is why it's a good measure of your health systems, because you're not at
#
the frontier of your ignorance.
#
You are well within kind of your knowledge frontier and you just have to learn how to
#
So that's why infant mortality is like a good measure of your health systems.
#
But life expectancy is a good measure of your population health.
#
Now there is an important modification to this, which the WHO uses called the HAIL,
#
which is the health adjusted life expectancy, right?
#
So you can have a long life, but have a lower quality of life, particularly because of things
#
like pollution, where, you know, you're just having higher morbidity, even though you may
#
kind of continue to live.
#
So that's, I think, an important additional metric to look at.
#
And then I think the other really important metric of future population health is child
#
So the mortality is, can I keep the children alive?
#
But if you keep them alive, so in a way, in a perverse way, actually, reducing child mortality
#
can worsen child malnutrition.
#
And that's because you're keeping the left tail of the distribution alive, right?
#
You know, that may otherwise have passed away.
#
So, you know, so the first step is to make sure children stay alive.
#
The second part is to then make sure that by age two or by age five, which are kind
#
of these key development, particularly age two, right?
#
I mean, these first thousand days of life, as they call them, the nine months in utero
#
and the first two years of life are kind of the critical developmental period where you
#
kind of want to make sure that the children are getting these right inputs in terms of,
#
you know, cognitive developments and brain development and other overall physiological
#
So that would be another key metric to look at.
#
So I think the glass half full story, as with most things in India, is that we've actually
#
made remarkable progress on many of these indications in the last 40, 50 years, right?
#
So in particular, life expectancy has gone up and mortality has gone down.
#
Almost every metric has improved.
#
But the question is, like, you know, like they say in the old grading of students, like,
#
you know, my teacher, most of the time, like, you know, if you get a hundred grade for anything
#
else, it'll be can do better, right?
#
Like, you know, I mean, so, and so, you know, I think the overall grade for India is can
#
But before we get sanguine, right, we can also get much worse.
#
OK, so we can also do much worse.
#
And so and sometimes I think there's a little bit of a naive push to criticize everything
#
we've done as a country.
#
And that's also incorrect, right?
#
You know, we have done a bunch of things that are correct and that we are laboring under
#
monumental other institutional weaknesses, which we can talk about.
#
So overall, my net net assessment of India over 75 years is that of most metrics, I think
#
we have a lot to be proud of.
#
At the time of independence, nobody predicted like, you know, we'd stay united or kind of
#
stay mostly democratic.
#
So there are basic sign of successes that we've had as a state.
#
And even on development outcomes, I think we've actually not done badly.
#
OK, the question is, what should our benchmarks be?
#
OK, and now, unlike, say, people in education who go and study education systems in Finland
#
or Singapore, and that kind of really irritates the hell out of me because I'm like, you know,
#
you're going and studying places with 20 times the per capita income.
#
But I think if you think about what would be three relevant comparisons for India, OK,
#
three countries I think we need to benchmark ourselves against would be China, Vietnam
#
And China, because it's the only other large country of similar population in terms of
#
just the sheer size of what they're trying to manage, Vietnam, because it has similar
#
per capita income as India, but does much, much better on health, education and other
#
And Bangladesh, because they had the same initial conditions as us mean and in many
#
ways were actually worse than India in 71 because, you know, the devastation of the
#
war kind of kept them at a lower place.
#
But Bangladesh has now steadily improved and is now like, you know, far ahead of Bengal
#
and even ahead of, like, all in the averages, like, you know, I mean, on a bunch of human
#
So again, so these are the three benchmarks I want us to look at when we are trying to
#
make a sense of can do better.
#
So it is not like, you know, I'm taking a student and saying a hundred percent, I mean,
#
I'm saying like, you know, look at these three other comparable students who are comparable
#
in their own way and saying, how can you get slightly better?
#
Then the other big part about health in India that's worth reflecting on is just how much
#
variation there is within the country.
#
I mean, so, you know, we have this paper, which I'll talk about, but the title of that paper
#
It's to India's in the sense that the quality of health care that you have in the southern
#
states, particularly Tamil Nadu, Kerala, right.
#
And I mean, there are some other states too, is essentially like a couple of standard deviations
#
about like, you know, what you would get in the north.
#
So I'll talk a little bit more about this measurement paper later.
#
But for example, we find that, you know, even like unqualified providers and nurses in Tamil
#
Nadu and Kerala do better than like, you know, fully trained doctors in UP and Bihar.
#
So that's a separate, and we'll come back to that in a moment.
#
But you know, there's obviously a lot to learn from the within country variation, right.
#
I mean, and what did some of these states do, right?
#
The way I like to think about health systems is what I've done in the book is to just have
#
a set of graphs where I put outcomes graphed against GDP per capita.
#
So, because in general, more income allows you to buy more health.
#
It means your parents are richer, the system is richer, everything is richer.
#
So I expect income outcomes to improve with income.
#
But the real lesson comes from looking at which are the states that are positive outliers
#
conditional on income, right?
#
So holding your income constant, which are the states that are doing much better.
#
So to give you a simple example of how that flips results around in absolute levels, outcomes
#
in Bihar look worse than Haryana.
#
But Haryana is actually much richer than Bihar.
#
And so if you kind of control for GDP and look at health outcomes, actually Bihar ends
#
up doing slightly better than expected for its very, very low GDP.
#
And Haryana ends up doing much worse.
#
Now, but then that starts getting not just into policy issues, but also into social and
#
cultural issues, into empowerment of women in particular.
#
So in fact, ironically, it turns out that women in Bihar end up being much more empowered
#
partly because there are so many more migrant workers who leave the state, right?
#
And so, which means that the women are running the show, like, you know, much more than they
#
are in some of the other states, because essentially the men are migrating and gone out.
#
So there are some indirect positive gender spillovers like, you know, that you see in
#
Bihar as a result of the structure of the economy.
#
And I think the larger point when we think about how do we get better as a country is
#
to say that we've done well, we have a lot to be proud of.
#
And I'll talk about the specific pieces of the system that have worked well, but we can
#
And then, you know, that's how I frame that whole discussion.
#
So that's kind of overall health, right?
#
But we can then start looking at the healthcare system itself in terms of the structure of
#
What does it look like?
#
Is there stuff you, I'll pause for a second before I start talking about that.
#
What I'll do is everything that you said, I'll fill in some figures which I've got from
#
papers partly written by you and from the chapter in your book itself, because I found
#
them very illustrative.
#
Like one, what is the glass half full?
#
The glass half full is that infant mortality in India has declined from 142.6 per thousand
#
live births in 1970 to 28.3 in 2019.
#
And you'd think 28.3 is good, but it's still four times more kids dying here than in China,
#
which is 6.3, Vietnam is 15.9.
#
Even Bangladesh is, you know, less kids die, 25.6.
#
You know, again, you talk about how life expectancy at birth was, you know, 41.4 in 1960, which
#
If we were like the median person, we'd be dead if we were in 1960.
#
And it's gone to 69.4 today.
#
But the point that you made about health-adjusted life expectancy was that India's health-adjusted
#
life expectancy, what are the healthy years of your life that you can expect, that's 59.3.
#
So there are 10 years of the 69.4, which you're losing to ill health and you're nowhere near
#
Again, this number is, you know, worse than China, Vietnam or Bangladesh.
#
And you give various reasons for this.
#
Similarly, you talk about sort of stunting, where I'll quote from your book, quote, India
#
has reduced a fraction of stunted children under five from a shockingly high 62.7 percent
#
in 1989 to 34.7 percent in 2017, stop quote, which is also shockingly high.
#
And as you point out, quote, India's stunting rate is 4.3 times greater than that in China,
#
you know, 34.7 versus 8.1, 46 percent greater than Vietnam, even 13 percent greater than
#
Bangladesh, which are, you know, the glass half empty, therefore does seem to be quite
#
Then you pointed to, you know, figures that reflect the variation that you gave within
#
You point out how infant mortality rates in states like MP, MP is 47, Odisha is 44, Assam
#
And you have Goa, which is 8, Kerala, which is 10, Puducherry, which is 10.
#
You know, Tamil Nadu is 17, Maharashtra is 19.
#
You know, the figures kind of speak for themselves.
#
And about those outliers also, I found that fascinating how, you know, in absolute terms,
#
Bihar seems to be worse than Haryana.
#
But when you adjust for per capita income, it's actually much better.
#
And one, you gave the Bihar side of the reason that so many migrants are there, so women
#
You know, I'll again give a brief aside.
#
I once wrote a feature story on Sakshi Malik after she won the bronze medal at the Olympics.
#
I think she won bronze, right?
#
One of the things I was trying to investigate during that is how the hell does a state which
#
is known for its misogyny, in other words, Haryana and the way it treats its women, how
#
the hell are they getting so many female wrestlers and all of that?
#
The answer actually does lie in patriarchy, where in the 1970s, there was this dude who
#
was a wrestling coach called Chandgiram.
#
And Chandgiram didn't have sons.
#
And in 77 or 78, the Olympic Committee announced that, you know, that they'd have a women's
#
So he made his daughter, I think, Sonika Kalikaraman.
#
This is the Dungal story.
#
The Dungal coach, Mahavir Singh Fogart was Chandgiram's assistant.
#
Yeah, so Chandgiram got his daughter into it.
#
I think her name was Sonika Kalikaraman or whatever.
#
And then Mahavir Fogart, who was his assistant, got his daughters into it.
#
So this whole thing was these men who don't have sons, who want to vicariously achieve
#
something they couldn't see the Olympics as this thing.
#
And therefore, they push their daughters into it.
#
But that doesn't mean that women are empowered in the state overall.
#
And of course, women's empowerment is actually, as you have pointed out, you know, one of
#
the key indicators of hell, the kind of food that they're getting and even the spillover
#
effects, because a mother who's not getting enough nutrition will, you know, affect her,
#
the childbirth, it will affect what happens to the kid after that, all these massive spillover
#
So I've, you know, everything that you said have kind of taken data from your chapter
#
and sort of put it there.
#
And I think it was important to give this data because it just speaks for itself.
#
It is so powerful that yes, glass half full, yeah, we're not dying at 41, we're dying at
#
But look at all these other countries.
#
What have they achieved?
#
One of the things in the last episode that we did on education, you told me what really
#
struck you was to look at, say, was to compare India and Indonesia and say, okay, we were
#
here at, you know, we were at a comparable level once upon a time, and now look at them
#
and now look at us and where did we go wrong and so on.
#
Having added to this dismal story, I'll kind of let you let you continue.
#
And thanks, because I think, yeah, I didn't give the numbers partly because I felt, you
#
know, that's a part of the chapter that's a little dry, but you're right, the numbers
#
I'm glad you filled that in, like, you know, so then I can talk about, you know, the concept.
#
So coming back to the research that I want to talk about, right, I think what most of
#
your listeners probably have very little idea of is just the landscape of health care provision,
#
particularly in rural India, okay.
#
So and, and this is something that not just your listeners, but the government frankly
#
And that's because, you know, the majority of health is provided by completely unqualified
#
Now, we know that they exist because in household survey data, we know people are going to the
#
private guys, but we have no data because they're technically illegal.
#
And so that means they don't show up in any official database about, you know, who these
#
Let me give you a little bit of my journey into kind of just the three major health studies
#
I have done, like, you know, so I can speak in a little bit more detail about my own work,
#
but also draw in on some of the work by Jishnu and Jeff in particular, to give your listeners
#
a little bit of a capsule of what we've learned about the functioning of the healthcare system,
#
So like I said, you know, I started this journey with kind of this project and looking at quality
#
of public service delivery, right?
#
And this was part of a project that was done for the World Bank's World Development Report
#
in 2004 on making services work for the poor or 2003.
#
And so it had kind of a cross country component.
#
And part of the challenge was coming up with metrics of quality that were measurable in
#
identical ways across contexts.
#
So, and the reason we picked up provider absence is it's a very blunt measure.
#
Obviously, the doctor being there is necessary, but not sufficient, right?
#
I mean, he could be there and still not do anything.
#
But it's a very, very basic blunt measure.
#
We said, okay, you know, like Woody Allen, I think, famously said, right, that 80% of
#
success in life is just showing up, right?
#
It means so, so, you know, we started that very, very basic measure.
#
And just like we did for education, we also did for health clinics, where essentially
#
trained enumerators made surprise visits to over close to 1500 clinics across the country,
#
about 80 in every state and just measure it.
#
And so what we got out of that was this estimate in 2003 that about 40% of doctors and roughly
#
the same fraction of health personnel in general were absent on any given day.
#
And that's a shocking number, because once you take away the vacancies, right, I mean,
#
the fact that these were conditional in the positions being filled, right, you've also
#
got vacancies that you can see how abysmal that is in terms of the amount of services
#
Now, one of the limitations of our all India study is we could not check because the doctors
#
could potentially be working, but just be on site, okay, and field visits.
#
They could be in villages where they're supposed to be.
#
They're supposed to be doing rounds.
#
So there was a complementary study, in fact, done.
#
Then nobody had them, but three Nobel laureates, right?
#
So Esther Duflo, Abhijit Banerjee and Angus Deaton, and though they fight on our cities,
#
they actually work together on that project, right?
#
I mean, in Rajasthan, in Udaipur, but there, because it was one district, they were able
#
to do much more detailed work of actually following to the village where the doctor
#
or nurse could have plausibly been, okay.
#
And they find that in only 10% of the cases, could they find them in the field?
#
And that the story was worse than just absence, right?
#
The story was also one of unpredictable absence.
#
So if I know that the doctors is predictably absent two days a week, okay, then that could
#
But I know he's there on Mondays, Thursdays and Fridays.
#
And so as a patient, I optimize around that.
#
And that's completely fine.
#
The problem is that they found that you could run a regression of all the observable data
#
and you would have zero predictive power of when this guy is going to be absent, okay?
#
So, and that makes it much, much more painful for the poor beneficiary who kind of has to
#
both forego a day of earnings and pay the transport to go to one of these clinics, right?
#
And then if you go there and find that the doctor may not even be there, you know, that
#
becomes much, much more problematic in terms of access to public health care, okay?
#
So those two studies, I think, basically focus the challenge on governance, right?
#
I mean, that the basic governance challenges we have in the public system.
#
Now, after that, I started working much more in education, right?
#
And it's Jishnu and Jeff, Jishnu was then the World Bank, both of the World Bank, then
#
Jeff went to Princeton, Jishnu is now at Georgetown.
#
You know, so they led a lot of the work on improving measurement of how do you just measure
#
And this turns out to be a really hard problem because measuring requires, there is a possibility
#
of undertreatment, there's a possibility of overtreatment, there's, you know, what the
#
quality of care is going to vary by protocol.
#
So long story short, the method that is most commonly used in the literature to assess
#
kind of doctor competence is what is called the method of medical vignettes, okay?
#
So, and the vignettes are the form that doctor qualifying exams will look like, where you
#
say, you know, you give the doctor a hypothetical case and you say a patient is walking into
#
your clinic and report symptom X, Y, Z, okay, like, you know, so what do you do next?
#
So the whole point of kind of good medical education is that you have a treatment protocol
#
or a diagnosis protocol, right?
#
So it's what is called a checklist of essential items, saying somebody walks in with the following
#
symptoms, then, you know, here are the first four things you do, right?
#
You take the basic vital statistics, you measure temperature, you take a little bit of history
#
and you know, the patterns of the fever, you know.
#
So you try to ask questions.
#
So the whole point of being an effective medical provider is the patient walks in with some
#
random symptom and mapping that symptom into an underlying diagnosis and then treating
#
only that and not over treating, right?
#
Is the core of what you want a good doctor to do, okay?
#
So the first step of assessing is this provider able to do that is to then assess provider
#
So that's what they do with these vignettes and they measure doctor knowledge in a bunch
#
This has kind of been used in multiple settings.
#
Now the problem with the vignettes is that the vignette only gives you a measure of doctor
#
It doesn't tell you anything about doctor practice, right?
#
I mean, what does the guy actually do like I mean, when a patient walks in, right?
#
So you know, you might know a lot, but choose to not exert effort.
#
And this goes back to again, something Ajay said, which is healthcare is not a public
#
Healthcare is an inherently private good, right?
#
Because I give it to you and not to others.
#
It is rival because if I spend time on you, I have less time for others.
#
So I can both exclude and it is right.
#
So it is not a public good at all, right?
#
It is a classic private good.
#
And the amount of effort you put in is costly, right?
#
And coming back, the historical anecdote, which I wanted to talk about, which people
#
may not realize is that the British colonial healthcare system, the medical system that
#
was set up, it was not set up to take care of the population.
#
It was set up to take care of the collector and the superintendent of police, right?
#
So it and it was a completely excludable good, right?
#
It was the doctor, the medical officer in the district's job was to make sure that
#
the ruling employees were healthy, right?
#
Like I mean, population to do like, you know, that would come maybe in times of some major
#
pandemics or something.
#
But for the most part, it was a private good that was allocated to the elite.
#
So what they then like, listen, knowledge is not enough because when people walk into
#
a clinic and you know, you've seen those papers.
#
So what they started doing was then measuring what they call participant observations, okay,
#
where you have people sitting in the clinic, almost like, you know, anthropology type,
#
you're sitting as, and you're just coding every interaction between the doctor and the
#
So patient walks in and you just observe.
#
Now for the first half day, one day you might get some, what are called Hawthorne effects
#
that they work a little harder because somebody's watching them.
#
But after some time, we now have evidence to say that it just goes back to normal practice.
#
So that gives you a measure of effort, right?
#
So what they were finding and the main kind of insight from a bunch of these papers.
#
Now of course, this level of measurement could only be done in smaller samples, right?
#
So they had samples in Delhi, they had, you know, other samples, but the punchline that
#
was coming out of all of this was that the doctors, A, the knowledge was low, but even
#
conditional and knowledge, the effort in the public system was incredibly weak.
#
So they kind of came up with this terminology, what they call the no-do gap, right?
#
The gap between kind of knowledge and practice was so large that it kind of highlights again,
#
the importance of incentives because you can hire people based on competence and knowledge,
#
but that doesn't kind of ensure that they will exert effort when they actually see a
#
Now, and this is a very difficult problem, right?
#
It's a difficult problem.
#
And I also sympathize with the doctors in some way, because in a funny way, you almost
#
have negative incentives for effort, right?
#
Because if you put in a lot of effort, you get a reputation as a good doctor and you
#
And you get more patients, but your compensation is not going up.
#
Your resources are not going up.
#
There is nothing in the system that is actually variably increasing with the amount of patients
#
So they went back to the separating equilibrium of how these incentives work out.
#
And then they also looked at the private doctors and see that the private doctors in general
#
But they actually work harder because they function closer to the frontier of their lower
#
So if quality is knowledge times effort, the public guys have more knowledge because you're
#
recruiting based on more formal credentials.
#
The private guys are much less qualified on average, but they work harder and they function
#
closer to the frontier of knowledge.
#
So this is the broad message that was coming out of that work.
#
Now the limitation with that is in the participant observation, when I'm sitting in the clinic
#
and coding the interaction, I don't know what is the underlying ailment the patient
#
So I can observe your bedside manner.
#
I can observe how many questions you ask.
#
I can observe how many medicines you give, but I have no idea if what you've done is
#
Because I don't know what the underlying disease is that this random patient is walking in
#
Now in the vignette, I know that in the vignette, I know what the underlying cases and I'm structuring
#
the whole case so that a good doctor will find his or her way to the correct case and
#
the correct diagnosis, right?
#
So the problem in the PO, PO measures effort, vignette measures knowledge, but I want to
#
measure both in the same case to then really get a sense of is the doctor getting the right
#
And so then the big innovation and this is, you know, it's been used in medical studies
#
before, but just for your listeners is a detour into the wonderful world of research, right?
#
Like, you know, is the, so the, the closest to a Bollywood operation, like, you know,
#
we've had in any of our recent studies is if you truly want to understand both doctor
#
knowledge and doctor practice, the only way to do this is with what are called standardized
#
So these are essentially actors like a mean who are trained to be fake patients, right?
#
Mean who go in and present a set of symptoms.
#
Now because you know the underlying case, okay, so now you can actually assess whether
#
the quality and the protocols the doctor is doing are correct as per the medical checklist
#
So you both see knowledge and you effectively see the real practice of what this person
#
And then at the end of the transaction, you kind of observe the medication.
#
So you get to observe both over medication and under medication, and you observe the
#
So then you can kind of analyze what are market prices, what are these prices reflecting?
#
And again, this is not something we came up with, right?
#
So standardized patients are a well established tool for medical education in the US and Canada.
#
So if you're a medical student at some point in your life, you will see a standardized
#
So it is part of your training and part of your testing, okay, that a random person will
#
walk in and like, you know, you're going to be assessed on how you do.
#
So the only thing is we adapted that same method for assessing the quality of actual
#
So this is a little bit of the research journey in terms of just the methods and the measurement.
#
So I think Ajay last time was talking about, you know, how little we know, you know, and
#
one of the reasons I thought is a good time to do the episode is it builds up on that,
#
you know, that we actually do know a lot, but not much, not as systematically as we
#
would like to, because these were these one-off studies and we need to do a lot more of them.
#
So, you know, and then there was this last kind of study.
#
So the two studies, which I did then after, so Jishu and Jeff did a bunch of these things
#
and then we got together in about 2010 and worked together on two major studies.
#
So one was an All India study looking at the availability and landscape of providers in
#
So it was a representative sample of villages and all we did was we just went and did a
#
Like, you know, so who is here?
#
Like, you know, and where are people going to?
#
And then on an All India basis, there was no way you could do standardized patients.
#
And so there we measured quality just using the vignettes.
#
So that was a measure of doctor knowledge.
#
But then in one study in Madhya Pradesh, in a second study, in MP, we went, did a much,
#
I mean, where we had these standardized patients then go and present to both public and private
#
clinics to go present to the same doctor in his public and private practice.
#
And we'll come back and tell you a bunch of stories about that.
#
But you know, from a research perspective, those were some of the studies that were done.
#
And yeah, I'll take a little pause and come and give you the insights.
#
So maybe I'll give you the insights.
#
You know, I'll, I'll start with the All India one, maybe.
#
And then take a pause and come in with the MP one after that, unless you want to come
#
No, I think I'll come in right now because there is so much that you mentioned to unpack.
#
I think let's kind of unpack them one by one and come to this and I'll get to the rural
#
part of it and absenteeism and all of that later.
#
But very quickly to just sort of elaborate on what you said about a vignette for the
#
sake of my listeners, what that in real terms would be that say, I go to a, I go to the
#
doctor and I'm the sort of actor as it were acting as if I have something and I will say
#
And then what we are measuring is what are the questions that person asked me?
#
How relevant are they to my problem?
#
So you know, these are the five questions that a good doctor would ask.
#
These are the six things that a good doctor would check the last of them, apparently being
#
some indentation in the temple or the head or something, which has something to do with
#
And, and you grade them based on how many of these they actually ask, which I kind of
#
There's this extremely interesting conversation, you know, in the 2007 study, Jishnu Das and
#
Jeffrey Hammer, where, you know, Das had a conversation with this doctor in a slum and
#
I love this conversation.
#
So I'm just going to read it out because it is, I think, so relevant to present times
#
also where this doctor says, yes, there is a lot of diarrhea and dysentery in this locality.
#
What can they do as well?
#
People don't know how to boil it.
#
That's why their children are always falling sick.
#
So Jishnu asks, so what do you do for children with diarrhea?
#
So the doctor says, what can we do?
#
We tell the mother to give water with salt and sugar to the baby and then also give some
#
So Jishnu asks, such as?
#
So the doctor says the usual, metrogel, loperamide, furoxone, forgive me for the pronunciations
#
I forgive you for your handwriting.
#
And then Jishnu says, but isn't ORS enough, ORS being oral rehydration, basically salt
#
And the doctor says, of course, the who and the others keep saying we should only give
#
If I tell the mother she should go home and only give the child water with salt and sugar,
#
she will never come back to me.
#
She will only go to the next doctor who will give her all the medicines and then she will
#
think that he is better than me.
#
And this, of course, speaks to a classic problem that you point out about how private doctors
#
often tend to over prescribe for different kinds of incentives.
#
And we'll discuss that in detail.
#
And I, in fact, reproduced, I took a screen grab of this and put it on Twitter today because
#
it spoke to me because yesterday there was an interview with Dr. Lancelot Pinto and another
#
And I'll link it from the show notes where they were talking about how one of the things
#
that is making COVID worse among the many things making COVID worse, one of the things
#
that is making it worse is that people are just over medicating where Dr. Pinto's point
#
is that even we know all the things that don't work like plasma and ivermectin and hydroxychloroquine
#
and even remdesivir is only in particular clinical situations is recommended.
#
But one of the things that does work is dexamethasone, which is a corticosteroid, but even that should
#
only be prescribed at a particular stage of the illness and what doctors are doing it
#
is that when young people are reporting fever, they are giving steroids from then only.
#
And what the steroid will do is it will reliably take the fever down, but it will also suppress
#
the immune system reaction and then you will be hit harder with COVID.
#
And so people are getting over that fever and then just falling apart.
#
And it's because of this over medication that is happening.
#
So what is required at the start is that that initial fever of four or five days, you just
#
have to live through it.
#
That's a body's immune system.
#
Unless your oxygen level levels go down, you don't really need to do anything.
#
You know, that's the natural response of the body.
#
If you over medicate at that point, you only make it worse.
#
And one of the things that is happening is patients are asking for all of these because
#
they will read that array my uncle got ill and he was given these five medicines and
#
remdesivir and plasma and all of that.
#
And doctors under that pressure, just because the workload is so insane, anyway, they'll
#
say fine, you know, you can also have that.
#
And then these people will put out calls on Twitter, plasma remdesivir and all of that.
#
So I thought that in these current times, that's sort of an important aside that one
#
I think the technical term for it is polypharmacy and that this seems to be a tendency that
#
They feel a doctor is competent if that kind of happens.
#
Yeah, no, I think, you know, you're just speaking to what makes health care so so vexing, right?
#
Because there is kind of both, you know, what you're describing, which is patient induced
#
demand, like, you know, for things that are suboptimal and doctor induced demand, you
#
know, in other cases, right mean where you're kind of, you know, all else being equal, you
#
recommend procedures where you make, you know, where the revenue is, you know, is more lucrative.
#
And so that is a very, very difficult problem.
#
And this is why if you go back to look at the core of doctor training, right?
#
Why is why do they spend so much time on medical ethics?
#
Why do they spend so much time on the Hippocratic oath?
#
Why do they spend so much time on do no harm?
#
It's precisely because the founders of the modern medical profession recognize just how
#
And that the only way to kind of even attempt this problem of kind of how much asymmetric
#
information and power there is, is to actually build a more noble doctor, right?
#
And so, you know, if I can inculcate in you a certain intrinsic identity that says deviating
#
from my intrinsic identity is costly, right?
#
That's the one hope we have in the long term of actually building a better medical system.
#
And this has to do both with doctors in terms of what they do, and also doctors ability
#
Inpatient demand and saying that, you know, so, and this is why customer satisfaction
#
that we normally use, okay, is not even always a good measure here.
#
I think Jishnu gave me this wonderful quote once I forget, I think from Kenneth Pendergast
#
as a paper, it says, you know, doctor satisfaction from drug addicted client who's assessing
#
the doctor based on is he getting his next dose, like, you know, I mean, is not a good
#
measure of doctor quality, right?
#
So in that sense, the measure of doctor quality is, like I said, to diagnose precisely and
#
give you exactly what is needed and nothing more.
#
And I remember an old conversation with Siddharth Mukherjee, he may not even remember, I think
#
I met him once in Boston.
#
He's at least five or six years older than me, but, and I remember very distinctly he
#
told me that he said a good doctor recognizes that every medication is a poison, right?
#
I mean, because it is being given to kill something, okay, that can be in the body.
#
So you kind of want to absolutely minimize what you do to just go after what the ailment
#
is and nothing more, right?
#
Yeah, but the flip side of that is now if you take that medical ideal into the practical
#
situation of kind of the poor in India is that you cannot blame them, like, I mean,
#
for wanting more medicines, because essentially if you are poor and you've fallen sick, you
#
are on the verge of starvation because you don't have income, right?
#
So the single most important kind of thing that is driving the demand for health care
#
for the rural poor is how do I get back to work?
#
It's not like, I mean, they're looking for all these medicines to say, I want the steroids
#
No, it is essentially every day I don't work is a day closer to destitution.
#
And so there's no unemployment insurance, there's no safety net.
#
And so which means that coming back to basic economics of kind of the time discounting,
#
I mean, the discounting calculation you're doing is very simple.
#
It says, yes, if I take over medication, there might be some adverse side effect in the future,
#
but I need to survive to get to that future, right?
#
I mean, so which means that the discount rate is very, very high.
#
And so what may seem medically optimal, right?
#
I mean, is often not what is optimal for the patient.
#
And one of the things we document in our NP study is that there is a ton of over prescription,
#
but the over prescription is not only driven by the incentives of the doctor to make more
#
money when I give you more medicines, right?
#
The over prescription is driven by the need for the patient to want to get symptomatic
#
relief so that he or she is able to go back to work.
#
And so then in a world where the doctor actually doesn't know what's going on with you, right?
#
Like I mean, he gives you a cocktail hoping that something will have an effect, right?
#
Like I mean, and that you'll feel better.
#
So in that sense, the over medication is not, it's kind of an ethically much more murky
#
Like, you know, because he's not just doing it to make money.
#
He's doing it because he wants to help you.
#
And what you want more than anything else is to get better enough to go back to work,
#
And that's what makes the ethics of this so kind of complicated and fraught, right?
#
So yeah, so it's like I said, there's only more and more profound confusion, right?
#
I mean, that if you ask me today, what should the policy response to this be?
#
I frankly don't have a good answer, right?
#
I mean, it is, it would be more along the lines and I'll come back and talk more about
#
training and education and how you can kind of see in a way what you want to do is improve
#
the doctor's ability to diagnose and credibly communicate that to the patient, right?
#
So that the doctor can act as this trusted filter of both what you need and what you
#
don't need in a world where like, you know, there isn't this immediate competitive pressure
#
or the sense of saying, okay, if you don't give the medicine, the doctor is not good,
#
Like I mean, so yeah, if you don't give the injection, the doctor is not good, right?
#
Like, you know, so there are these tropes that people have and these, you know, these
#
are very, very, very complicated issues to solve.
#
It was really fascinating.
#
I mean, in fact, this is a classic example where real medicine might have a placebo effect
#
because a patient, you know, feels that if they are given an actual, you know, medicine
#
instead of just being told that go home and, you know, put some salt in your water and
#
drink that and like, come on, I mean, if I'm paying good money to see a doctor, how can
#
you tell me to drink salt water?
#
The other sort of observation that you made before I come to my question and I found this
#
like so fascinating and so telling is this rule of three.
#
This is from the Das and Hammer study where Das and Hammer write that quote the average
#
interaction with a doctor in our sample is governed by the rule of three.
#
And then they write the patient arrives three days after becoming sick.
#
The doctor spends three point eight minutes and ask three point two questions after which
#
the patient leaves with two point six different medicines, zero point five of which are antibiotics.
#
And such a fantastic illustration.
#
Now question number one, what does this tell us?
#
This rule of three and question number two in a larger sense.
#
And this goes across multiple studies which you'd shared with me.
#
What do we know about the difference between private and public doctors, both in terms
#
of their competence, in terms of their effort?
#
Is there a sort of a distinction depending on, you know, the locality they are in, if
#
they're in a rich locality or a poor locality?
#
Is there a distinction between some, you know, a doctor in a hospital and a doctor with his
#
And is there a distinction with a doctor who is both by day, he's, you know, a doctor in
#
a public clinic, but also has his own private practice outside?
#
Now all of these are nuances you have explored.
#
Yeah, so I think the let me just start with some basic facts, right?
#
I think the that's the easiest place to start, right?
#
So what did we do in this kind of all India study that just was published in the social
#
science and medicine last year?
#
And this is the paper that's called The Two Indias, right?
#
So what we did was had a representative sample of villages where we just went and first counted,
#
I mean, who are all the providers and then did a random sample where we went and administered
#
So I think the goal is just to characterize the landscape, right?
#
Who are these providers?
#
Ho kya raha hai, right?
#
And so the main punchline that comes out of this is that over 70% of the overall provision
#
of health care in rural India is coming from the private sector.
#
And the majority of these private providers have no formal qualifications whatsoever.
#
So that's what you would call quacks.
#
But we need to kind of put a lot of nuance into this term quack, okay, which I'll come
#
About who are these quote unquote doctors.
#
So that's fact number one.
#
About the structure of the health care system is that 70% is private.
#
Most of them are quacks.
#
And this is not even including people with legitimate qualifications who are practicing
#
outside their area of training.
#
So you have the traditional Indian medicine, people who have Ayurveda and Unani and homeopathy
#
kind of degrees, but are actually practicing allopathy, right?
#
And that's again reflects the fact that, you know, these alternative systems are meant
#
to be more about wellness rather than cure.
#
But the patient is not interested in wellness.
#
And so they end up kind of dispensing the same allopathic medicine that gives you the
#
So that is fact number one.
#
The share is major, mainly private and more than 50% are unqualified.
#
The other important fact is that this is about 50% in literally every state in India.
#
So except Kerala, which is about 45, 47%, even then, so the private share is very, very
#
And this is a little subtle.
#
It's not the fact that, but Tamil Nadu and Kerala that have better public systems have
#
They also have more private, but their private is much better.
#
So, and this is a very deep and subtle point I want to make, which is coming back to your
#
story about when the elites have exited from the private sector, why do we care about improving
#
And the subtle point is that the reason I care about state capacity and improving public
#
systems is that not only does it directly benefit those who use the public system, it
#
also indirectly benefits everybody who uses the private system because the private is
#
forced to compete with the improving public.
#
So when the public option gets better, you're also automatically forcing out the left tail
#
of the distribution of the private providers and forcing the private distribution up.
#
So there's that overall spillover in the economy that is not being accounted for in terms of
#
the value of investing in strong public systems.
#
So, so the second key fact is that as your systems get better, it is not that the private
#
goes away, but the private gets better.
#
So the, so the public is pushing up the base and it's the, that is in fact a tide that
#
So that's the second key fact.
#
The third is just how much variation that is across the country, right?
#
And so if you look at these vignettes and look at the quality, what you'll see is that
#
nurses and sometimes even the unqualified providers in Tamil Nadu and Kerala do better
#
on these vignettes than even people notionally with MBBS degrees in UP and Bihar.
#
Like, you know, now, and again, this is not to malign UP and Bihar, but I suspect it speaks
#
to the quality of medical education or the fact that you can get through, there are probably
#
more kind of fake colleges.
#
I don't know enough about this, right?
#
I mean, but you know, it certainly highlights, I would say the value of say some of these
#
national systems like NEET and NEXT.
#
Like, you know, that says let's have a standardized entrance test, a standardized exit test.
#
So that regardless of what your kind of training program is that you have to show on a certain
#
common standard, right?
#
Mean that you've achieved.
#
So the other key fact, which I want people to understand, and this is more from my fieldwork
#
rather than systematic data collection is a little bit more about who these quacks are.
#
Like, you know, so how have they become so prevalent and why are they all over the place?
#
So two things, one that the state is often not there.
#
And second, even if the state is there, the state is not accountable, right?
#
So one of the striking findings we have in our MP study is that the share of the private
#
guys is above 60 or 70 percent, even when you have a public MBBS doctor posted in the
#
So it's not just a question of have you posted a public doctor because the public doctor
#
So coming back to who these private quacks are, see, most of them are have done what
#
you would call an informal apprenticeship, okay, that they have shadowed a regular doctor,
#
like, you know, and kind of been in this guy's clinic for a year or two, observed a bunch
#
And then they say, okay, I'm ready to go off and kind of do my own thing.
#
Now, as with any skilling program or apprenticeship program in India, there is so much variation
#
in the quality of who you shadowed that that automatically then means that there are some
#
of these guys who are not bad at all because they've actually informally imbibed some pretty
#
And then there are others who are abysmal.
#
So that's one part about who these people are.
#
Now, but the second thing that's important to understand about why they are not these
#
evil kind of, you know, there's a sense that, oh, they're these evil profiteering quacks.
#
But that I think is far from the case because they are often highly embedded and very, very
#
respected in their communities because they are the only people providing care and they
#
have to live in that community.
#
So, in fact, the best incentive for them not to do something wrong is if something bad
#
happens in their watch, they're going to get beaten up.
#
Like I mean, and that's often a better insight.
#
So Jeshu has sent me stories about like, you know, people getting beaten up, like, you
#
know, if the community feels that the doctor has done something bad.
#
So in that sense that they are not formally qualified, but they are embedded in their
#
They have kind of some experience, some practice.
#
And then the other source by which they learn things is the informal education.
#
And again, this is not in the data set, but it comes from my field conversations with
#
these guys is they actually get a fair bit of continuing medical education from pharmacy
#
Because the pharmacy guys, the government might not know who these guys are, but every
#
pharmacy rep knows them because that's where the medicines are being pulled through.
#
So the pharmacy reps will have in their beat the data of all of these informal providers
#
and they will make sure they've gone and visited them.
#
And so that becomes like a source of now.
#
Of course, it's petrifying to think about then how much influence a pharmacy rep has
#
in terms of like, you know, over medication, because these guys then have incentives to
#
So, but it's just giving you some sense of the lay of the land, right?
#
I mean about who these guys are and that they, that they account for the lion's share of
#
healthcare in the country.
#
Now they're coming back to the research about the quality, right?
#
So it is one thing to say that they exist and this is what they look like on paper.
#
But what can I say about that actual quality?
#
And that is what the MP study does, right, because what the MP study does is it takes
#
measurement very, very seriously.
#
And this is a paper in the American economic review.
#
And what we do is we have these standardized patients and we draw a random sample of villages
#
and we make a census of all of the providers.
#
And then we randomly draw samples of public and private clinics.
#
And then we send these SPs over a period of time to go present their cases, right?
#
So then we can observe exactly the quality of care that's happening in the public and
#
the private, the qualified, the unqualified, and then just understand that broad gamut.
#
So again, this is one state, it's, you know, about 160 villages actually, but it's better
#
data than we have anywhere else.
#
So it's the best we have right now, but the key kind of big picture messages are the private
#
guys know less, okay, because they're less qualified, but they systematically work harder.
#
And now by work harder, it means they're spending more time, they're asking more questions,
#
they're getting more of the items on the checklist.
#
And so when you put that picture together, the overall quality of care that you get from
#
these quacks is not different from what you get from the public health clinic.
#
And now this is because when you go into the PHC, most of the time the doctor is not seeing
#
you because the doctor is not there.
#
So even in the public system, you're being seen by a ward boy or a compounder or somebody
#
who's not actually formally qualified, but who just happens to be there, like, you know,
#
So in that way, the private guy who's less qualified, like, you know, is not giving you
#
any inferior care now, but going back to something we said in the education episode, the absolute
#
quality of care is abysmal, right?
#
I mean, so it's just like those two bears I talked about last time, right?
#
The private guy just has to be slightly better than the public guy.
#
That's why improving public is so important, right?
#
Because it forces the private guy to up his game as well.
#
So what you have today is overall, the public system hires more qualified people, but because
#
it doesn't have its governance act together, right, I mean, and you can't get accountability
#
for a variety of reasons, that the effort is really low.
#
The private guys have lower knowledge, but much higher effort.
#
And so put together, the composite result of that is that the private guy is at least
#
Slightly better, but at least no worse.
#
And just like education, right, so just like my school voucher study, we found exactly
#
That the private school teachers are much less qualified, but they work much harder.
#
So the net-net result is that the value added is comparable, but they do it at a much lower
#
And that's because the salaries of the private school teachers are about one fifth.
#
And so similarly here, what you'll see is that the cost per patient in the private sector
#
is substantially lower because most of these guys, you know, they charge fees, whereas
#
once you add all the salary costs in the public system and divide that by kind of the utilization,
#
you get a much, much higher cost per patient for comparable quality.
#
So I'll stop there and then come back to this other even more cool study where we then look
#
at the dual sample of the same doctor in his public and private practice.
#
But let me just pause for a moment on this representative sample and get you in there.
#
Like, first of all, as an aside again, like one of the statistics that we know about education
#
is also true of healthcare.
#
I found out while reading your chapter in these papers, which is that even though people
#
have free public options available, they still prefer to go by and large to private doctors
#
and that in itself is telling.
#
Now I'll quickly summarize in a sort of two separate buckets, a bunch of insights that
#
really struck me and you can tell me where I'm making a mistake or you'd like to add
#
And one is the urban bucket, which is what these DAS and Hammer studies look at specifically
#
in Delhi and all of that, where they kind of point out that look, there is a stark difference
#
between the private guys and the public guys, but there are nuances within that.
#
Like number one, the private guys always show more competence and more effort.
#
But then within the private guys, there are two categories.
#
One is the MBBS guys who are actually trained and the other is the untrained guys.
#
Within the untrained guys, there will be the Ayush guys, who you could argue have been
#
trained wrongly and the apprentice type people who are, you know, picking it up from a proper
#
doctor and kind of taking that.
#
And everywhere you find that not only does private outperform public, but in different
#
localities, you know, private and public will both rise, so the difference between them
#
And in poorer localities, whether it's private or public, it will be much worse than in a
#
So the locality matters a hell of a lot.
#
Now that is the urban India card, the little snapshots that we got.
#
In rural India, what I kind of found fascinating was how in your paper on two Indias, you've
#
described how people travel for medical care, where you write quote, people may visit the
#
clinic in their village, but if they have something more serious, they will go to the
#
highway and if they have something really major, they will take a bus and go to the
#
So people will almost never travel from one village to another, stop quote.
#
You also point out that quote, a high quality provider who is posted to a village will spend
#
a much smaller portion of the day seeing patients compared to a provider in a district hospital,
#
And then you make a point that you also made in the education episode in that context,
#
that sometimes better roads are more useful than more rural clinics, that you want to
#
connect a patient with a healthcare.
#
And if you can't take Mohammed to the mountain, you get the mountain to him, you know, so
#
if you can't provide that excellent medical service at his village, if you just provide
#
better roads, better maybe mobile connectivity or whatever, then he can access that better
#
But here I have a question, like the vast majority of the care that is kind of being
#
accessed is private here and you've pointed out that, you know, in your study, you find
#
that there are 3.2 doctors per village and only 2.8% of them have an MBBS, right?
#
And while there is variation between states, the state of affairs is kind of ubiquitous.
#
So now when I connect this with how you describe the travel, that if it's something immediate,
#
like if I have diarrhea or a cold, I'll go to the local guy.
#
And if it's something a little more serious, I'll go to the highway.
#
And if it's something really, really serious, I'll take a bus to the nearby town if it's
#
Now what this therefore would imply is that even if the so-called doctor is not that trained,
#
doesn't have a degree, the point is that, you know, a common cold gets better on its
#
It's a reversion to the mean kind of thing.
#
Many of these common things that they would go for, like even diarrhea will kind of get
#
better and it will almost be folk wisdom that have, you know, salt and water and all of
#
So you know, is that's what playing out there that they're going out to a private doctor
#
in a situation where it's not serious to begin with and it doesn't really matter if the doctor
#
I mean, if you have 3.2 per village and only 2.8 per have MBBS, they obviously can't treat
#
For the serious stuff, you're going elsewhere anyway.
#
So it's almost like placebo doctors.
#
I mean, the reason a lot of quackery appears to work to people and I have written frequently
#
on this in various places, the reason a lot of quackery appears to work is two reasons.
#
One is of course a placebo effect, which is far more powerful than we realize, which is
#
why every proper piece of medicine has to pass a double blind placebo test.
#
And by the way, no homeopathic medication has ever passed something like that.
#
And the other reason is reversion to the mean in the sense that many diseases just happen
#
and get better on its own, including in the vast majority of cases, even COVID where you'll
#
get fever for a week or you'll be asymptomatic and it will just get better.
#
And if at the peak of your symptoms, you take some medicine, you will ascribe it to that
#
correlation and causation and you will imagine that's what is causing the problem.
#
So is this part of the situation that the fact is at one level, I get it that Indian
#
Medical Association now keeps protesting that there are so many quacks.
#
Now at one level, I get the argument that listen, the state isn't there, proper doctors
#
aren't there, but these are people who might be able to help in some way, so let's not
#
But at the same time, could these just be placebo doctors in the sense they're not doctors
#
And these are not, you know, you feel good that you're going to someone and he's giving
#
you medicine, but you were going to get better anyway.
#
So I think, you know, I don't know the full answer to that question, right?
#
Because we didn't do a study on placebos.
#
Now what I can say is that because we are codifying the quality of treatment, right?
#
Because in these SP cases, we know, right, we know the underlying ailment and see, you're
#
absolutely right that in the majority of the case, you actually don't need medication,
#
But the truth is, the reason you want the doctor is not for your average case, but to
#
be able to say, are you in that 5% of cases where you actually need the attention, right?
#
So in fact, the SP case that we have is a case where the patient walks in and says,
#
I have had some pain in my chest.
#
And now this could be anything from a heart attack on the extreme, or it could just be
#
that I had a slight sprain and that I slept in the wrong posture and like nine times out
#
of 10, it will turn out to be completely benign, correct?
#
But the reason you go to the doctor is not to get the treatment, but to kind of get the
#
pathway of questioning to say, is there so, which is why the right questions you ask her,
#
you know, what is the nature of the pain?
#
Is it a throbbing pain?
#
Is it a vibrating pain?
#
Like is it a pulsating pain, right?
#
So the point of the doctor is not to cure you, but to then identify whether you need
#
referral to something more, right?
#
I mean, so the, which is why the cases are designed in a precisely, right?
#
I mean, to reflect symptoms that I think in one of the cases, it's like, I have a hard
#
So now it could be asthma, which needs training, or it could be just an allergy, or it could
#
be, you know, some short-term adverse reaction to something.
#
So again, the cases are designed in a way as to be cases where you can measure both
#
omission and commission.
#
The bottom line is coming back to the absolute quality of care in our data, right?
#
If you say what fraction of these providers correctly give you the correct treatment,
#
It is only about 15% in the public sector and about 22% in the private.
#
So that's why I said the private does better, it's 50% better, but in absolute levels,
#
it's still abysmally low.
#
Now, but these guys are mostly unqualified, but where it gets very interesting, if you
#
look at public versus private is because the problem here is the qualifications of these
#
guys are different, right?
#
So if you want to really isolate the effect of private sector incentives.
#
So the really cool part of the study is then we found the same doctors, right?
#
So the same guy in his public practice and private practice.
#
So now the knowledge is constant.
#
And what we did it over a period of time, he got the same case in both his public and
#
private practice, obviously with different actors, like, you know, it's not Amit going
#
to both the practices, like, you know, I mean, but the same case.
#
And what you saw was that the quality of care was better in the private practice on every
#
So there is higher effort, there is more likely to be a correct treatment, you know, there's
#
less likely to be unnecessary treatment.
#
So in every way that there is more effort being put in.
#
So to answer your question about the larger gamut of rural healthcare and saying how much
#
of this is just placebo, I think, you know, my view on this is that, yes, only about 15
#
to 20% of the cases actually got it correct.
#
Now, does that mean the remaining 80% of placebo?
#
Possibly no, because a lot of what they're giving is also what we call palliative relief,
#
Like I mean, so it may not cure your underlying symptom, but it is still giving you some utility
#
in terms of, you know, making you able to go back to work.
#
Now, the question, of course, is that steroid or whatever thing they're giving you as a
#
painkiller, what long term cost is it coming at?
#
And then it goes into this very complicated issue of what is the patient's discount rate,
#
So if this is a patient who says I need to get better today, because if I don't work,
#
I don't have food to put on the table, then can you really blame the doctor or the patient?
#
Like, you know, I mean, for over medicating on the palliative stuff, if that's what allows
#
the patient to get back to work.
#
So that's what makes this so difficult.
#
Now, maybe I'll pause here and then we can come back and talk.
#
Maybe we should switch to talking about policy, right?
#
Because I think it's very, very, very difficult.
#
I think Ajay's view on this was we just don't even know enough that we just need to spend
#
Like, you know, I have some views at least on principles of how I would structure policy
#
based on kind of the work and the evidence we have.
#
And you know, maybe we should switch to that.
#
And that way we can, you know, bring all these facts together to say, because right now it
#
almost feels like I've given you this, you know, broad smorgasbord of facts.
#
But organizing this into saying, how do we take this understanding into thinking about
#
policy is then something that I need to spend some time on, you know, which we can do next.
#
So a couple of things I want to double click into, and this might take us into the realm
#
of policy, but even if it doesn't, we'll talk about policy anyway after this.
#
And here are the two things I want to double click into.
#
One is talking about public incentives and the other is private competence.
#
And when we speak about the public health system, right, we've got the five tier system.
#
We've notionally got all these doctors in every village, in every town, every district,
#
But there is high absenteeism.
#
The classical explanation for why private always performs better than public is incentives.
#
Because in public, you've got a tenure job, there's no accountability.
#
And you know, those are your incentives and that's how it functions.
#
So first question, what are the ways to improve these incentives?
#
Like once we take it for granted, because you know, the public reflexes that repeatedly
#
like an education, they will ignore the free option and they'll go to a paid private guy.
#
Clearly, something is seriously wrong.
#
Now, if we accept that the market can't take care of everything in something like healthcare,
#
then in that case, we have to focus on this and we have to figure out better incentives.
#
One way like we discussed is more local government, but that isn't going to happen anytime soon.
#
And that in any case, we will discuss in our episode on federalism.
#
So what are the sort of ways of improving these public incentives for public healthcare
#
So this is a great question.
#
And maybe I will park it for the policy piece.
#
But let me say one other important thing that what is really interesting about why we call
#
the paper to India is like if you look at say, Tamil Nadu, Kerala and some of the better
#
performing states is that not only do they deliver better care, but they also deliver
#
it at a much lower cost per patient.
#
So how are they achieving higher quality and lower cost?
#
It's because the utilization of the doctor is higher, because the biggest cost in the
#
system is your doctor's salary.
#
But if people have confidence in the public system that everybody is coming, then your
#
cost per patient is actually lower and the equilibrium has unraveled where not only are
#
they not that good, but if you don't know if they're there, you're not even going to
#
So the patient doesn't go and then you kind of the whole thing unravels into a low level
#
So I think it just highlights the broader value of investing in public systems.
#
Now this is spoken from somebody who has documented that the private incentives do better.
#
But my overall view in terms of architecting systems of service delivery is that the public
#
and private both have their weaknesses.
#
They have different pathologies, right?
#
So at one level, a public provider should be able to do better.
#
Because there is no profit motive and therefore you don't have to kind of charge more to make
#
the fat cat salaries of your corporate executives who are running the private firms.
#
At another level, they have scale in procurement.
#
So it gives you volume discounts, whether it's on your medication or your supplies.
#
You have scale in terms of cost of capital, if you need to borrow for infrastructure and
#
you have the natural scale that comes from being.
#
So combination of lack of profit motive and combination of economies of scale, all else
#
being equal should make the public a potentially better provider of value for money than the
#
But the elephant in the room, of course, is incentives.
#
So what the private guy has is much better incentives, right?
#
So essentially he doesn't get paid if he doesn't show up, right?
#
I mean, whereas the public guy, once you have the salary.
#
So you know, one very interesting way of thinking about public versus private in the COVID context
#
is like, you know, is the public unions fight to stay shut?
#
The private guys fight to open, right?
#
And that itself is a very telling this thing about, you know, the distinction of the incentives
#
because these you only get paid, right?
#
Now, and which is why I think the public versus private debate itself is so different and
#
should be different in different parts of the world.
#
So I think if you and part of the nuance in that AER paper, right, where we kind of allude
#
to the fact that Ken Arrow was not fully right.
#
It's always a dangerous proposition to start a paper saying that Ken Arrow was wrong.
#
So, you know, like the most brilliant economist of the 20th century.
#
But you know, the caveat is not that he's wrong, but that he was right only in the setting
#
You know, he had not actually seen what the wild west of kind of healthcare in rural India
#
So in fact, as an aside, sometimes it's interesting, right?
#
How some of the great ideas in modern applied microeconomics that won Nobel prizes, like,
#
you know, for people like Stiglitz and Akerlof actually came when they spent a year in developing
#
So Joe Stiglitz spent a year in Kenya.
#
Akerlof wrote his Lemons paper from his time in India when, you know, it's like studying
#
development economics is a little bit like studying pathology in medicine.
#
So pathology is you look at the dead patient and saying what did not work, right?
#
So when something is working, you don't see how beautiful it is that all parts of the
#
system are working, right?
#
So it's only when something is dysfunctional that you then go and understand the importance
#
of things you took for granted institutionally, right?
#
And so similarly, for kind of Western educated economists who kind of model the world in a
#
certain way, you know, you put them in a developing country where these kind of markets don't
#
exist and then you kind of say, oh my God, wait, like, you know, there is all of these
#
problems of asymmetric information and then they kind of embellish their models for this,
#
you know, work that then went and won a Nobel prize.
#
But anyway, so maybe my point here is if Aero had spent time in rural India, like, you know,
#
he would have modified like, you know, some of his, his, his theorems as well.
#
But what we are basically showing, we have this theoretical model in that AER paper,
#
When we say that, listen, market incentives are both good and bad, okay?
#
Because what we see in the pricing equation is that the price a doctor charges is correlated
#
with observable measures of effort, right?
#
So the price is higher if I spend more time, the price is higher if I do more checklist
#
items and the price is higher if I give you more medication, okay?
#
Because those are things that the patient can see.
#
Now what is really ironic is that there is no price premium for giving the correct treatment
#
because the patient doesn't know the treatment is correct, okay?
#
But unconditionally, you are paid more if you give the correct treatment because when
#
you give more medicines, you are mechanically more likely to also give a correct treatment,
#
So it's like, you know, you throw the kitchen sink at it and something will stick, okay?
#
So but the subtle point is this that doctor incentives of say private incentives are good
#
because it gives you incentives to work harder for the patient to come back to you.
#
But it also gives you negative incentives for overprescribing, right?
#
And so the question is how do you find the right balance between having enough incentives
#
for effort but not enough incentives for over treatment, right?
#
And that's the delicate balance we're trying to find.
#
Now the traditional modern Western medicine approach for this is you train a medical ethics,
#
you know, you try to train with say Hippocratic oath and you kind of have a group of peers
#
and you have a strong liability regime that says if you are found to have violated what
#
is kind of standard practice, so there are other institutional ways of kind of checking
#
But the general sense in the US is that the better working health systems are ones that
#
don't give doctors too much incentives, but you want to weaken their incentives and focus
#
more on kind of intrinsic motivation of quality of care.
#
So if you look at say a Mayo Clinic where the doctors are paid on a flat salary with
#
not that much kind of, you know, bonuses based on procedures, the sense is that you get better
#
What Ajay was saying, what a CMC Valor does, like, you know, I mean, so at the high end
#
of kind of the healthcare system, the best systems are where you have well-trained doctors
#
who have a strong sense of ethics and who are practicing in a community of peers whereby
#
you kind of reinforce good practice because of the norms that are built around you.
#
Now the problem is like, you know, when you come to India, we are at the opposite end
#
Which is that the effort in the public sector is so low that the marginal incentives, so
#
here's what happens, right?
#
When I increase your incentives, I'm increasing incentives for diagnostic effort, but I'm
#
also increasing incentives for overtreatment, right?
#
But the returns to the diagnostic effort is diminishing in how much time I'm spending,
#
Because if my default interaction is 20 minutes with the patient, then the marginal returns
#
to more diagnostic effort is not that high.
#
But in India, if it's two minutes, right, then the marginal return to that extra diagnostic
#
So part of what we argue in that model is there's a crossing property that if you're
#
at a very low level of baseline effort, then having more incentives is actually better
#
because those benefits outweigh the costs, right?
#
I mean, whereas if you are at a point where the baseline effort is already high, now putting
#
incentives backfires because now I'm incentivizing overtreatment, okay?
#
So it just kind of highlights how difficult these issues are.
#
But coming back to the Indian setting and policy, and maybe we'll take a short break
#
and come back, is that architecting systems so that you have some incentives but not too
#
much is kind of what is the really delicate balance of this.
#
And hopefully we can talk, I'll talk through some ideas for that when we get there.
#
And my other sort of double-click question before we get there, one of course was about
#
incentives in the public sector, but the other is about the competence of the private sector.
#
In the sense you pointed out that look, most people are going to private providers, a very
#
small percentage of these are properly trained MBBS guys.
#
And even a large percentage of the MBBS guys aren't particularly competent when it comes
#
to all these checklists, vignettes and all that.
#
So then the question is, how do we solve for this?
#
Because like Ajay pointed out in his episode, he used a telling phrase that look, it's expensive
#
to manufacture a doctor, which is true.
#
And there are supply constraints like the Indian Medical Council had run an episode
#
on this once, you know, a group of doctors and they act like they kind of cartelize the
#
They, you know, restrict the supply of doctors and that's a separate problem.
#
And assume for a moment that that is just there, it is how it is, it is expensive to
#
So can something like AI play a part in this, like Ajay spoke about AI plus nurse.
#
Now that seems an interesting formulation to me because one, AI has come a long way,
#
like Eric Topol has this great book called Deep Medicine, for example, about how AI and
#
doctors can work together.
#
And two, when I think of this vignetting, for example, you have a chest pain or you
#
have a diarrhea and you go to a doctor and there's a checklist.
#
Now a checklist can easily be encoded in an app where, you know, it doesn't have diagnostic
#
authority or whatever, but you can go through the checklist and at the end of it, you can
#
say, okay, I don't need to go to a doc, I'll just drink some salt water and I'll see if
#
Or it's not pulsating, so I don't really need to worry, I probably slept in the wrong
#
I'm just thinking aloud, this is by no means a medical thing.
#
So can technology, in a sense, save us, not in terms of replacing doctors, but maybe as
#
a tool to doctors, maybe as a tool for these informal doctors where it's there with them?
#
And I think most of the policy solutions I'll talk about will have a technology layer, right?
#
And I think one way to think about this is that technology in any sector, whether it's
#
nutrition, whether it's health and other examples, technology is a force multiplier, but again,
#
for the technology to be effective, you need to first understand what the binding constraint
#
is, and then make sure the technology is targeting that binding constraint.
#
And one of the things we did not talk about in our education episode is we didn't talk
#
about education technology, right?
#
And so since I haven't done research on technology and health, but I have done research on technology
#
and education, let me give you this quick analogy, right?
#
Which is in theory, the technology can do amazing things, okay?
#
It can help give you high quality content with the best professors for everybody.
#
It can help make the material more interactive, more engaging, it can customize the learning,
#
it can shorten the feedback loop, it can do all of these great things, okay?
#
And therefore, you have a lot of tech evangelism in education, till you go look at the data,
#
if you go look at the actual RCT evidence on technology in education, you'll find that
#
the results are all over the place, okay?
#
There are cases where the impacts on learning are negative, there are cases where there
#
are zero, and there are cases where it is positive, right?
#
So you know, and it is negative in cases where people are blindly giving computers to teenagers
#
and who are then spending a lot more time playing games, you know, it's neutral in
#
cases where I just give you a laptop, okay?
#
And it is positive when I take the effort to integrate pedagogy with the technology,
#
So I think my quick answer to this is technology can be a huge force multiplier, but there
#
are two or three key caveats.
#
One is kind of making sure that it is designed to address the binding constraints and making
#
sure that we are designing the incentives for the innovation to solve at scale for populations
#
without purchasing power, right?
#
Because the problem is, if you look at innovation, innovation is typically, and this is true
#
in ed tech as well, right?
#
That innovation is driven at the right tail of the distribution where there is the paying
#
customer, that is what you really want is to direct that innovation, like, you know,
#
at the bottom of the pyramid.
#
And that's where, again, you need the government to kind of play a little bit of an enabling
#
role and maybe that's a good place to start, right?
#
Like, you know, when we get into the detailed discussion on policy.
#
So let's talk about policy now, you know, we know the problems and we know some of the
#
problems we in fact, we think we know the problems and then like you said, is very vexatious.
#
Where do we begin to think in terms of fixing these problems?
#
You know, so let me start first with some core principles of public finance, right?
#
Which is so Indian health expenditure is overall health expenditure is comparable to other
#
say South Asian countries, but public health expenditure is lower.
#
Okay, so there's a lot more out of pocket spending.
#
So I think there is definitely room for increase in public spending on health now.
#
But it is incredibly important to think about the cost effectiveness of that spending, because
#
again, a big part of the research is how much variation there is in different ways of spending
#
So let's start with the first principles of public finance of saying, if I were to have
#
a marginal increase in health expenditure, where should that go?
#
And where it should first and foremost go are in the public goods, right?
#
And that is public health, preventive health that works at a population level.
#
And again, as I said this, but this is so important, it's worth reinforcing again and
#
again and again, right?
#
I mean that, you know, oxygen cylinders and ventilators are not public health, right?
#
I mean, so the mind gets concentrated on the immediate life-saving device and the constraint.
#
But that is still essentially a private good that provides health care to one person.
#
And it is rival, it is excludable.
#
Now, what is public health?
#
Say in the COVID context, public health would be things like, and I'll talk specifically
#
about COVID later, would be things like universal masking, right?
#
That can mean where there are public goods and externalities, and we'll talk about that.
#
But even more basically, things like air pollution, okay?
#
So my first and almost counter-intuitive message on health policy is that sometimes the most
#
effective return on investments may lie completely outside the health department, okay?
#
So when we think about health policy, we think about what does the Ministry of Health do?
#
But the Ministry of Health basically, again, like Ajay said, thinks of itself mainly as
#
a PSU running the public health care system, okay?
#
But that is not health, that is not wellbeing, that's not quality of life, okay?
#
So for example, I would say the single most important thing we could do for population
#
health is actually fix air pollution in North India, okay?
#
And so the calculations on this just show that the costs are of the orders of, you know,
#
just several hundreds of billions of dollars, right?
#
I mean, and whether it's the life expectancy, whether it's the health adjusted life expectancy,
#
the returns on fixing that are huge.
#
Now, the problem is that we have boxed our environmental discourse in two ways, okay?
#
One, it has been framed as a trade-off between kind of industrialization and economic activity
#
versus environmental regulation, okay?
#
So the greens are essentially presented as being anti-growth, okay?
#
And so somehow there is this trade-off and then, you know, and there's also obviously
#
virtual signaling sometimes from the environmental folks.
#
And so that's one kind of trap we get stuck in, okay, of green versus growth, okay, which
#
Second mistake we make is we kind of get caught in the politics of kind of global climate
#
change negotiations, right?
#
I mean, where we say that, listen, the West has polluted.
#
And so now, like, you know, we need our space to pollute, okay, like, you know, for us to
#
But, you know, but that is like cutting your nose to spite your face because forget the
#
global externalities, right?
#
Just the domestic health externalities alone, like, you know, make this a first order crisis
#
that you want to work on, right?
#
So now, and that brings me back to why I actually think the high petrol taxes are a good thing,
#
And that's because, see, holding as a government when you need to raise revenue, right?
#
Any government needs revenue for its core operations, right?
#
So, what is public finance 101?
#
Again, it's very obvious, but we don't really think about it this way because most governments
#
in the revenue department thinks about their target for how much revenue I'm raising, okay?
#
But there is a separate point of not quantity of revenue, but quality of revenue, right?
#
What is the quality of your revenue?
#
And there the public finance 101 is to say that you want to tax bags and not goods, correct?
#
So if I tax labor supply, I am reducing your incentives to work and that is bad, okay?
#
But if I tax alcohol or if I tax pollution, right, I mean, then I'm raising my revenue,
#
but I'm also getting the price right or closer to what is socially optimal, okay?
#
So just like those first principles, right?
#
Now, of course, you could argue that with, say, high petrol taxes, there is inflationary
#
pressures across the economy and all of this kind of stuff.
#
But overall, right, I mean, my meta view is given that the government has certain revenue
#
requiring, you know, revenue needs, that you're much better off, right?
#
Like in taxing carbon, right, I mean, through the form of high gasoline taxes.
#
Now, it's a separate matter that that is being used to hide and cover a bunch of other fiscal
#
But my preferred way of communicating the health value of the petrol taxes would be
#
to say that you justify this and saying we are doing this because we're going to take
#
these taxes and reinvest in public transport, reinvest in, you know, public health.
#
So one of the chapters in the whole book is on the revenue side of public finance, right?
#
So I think one of the other problems of state capacity we have in India is that the fiscal
#
compact between the state and the citizen is broken, okay, which is because the tax
#
paying classes have completely seceded from being recipients of services, you kind of
#
see that the taxes going into this black hole called the government that waste all your
#
money, which means tax evasion almost becomes, you know, quasi justified in the evading classes,
#
saying, like, you know, anyway, the government is going to waste this money, okay?
#
So part of state capacity is also reestablishing the fiscal compact, right, mean that as citizens
#
when you pay taxes, you can expect a certain amount of public goods and services.
#
And this takes me back to urban, right?
#
So one of the I talk much more about property taxes.
#
And why I think we need more decentralization property taxes is it allows you to build a
#
better connection in the taxpayers mind between the taxes that are being paid in the services
#
that are being received, as opposed to taxes going into this black hole called the government,
#
So if you did the same thing with the carbon taxes and saying, yes, I am charging this
#
for petrol, but there is a reason.
#
And that's because here are all the environmental costs, here are all the congestion costs,
#
and we're going to take this money and we're going to, you know, I think government of
#
Delhi got pilloried a little bit for saying, like, you know, free bus tickets for women
#
But you know, subsidizing mass public transit is actually a wonderful thing, like, you know,
#
because you're also cross subsidizing that from the car owning class.
#
So they might cry, you know, from the rooftops, but that is good policy.
#
It's good policy to increase the cost of carbon, like I mean, and use that for, you know, higher
#
density, more efficient mass transport.
#
And if you were able to make that ring fencing so that the taxpayer understands that, I think
#
that's actually a good policy.
#
So that's, you know, just one aside on that.
#
Now the other places where the highest returns to public spending may happen outside the
#
health ministry, we already talked about this, is in the context of infrastructure, like,
#
you know, so roads was one piece, but broadband is another.
#
So then coming back to how do you get quality of care to remote areas, right?
#
Because the problem today, we talk about doctor absence, but frankly, you know, I remember
#
presenting this doctor absence paper in 2004, some at the World Bank.
#
And I think somebody from Russia had this very insightful comment where he said, you
#
know, even Stalin could not get his doctors to go to remote areas.
#
Like, you know, now maybe the Tamil Nadu Stalin will succeed because it's, you know, it's
#
But you know, even the Russian Stalin could not get doctors in remote rural areas because,
#
you know, they're just, they're too educated to want to live in the rural areas and the
#
utilization is too low.
#
So one of the ways technology or broadband or these things can help is them leveraging
#
that kind of higher qualified doctor to be able to cater to larger populations.
#
So now the question is, when we look at the infrastructure budget, the health ministry
#
is not even thinking about this, right?
#
So at the very least, what I would like is to, a finance department is thinking about
#
these allocations, find a way to quantify say the health benefits or something like
#
And, you know, and therefore put correspondingly more money in some of those broader public
#
goods that have higher returns.
#
So that's one principle of public goods.
#
Now the other piece which I'll then come to is public health, preventive health, primary
#
And I'll come to that in a moment, but I'll pause and let you come in.
#
No, I will just sort of express my disagreement about the petrol thing because number of reasons.
#
Number one, even this government, I hope they don't get the idea from listening to this
#
podcast, but even this government hasn't presented it as a syntax or a carbon tax or as equivalent
#
to, oh, we are taxing alcohol so the guy won't go and beat up his wife.
#
Secondly, I think you think of it only as something that emits carbon and pollutes the
#
air and is bad for health is a problem because I think the positive externalities are far
#
greater in the sense that I don't see how you can argue for building roads at one end
#
and at the same time for taxing petrol more at the other.
#
I think you should build roads and not tax petrol or keep the taxes as low as possible
#
for the same damn reason because people getting around becomes, you know, so much easier and
#
that's so much better for the economy, that's so much better for everybody's overall.
#
But the simple thing here is you build the roads and kind of you subsidize the mass transit,
#
So you do the buses and not the cars.
#
You see, that's the basic point.
#
So there is still higher density transport, right?
#
All else being equal, higher density is going to be better.
#
Sure, but my point is that after you fix that public transit, then you do whatever when
#
See, what is happening here is that the state is failing in all these ways and then it is
#
and this is a common story in India.
#
The state will fail completely and it will tell private guys you also can't do and this
#
We don't have that public transit, which brings me to my other objection where you speak about
#
how it's a good idea if you tax carbon because then you can reinvest some money in public
#
transport and this and that and all of that.
#
But no, the money is going into statues and central Vista and all of these other things.
#
You can't have it both ways, right?
#
You can't say I want state capacity and you can't.
#
So I think, but this is the challenge, right?
#
So there's the public choice view that says, listen, like politicians are veenal and they're
#
going to take our tax money and essentially kind of do their vanity projects, in which
#
case you should shrink the state and just kind of get away, you know, just forget then
#
why are we talking about state capacity, right?
#
Because we don't trust the state.
#
Now the flip side of that is that this goes back to my point about the plane that is India,
#
So the private option essentially has worked for the people who are flying first in business
#
I mean, who have kind of seceded from the state.
#
But the reason there is no question of kind of giving up on the state is multiple things,
#
So one is there are certain core functions like justice and law that the state has a
#
Second, in terms of actual public goods, public health, environment, vector control, you know,
#
again, private for all the private philanthropy and initiative, there is no substitute for
#
the sheer scale and coordination that comes from getting the government to do its job.
#
I think the answer is just more engagement by the elites, right?
#
I mean, so in the end, so in this way, I think I'm more, you know, on the Ashwin Mahesh side
#
I think you guys had the same back and forth, right?
#
Where Ashwin was talking about, you know, what we need to do to take back the government,
#
No, no, I agree with that aspect of it.
#
My problem here is that you are proposing a real world action, which is tax patrol more
#
based on a utopian argument that, you know, that it will be okay to tax patrol more if
#
the state can do all of these things.
#
No, no, but think about...
#
The point is the state is not doing any of those things.
#
And the other point I want to make is that when you talk about the India plane and the
#
rich can buy private goods and all of that, the point is roads are good for everyone,
#
And I would argue that keeping petrol prices low are also good for everyone, including
#
So I think, let's put it this way, right?
#
There are two separate questions, okay?
#
One is what does the government do with any amount of money, okay?
#
The second is saying that we want the government to make certain core investments, right?
#
So suppose there is a menu of investments post-COVID, right, that we say these are clearly
#
high return public investments, okay?
#
Now there is always a possibility of saying we will not raise taxes, let's cut down all
#
of our inefficient spending, okay, and then move it to the more efficient ones, okay,
#
which would be my preferred outcome, okay?
#
Now the political reality of reforms and expenditure is that it's very, very hard to change existing
#
What you can do is when there is additional money, you can use that in a better way, okay?
#
Now one way is to say that along with growth, as the budgets grow, you know, you use that
#
marginal extra money to do better things and don't raise taxes.
#
But on the other hand, if the Department of Revenue is given a target and saying like
#
I want you to raise so much money, okay, and that is coming from the political bosses.
#
Now the Revenue Department then has a choice of saying am I going to raise income taxes,
#
am I going to raise GST, or am I going to raise petrol taxes, or am I going to raise
#
My submission is very simple, that conditional on a given target of revenue, the petrol tax
#
and the alcohol tax are better forms of tax than income tax and GST because this is disincentivizing
#
a good and that is disincentivizing all else equal, something that is generating a negative
#
So for example, look at the Tamil Nadu versus Bihar, right, I mean the Bihar model of prohibition
#
is like fiscal folly written all over it, right, whereas the Tamil Nadu model is saying
#
like, you know, people are going to drink.
#
So instead of taking that underground, like I mean, you know, the TASMAC system is responsible
#
for a very large part of kind of state revenue in Tamil Nadu, but then they send it back
#
into welfare programs and put money into women's accounts, which is, I think, overall a much,
#
much better way of doing things like, you know, I mean, in the Bihar model.
#
So anyway, I think there is a meta discussion about the size of the state and the fact that
#
the state is this and this is classic Musgrave Buchanan, right, public economics versus public
#
And I think here's the other problem, right, the other problem is where you are in India
#
is again, because of the political pressures, the pressure on the government doing things
#
is always going to be there.
#
So you and I can sit and have a hypothetical discussion of saying government should not
#
be doing this, but pressure to IEGA revenue to the taxman is coming after you, you like
#
So all I'm saying is that holding a revenue envelope constant, the revenue, the, the petrol
#
tax is a better way of hitting that than something else.
#
But what it really amounts to is, see, I think a tax on petrol is a tax on the poor, because
#
it will have an inflationary effect on everything else.
#
The prices of everything goes up, including the salt and the soap.
#
But it depends on what you're doing with that revenue, right?
#
So if that revenue is then being coming back into funding, say, like, you know, the primary
#
health care, the poor use, if the revenue is coming back to fund, say, an extra worker
#
in the ICDS system, then it in net net ends up being much more progressive.
#
So again, if you know, but you know, the thing is, if my aunt was my uncle, she'd be a man.
#
This is like that only farfetched.
#
My point is that what is really happening here is that the rich sitting in their first
#
class in the plane are saying fine tax support, you know, so anyway, but we can discuss it
#
But again, but it goes back to kind of saying all I'm saying is, yes, we have a real problem,
#
We have so and what is interesting is I get attacked from both the left and the right,
#
Like, you know, so the left thinks like, you know, sorry, I get attacked from the left
#
for pushing so much in terms of cost effectiveness, cost effectiveness, cost effectiveness, right?
#
Like, you know, I get attacked from the right, like, you know, for feeling that I have this
#
naive kind of optimism in the state, right?
#
And so I guess the middle ground I'm trying to occupy is to say, at one level, the state
#
needs to do more things at another level, it is so inefficient, right?
#
Like, I mean, that let's push it to kind of increase on the things that are cost effective
#
And then conditional on doing that, like, I mean, is it worth funding?
#
Like I said, in the context of the book with Kelkar, if the marginal cost of raising every
#
rupee of tax revenues, in fact, three rupees, then the ROI threshold of what you do with
#
tax revenue needs to be over 3x for you to actually do the tax.
#
So that I'm fully agreed that we should resist kind of a blind expansion of the state because
#
the default quality of expenditure so bad.
#
Now, you may come back and say that, listen, in fact, I was talking to a finance minister
#
last year who wanted to raise tax revenue, I told him, I said in good conscience, I cannot,
#
you know, I cannot just commit to raising revenue unless you commit to me that you're
#
going to do good things with it, because otherwise you're just kind of attacking the poor taxpayer,
#
Like I mean, for your next vanity.
#
But this was a government that had some commitment to, you know, service delivery, right?
#
But that was a side conversation in which we did have.
#
So anyway, so I think coming back to where we are in health policy and yeah, I'll put
#
a final aside because you mentioned Ajay's episode.
#
And I'll again point out that Ajay and Vijay Kelkar wrote this great book called In Service
#
of the Republic and we've discussed this concept of how every rupee spent by the government
#
has an opportunity cost as it were of three rupees from the economy, three rupees of growth.
#
So we discussed that in an episode which I will link from the show notes.
#
And finally, I will just, you know, make one little point that it is not that I am criticizing
#
your optimism of what the state can do.
#
I am saying that, yes, I admire the effort to make the state more responsive and to make
#
it do better things, but that should not take the shape of the assumption that the government
#
will do everything right, which then becomes a justification for all kinds of nonsense.
#
You know, just as we talk about market failure, we never talk about government failure.
#
We compare market failure with a European idea of what the state will do.
#
But the point is government failure is more ubiquitous.
#
I'm not making that case at all.
#
Most of my research has in fact been about documenting the government waste, right?
#
So I think in that way, again, my view is closer to Ashwin, right?
#
I mean, that you can't say that, listen, the state does bad things because we are the state,
#
I mean, so to somehow I think to assume that at some level, the state can do it.
#
You know, I started to interrupt you, but I'll object to that, that saying that we are
#
the state because you could have made the same argument during the British times.
#
I would argue that the state that we are saddled with today is the same colonial state with
#
a different set of rules.
#
No, no, but the big difference, now I'm forgetting, you know, how would that argument have gone
#
down if you were to tell Gandhi or Nehru or Savarkar or whoever, that, listen, that, you
#
know, this is, we are the state.
#
No, no, but here's the big difference.
#
Here is the big difference, right?
#
And I can't remember which, I think it was Ajay again who said this, right?
#
Because you keep saying state coercion.
#
I mean, the whole point of a democratic state is that any coercion that happens, happens
#
at the willingness of the people.
#
In fact, one of the sections of my final chapter is on making democracy work better, right?
#
And some of this has to do with institutional rules, reforms, whatever.
#
So essentially when I'm saying we are the state, I'm just saying that in a democracy,
#
If people will not engage, right, you can't then expect that, you know, magically good
#
things are going to happen, right?
#
So, you know, a different way of saying this is if the competent refuse to participate,
#
the incompetent will take over, right?
#
I mean, so you can't sit and then complain about what the state is.
#
And which is why my steer to people in civil society who care about improving the public
#
good is yes, you see, in a way, competent people want to kind of be in a setting with
#
their locus of control is high, right?
#
So you don't want to take on the state because it's this beast that is so far outside kind
#
of your your grasp, so to speak, right?
#
But again, given that that is the beast that is spending these hundreds of thousands of
#
The returns to fixing that beast or tweaking the beast to start getting more effective.
#
And like Ashwin said, like starting to engage with local government, starting to engage
#
with like, you know, the public sphere in your immediate vicinity is in fact a good
#
So in fact, in the education episode, I didn't mention this, but Howard Gardner at Harvard,
#
I think had, you know, recognizing that education is the needs of an education system are different
#
to different parts of the socioeconomic distribution.
#
I think had this wonderful quote once where he said that, listen, first generation learners
#
need like functional skills, right?
#
Middle classes need more engagement.
#
And that's because you're all caught in your own treadmills of, you know, meeting your
#
whatever private deadlines and kind of oblivious to the public.
#
And elites need more ethics, right?
#
Because the elites kind of have the luxury of kind of not being on a daily treadmill.
#
But like, I mean, if you have kind of created a culture where the elites are more about
#
how do I park my money in tax havens now at an individual level, you don't blame them
#
because they're like, okay, this beast of the state is going to waste my money.
#
So, but then you're starting to see at least in India, a lot of philanthropy, a lot of
#
kind of elites really starting to engage in shaping kind of the larger public ecosystem
#
My plea then is to the middle classes to say that we need more engagement, right?
#
You can't step aside and somehow think because if you step aside, then your only option is
#
to shrink the state because if you're stepping aside, you're basically saying, listen, the
#
state is unredeemable and it is kind of downward spiral.
#
And then we want to shrink the state and then give up on state capacity altogether.
#
But before you move on, one final statement that, you know, I think you're being a little
#
totalitarian when you speak of the people as if it's a monolithic hole and the people
#
voted these people in or the people keep the state accountable and all of that.
#
You know, that's not true.
#
We have to think at the level of individuals.
#
You know, there have been tens of thousands if not hundreds of thousands of deaths that
#
have been caused in India that are needless because of the apathy and neglect of the dysfunctional
#
I'm not responsible for that.
#
You can say, hey, the people are the state and all of that.
#
But I am simply not because I have at a personal level been warning that there will be a second
#
wave, been saying vaccinate everyone, been railing against the dysfunctional state ever
#
since I started writing or possibly since I was in the womb, because that's just what
#
But anyway, as you and I are going to have a continuing discussion, I don't even think
#
we deeply fundamentally disagree in terms of the facts.
#
I think we actually agree remarkably in what the facts are.
#
And then I think, you know, the creative tension is more in terms of how do you move forward?
#
Like you know, I mean, from this morass, like, you know, of where we are.
#
So, you know, I don't think I disagree.
#
I don't think we disagree on the core diagnosis at all.
#
So yeah, no, I'm with you on the effort that we have to make in reforming the state.
#
But I also sort of anyway, let's leave it for some other discussion as my whole podcast
#
is a continuing evolving discussion over hundreds of episodes with different people.
#
So God knows which guest will next contribute to this discussion.
#
Let's move on to sort of the policy.
#
So coming back to the policy on health care, right, I think so we've talked a little bit
#
about public finance, public expenditure and value for money right now.
#
But thinking about the creation of, you know, so I'm going to say two or three, I'm going
#
to say a few things on strengthening public systems.
#
And then I'm going to say a few on the private system, given that the private is 75% of actually
#
the distribution of health care in the country, right.
#
So on the public side, I think the single most important insight is the following, okay,
#
which is we have the most expensive part of the health care system is the doctor, right,
#
both in terms of the cost of the training and in terms of the scarcity of supply.
#
And we are living in this model of saying we're going to provide universal health care
#
by staffing rural places with these highly qualified doctors.
#
And that has just not worked and it is not going to work.
#
So the way we need to think about building our human resources for health care, instead
#
of saying, I'm going to take a talented urban doctor and put them in a rural area.
#
It's like, can I take a talented, motivated rural person and upskill this person in a
#
way that allows them to be effective in their communities?
#
So I think that's like the core mental shift we need to make.
#
And we have, and this is not just me, right.
#
We now have good data and evidence.
#
If you look at what is the part of the healthcare system in India that works best, okay.
#
What works best is we do well on vaccinations, we are doing well on institutional deliveries.
#
So we have key metrics of vaccinations, we managed to do this year after year with kid
#
Now, why are we succeeding is because the most successful part of the healthcare system
#
right now, I would argue are the ASHA workers.
#
So the ASHA workers are hired locally in the village.
#
They're typically eight, 10th or 12th pass, like, you know, they're in that range of
#
They provided a modest amount of training and importantly that they are paid on a piece
#
So they get a modest stipend, but they are paid for every institutional delivery.
#
They are paid for every vaccination.
#
So which means that you have taken the local talent, you have upskilled them and you have
#
put them in an overall incentive regime whereby like I mean, there is some continuous reward
#
for ongoing performance.
#
You know, this model was developed by Dr. Abhay Bang and Rani Bang, like, you know,
#
in Gachiroli district in Maharashtra, you know, they have search and they've done tons
#
of work over the years of building health systems and capacity in the back of the beyond.
#
So taking that insight into how do we build human resources for kind of the public healthcare
#
My most important push would be something very similar to what I said for teachers.
#
I mean, which is, you know, think about aggressively expanding nurse training programs, right.
#
I mean, but with the nurse training programs should be heavily practicum based.
#
So specifically the model that I have in mind, whether it's for education of a nursing is
#
to say that imagine at every district level, you create a district nursing college, right.
#
I mean, that is equipped to train a few hundred nurses a year.
#
Right, you say that you're going to admit students, the top candidates from every panchayat.
#
And you will prioritize panchayats where they currently don't have staff in their health
#
So you are getting an inflow of the top talent, 10th pass or 12th pass as the case might be.
#
And the one good news from expanding school education over the years is now you kind of
#
have at the panchayat level, like, you know, candidates who have passed 12th and particularly
#
for girls, this is really important because the cultural barriers to working outside the
#
village means that the labor force participation falls sharply and you kind of get married
#
and have kids at 16 or 18 partly because of that barrier.
#
So being able to create the structure whereby you say this is a prestigious program that
#
you're admitted to, but the government is paying your fees.
#
So you're not actually paying out of pocket.
#
But what you're doing over a three year or four year period is you're saying I'm going
#
to get three months of modular nurse training.
#
That teaches me in terms of the basics and then nine months I'm back in my community
#
attached to the sub center, like reporting into the PHC.
#
So there's a regular nurse and doctor I'm reporting into.
#
So this then becomes part of your practicum training where you are starting to provide
#
And you do this for two years or three years.
#
And then at the end of three years, in fact, I can literally imagine, right, that they
#
have their own uniforms and they kind of, you know, sort of in a quasi official way,
#
you get like one stripe and two stripes and three stripes based on how many years of your
#
training you're finished.
#
So it also, these little subtle markers of status, right, mean that it gives you certain
#
official recognition in your community, in your village.
#
And so what you're doing, and this is where then technology comes in, right, because what
#
you're doing is not just doing the three months training, but you're then creating these kind
#
of smartphone based apps as well as say peer groups.
#
So that you are constantly exposed and have access to a community of peers as well as
#
So when you're in practice, you can refer both to your immediate supervisor, as well
#
as have access to a virtual network of inputs for how you deal with this.
#
The technology can become incredibly effective as a way of supporting continuing education.
#
So you have to pass an exam and you come in, but once you're in the government service
#
for 30 years or 35 years, there is no kind of ongoing test of whether you're current
#
So again, what the U S does is saying doctors have to be recertified every 10 years.
#
Now we don't have the apparatus to do that, but imagine that now every nurse in your government
#
system has a unique ID and they are expected to do a certain say 10 or 15 days of training
#
But now the training is essentially on an app where you have curated content.
#
And then at the end of every module of the content, you have a test.
#
So you can take this on your own time.
#
And in fact, my chapter on police, I talk about the same thing, right?
#
That one of the reasons of a police are so horrendously undertrained is because the nature
#
of their job is they have to be on call.
#
So they have to be available and they are so understaffed.
#
The system cannot spare these poor cops like I mean, to attend a training workshop.
#
But if you can modularize the training to fit it into the downtime during the day when
#
they're sitting in the Thana, like, you know, just waiting for things to happen.
#
And that's the same for nurses is the same for a lot of service delivery staff that there's
#
a lot of downtime, right?
#
When you're just sitting there waiting for a patient to come in, there are some times
#
when you have a lot of load, sometimes when you're just sitting, twiddling your thumbs.
#
So how do you use that time efficiently by kind of having modular training that you can
#
take and you take an exam that you show you have this done.
#
So and then once you've done this kind of modular nurse training, you give them a nursing
#
And the good thing about this is this is then valuable in both the public sector and the
#
So if the private sector is 70% of the system, you can say that you are creating the human
#
capital, the human resources to absorb into the public system and post them in the sub
#
centers and areas that are underserved.
#
But at a point, you're also creating the talent that allows the private sector to absorb them
#
So coming back to what the government can do, something that Ajay said again, is that the
#
short run elasticity of medical workers is basically zero.
#
You can't manufacture a doctor in a crisis.
#
So the most important lesson from all of this is like, how do we build a pipeline of human
#
resources for the health care sector, but do this in a way that doesn't make the mistake
#
of saying that I need super fancy qualified doctors everywhere.
#
What you really need is kind of the locally connected public health worker who is then
#
trained and leveraged with technology, like, you know, to kind of have access to.
#
And then this is a person who knows when they need to refer a complicated case.
#
So coming back to doing your diagnostics, doing a checklist, doing a bunch of things,
#
the technology then becomes a force multiplier that you put on top of the human resource
#
So that would probably be the single most important thing I would do is to kind of massively
#
So going back to your question of a philanthropist with a hundred million dollars, you know,
#
I would actually say creating these modular training programs and the templates for this
#
and working particularly in states like UP and Bihar, like I mean, to say, how do you
#
make these things happen?
#
And in the beginning you will need to bring, you know, you'll need to import your trainers.
#
You'll need to have some high quality people creating the content and the modules and stuff
#
So fundamentally, right, 90% of our problems in India are about supply constraints, right?
#
You expand supply and then, you know, the prices will fall and things will take care
#
So I think that's one core point.
#
Now a similar point with regard to doctors, okay, is what Chhattisgarh did with kind of
#
their rural medical assistant program.
#
And I sent you that paper as well.
#
So this they had done for two or three years where you say, again, you don't need like
#
this fully trained doctor sitting in the rural area twiddling their thumbs, but can you modularize
#
the training and provide enough medical training that they can be more effective?
#
And I think, you know, coming to doctor training, I think the same thing applies in the Chhattisgarh
#
experiment, which was unfortunately stopped after two or three years for, I think, complex
#
But the core idea was very, very simple, that you don't need this fancy six years of training
#
and put an MAPS in a remote rural area where this person has no inclination to be and is
#
not being effectively utilized, right?
#
So they had this rural medical assistant program where you had people trained for about three
#
years and then were eligible to practice in the rural areas, right?
#
So you have to start thinking creatively about expanding the supply in a way that the most
#
important mistake, let me repeat again and again, and this is true with all of our public
#
And that's because we recruit on the basis of a test and the basis of knowledge.
#
The people who will get this pass the exam will almost invariably be from urban areas.
#
And even like those who get in from other areas, the salaries will be so high that their
#
aspirations will be open.
#
So a big part of the absence problem is not that people are bad, it's that they're just
#
completely disconnected from the communities they're serving, right?
#
So to kind of take what we've learned from the ASHA model, to say take local talent that
#
is connected to the community and invest in upskilling them in a way that there's a pipeline
#
from the skilling to the employment, I think is one of the most important structural points
#
that we need to do in terms of building our healthcare capacity.
#
Again, these are more complex kind of reforms.
#
Now the other thing that is immediately actionable for a government, given that at the end of
#
the day, governments, they know how to build things, they know how to kind of just do more
#
The one thing I would say, which Abhijit Banerjee has also written about is particularly in
#
UP Bihar and places where the population has grown so much over the past kind of 20, 30
#
years, that there is a strong case for building a second district hospital, right?
#
Because right now you have one district hospital, but there's been this in the seventies and
#
the population has doubled.
#
Now the nice thing about a district hospital is you have scale.
#
And when you have scale, you can have kind of the community of practice, the absence
#
is typically much, much lower in these district hospitals compared to the remote rural areas.
#
The utilization is higher and it also becomes possible to then do my third idea, which is
#
kind of attach medical colleges, right?
#
Because the medical education problem.
#
So I'll come back to that.
#
So the second thing is build district hospitals.
#
This is something that government knows how to do.
#
Like, you know, I mean, just do it, build a second district hospital in kind of the
#
And given the investment in roads and people's willingness to travel, that's a case where
#
concentrating your resources in a scale facility makes more sense than saying, I'm going to
#
build these itsy bitsy little P H C's that nobody goes to.
#
So that's the second point.
#
The third then coming to medical education itself.
#
And if you go back to the two India's paper, I think one of the big reasons for why you
#
see this massive gradient in healthcare quality is the substantial investment in medical education
#
in the Southern States.
#
Now some of this is driven by private investment, which reflects the fact that it's expensive
#
and the Southern States are richer and so people are just able to pay more.
#
But from a policy perspective, medical education is controlled pretty much by the government.
#
Now of course, there's a ton of rent seeking there and like, you know, and that's been,
#
you know, the tragedy, but hopefully this is a crisis where the chief minister says,
#
okay, boss, I'm not going to allow rent seeking here because this is now really important.
#
And then the three things that determine kind of your medical education capacities, you
#
know, there's the facilities, there is the, you know, there's the actual human capital,
#
You know, who are the people available to do the training and typically these have been
#
attached to district hospitals, partly because that's where you have the doctors available.
#
So I think on the public system, those would be kind of the three things I would do as
#
far as healthcare is concerned.
#
Now there is obviously the technology piece of this and there is a question about does
#
the public system have the dexterity and the speed to kind of really leverage technology
#
the way you want to at scale.
#
So I think one thing which I haven't discussed in the book, but just like I had done in the
#
case of charter schools in the private school, there may well be a case to kind of partner
#
with public spirited foundations to come and say that if you were to kind of take over
#
a management contract, okay, for running a public health system in one district, right?
#
So it would still be public in the sense that you wouldn't turn away patients, you wouldn't
#
charge fees, but you would basically bring in a publicly motivated person to come and
#
look at their technology and innovation landscape and saying, how would I re-architect the system
#
with modern principles of technology and private sector, right?
#
And I don't think we've done enough of that.
#
I think we've done PPPs in terms of outsourcing, say tertiary care and saying in Aishman Bharat
#
or Arogya Sree that we will reimburse private hospitals.
#
What we have not done enough of is saying, can I bring real private sector management
#
expertise and horsepower into re-engineering the public system to leverage technology and
#
do things at scale, right?
#
So I think that might be something else that's really, really worth thinking about.
#
So many things to unpack there, but I won't unpack too many of them because you've been
#
so sort of clear and concise.
#
But I will point out that one thing that delighted me was that I asked Ajay a question in the
#
last episode about philanthropist and 100 million, and you picked up on it in this episode
#
and answered it without my asking it, which I find delightful, which is, you know, that's
#
how I like to see the scene in The Unseen, a conversation that goes across time and across
#
episode and across guests.
#
So I wonder what question that I asked you will be answered by some other guest who knows
#
maybe we could have an answer next week.
#
I also want to quote a little bit from your Two India's paper because just for the listeners
#
to clarify that there's a lot of activity happening at the local level, which isn't
#
The state also is not a monolithic beast, just as individuals are.
#
So you speak about the different views of primary health care and you say, quote, one
#
view of primary care in rural India is that it is available mainly through publicly operated
#
private health care centres or subcentres, which are sparsely located and understaffed.
#
According to this view, qualified doctors in India are mostly located in urban locations
#
and access to quality care in India is poor.
#
The other view is, quote, an alternate view agrees that even though access to qualified
#
providers in rural India is low, a wide variety of health care providers with diverse qualifications
#
has arisen to fill this gap.
#
And where what I found interesting was how you detailed out how different states are
#
treating this differently, this situation in the sense that West Bengal is trying to
#
Chhattisgarh, like you said, is reducing requirements, making it modular and saying, OK, you don't
#
have to do all six years.
#
You can just do three years or whatever.
#
You talk about how UP and MP are increasing transportation.
#
So that's how your access goes up.
#
You don't need a doctor there immediately.
#
And how the government of India is allowing alternate medicine, docs and wellness centres.
#
And I love this quote that you quoted the Indian Medical Association, which is resisting
#
this, obviously, because, you know, they want to keep it supply as restricted as possible.
#
And you've written, quote, in their view, for instance, training informal sector providers
#
is like teaching burglars to steal better.
#
Teaching burglars to steal better.
#
Isn't this what the IAS Academy in Mussoorie does?
#
No, no, to be fair, I'm kidding, I'm kidding.
#
This is my distrust of the dysfunctional state has also become a kind of a meme, I guess,
#
You know, there's lots to think about.
#
And once again, what I will say from having read these papers and having read the chapter
#
in your book, but even the papers itself, even your two India's paper is that there's
#
a lot of depth and nuance there.
#
And so I would just encourage listeners to get into the weeds because this is a problem,
#
I think, that is not just a problem for academics and policymakers.
#
We've seen that in this specific season, all our lives are affected.
#
The quality of our lives and sometimes our very lives themselves.
#
So do check out those papers and go a little bit into the weeds.
#
There are a lot of insights for me.
#
Let me turn your attention now to the current crisis, to COVID.
#
And of course, I have done many episodes on this.
#
And there are just like so many aspects of this to talk about.
#
But I have spoken about the economics aspects of this, the policymaking aspects of this,
#
So once before COVID, because I had finished the public part of the policy.
#
Maybe I'll take five minutes to talk about the private part of the policy.
#
And then come back to COVID.
#
Because I think, and again, see, in many ways, the episodes that I'm doing in education and
#
health are kind of mirroring the structure of the book in the sense that the book has
#
these six thematic chapters and the themes then cut across six different sectors.
#
So in that way, you'll see the parallels that are happening here.
#
And one of the most important kind of points I want to make is, and this is to any health
#
minister or health secretary anywhere who may be listening to this at some point, is
#
that the biggest kind of, I would say, myopia we have in health policy is that the government
#
for the most part only thinks about its job as running the public system.
#
Whereas we need the clarity again, that the government plays three different, very different
#
roles in the ecosystem, right?
#
As a policymaker, as a regulator, and as a provider.
#
And the way you see the private sector is very different in these three lenses, right?
#
So as a policymaker, the private sector is your ally because your job is to improve outcomes
#
for everybody, regardless of who provides it.
#
As a regulator, the private sector is equal to the public because you want to treat them
#
And as a provider, they are your competitor.
#
Now, the problem is that the government, the majority of its staff and human resources
#
are thinking and functioning as government as provider, right?
#
And so, which means you automatically distrust the private sector, you automatically want
#
to kind of cramp them in multiple ways and, you know, and sometimes justify it because
#
of course these are profit, you know, for every good private guy, there's an equally
#
So and policymaking by anecdote, you can do whatever you want, right?
#
Because you can find, you can find any story to justify what you want to do.
#
So, but coming back to the stark reality that 75% of healthcare in this country is delivered
#
If you are a thinking health minister or health secretary tomorrow who says that I care about
#
improving my health systems, the elephant in the room are these private guys.
#
So, what are you going to do about them?
#
Now, yeah, so the default approach to the private sector is basically to pretend that
#
Because they're illegal.
#
They're not supposed to be practicing.
#
And every once in a while, you'll have these drives of going and shutting down these clinics.
#
But the problem is they will come back a week later or two weeks later because they're in
#
They are highly respected.
#
They're the only source of care.
#
And so, there is demand, right?
#
I mean, so, you know, and I think the denial we are in is really, really counterproductive
#
given that if we care about population health, the largest segment are these folks.
#
And this is where, again, we have evidence.
#
So, there's this lovely paper again by Jishnu and Abhijit Banerjee and co-authors in Bengal.
#
I mean, where they partnered with the Liver Foundation and they ran this program.
#
Again, it was a randomized control trial of basically training these unqualified providers.
#
So, they ran a modular training over about a nine-month period.
#
And the whole point was to say, listen, you know, you are in the community, you care about
#
So, can we give you some basic training?
#
And just does that make you better?
#
And what they found was that this modest training had a substantial positive impact, right,
#
on their effectiveness as measured by standardized patients.
#
And it almost halved the gap between the completely unqualified and the fully qualified in about
#
nine months of training.
#
So, let me come back, therefore, to this core point that I made, that quality is a function
#
of knowledge and effort, right?
#
Public sector has high knowledge, low effort.
#
Private sector has low knowledge, high effort.
#
So, if you do a simple partial derivative, I'll ask your audience to do in their heads,
#
The partial derivative of quality with respect to effort is knowledge, right?
#
So, which means the returns to knowledge is actually higher in the private sector because
#
the effort is high, whereas the returns to effort is higher in the public sector because
#
the knowledge is high, right?
#
So, if you want to think about system architecture, we need to be thinking about how do you improve
#
the incentives and accountability in the public sector and how do you improve the knowledge
#
Whereas in practice, we do exactly the opposite, right?
#
I mean, because we say the private guys, we ignore them and we keep doing trainings of
#
Like in the public sector.
#
So, now, of course, this is politically fraught, right?
#
Because it will obviously be like opposed by different kind of interest groups.
#
And so, it's very important to be very clear that I'm not talking about training these
#
guys and calling them doctors, okay?
#
Like, I mean, it is going back to the spirit of modular training, recognizing that, listen,
#
these guys are in the community, they are respected, they don't want to do bad, right?
#
I mean, the last thing they want is to hurt their patients because they're living in the
#
So, giving them the ability to kind of be better and be able to refer complex cases
#
almost says that, listen, these private guys exist anyway, okay?
#
So, why not leverage them to be like the first point of kind of a structure of care whereby
#
I train them to say, okay, don't do this, don't do this, don't do this.
#
And now, here are things that you can do.
#
Now, there is a question about what credential you give them.
#
Like, you know, maybe you create a cadre of paramedics or a cadre of like, you know, something
#
that provides a certain legitimacy but doesn't say like, you know, that, but then you say,
#
if you call yourself doctor, like, you know, then I'll penalize you.
#
So, again, I can just give you principles, right?
#
Now, the details of this will then have to be worked out in specific states, specific
#
Now, one way to hybridize the two ideas I talked about was to say, suppose we have a
#
modular training program for nurses or we have this RMA training program, okay, whereby
#
say we are going to create these two-year programs or diplomas for these rural medical
#
Then in terms of who gets admission into that program, one route could be to say you do
#
well in a 12th standard exam.
#
The other route could be that if you are one of these unqualified but registered medical
#
practitioners who's been in your community for 10 years, that maybe will give you some
#
preferential access into this two-year training program, okay, so that we are able to take
#
that informal workforce and start giving them some formal skills, okay.
#
So, that would be one piece of how you may take this private system and make it function
#
better, okay, in a way that is constructive as opposed to saying, okay, I'm going to play
#
whack-a-mole with you and shut you down, shut you down, but then you'll pop up in another
#
Then I think the other things in the private sector, you know, one is obviously I think
#
something Ajay alluded to, you see, the biggest challenge in designing healthcare systems
#
is at one level you want some incentives for good care, at another level you don't want
#
the incentives to be too strong for over medication, right.
#
So, structurally, one way to do this is if you have insurance companies that are essentially
#
signing agreements with providers, okay, and saying that you are my designated provider.
#
Now, what happens is the members of the policy at the beginning of every year have a chance
#
to choose their provider, but the good news is you're now choosing not when you are sick
#
and need to be hospitalized, when you have very little bargaining power and very little
#
You are choosing at a time when you have kind of the ability to think about the other way.
#
So, the practice then has an incentive to have a reputation for providing a certain
#
standard of care because if you are somebody who's seen as crimping, then people will not
#
choose you, but the way you compensate the practice is not by saying I'm going to pay
#
You say I pay you a certain amount for every patient you enroll in the course of a year
#
so that you are not incentivized to overtreat, right.
#
So, again, if you think about, say, the British National Health System, the NHS has a similar
#
so as a doctor, as a primary care practitioner, you get paid a certain piece rate for every
#
patient who signs up with your practice, but then your reimbursements are relatively flat,
#
So, you don't get a steep gradient for doing too many procedures.
#
So, again, going back to what Ajay said, these are incredibly complex issues and, you know,
#
Western countries have entire communities of health economists, like who are, you know,
#
working with common data sets and starting to, you know, these are incredibly complex
#
design detail questions, right, but the broad architecture would take that form that you
#
say one of the ways to solve this asymmetry problem is you put a kind of a bigger bio
#
on the other side and over time you create the right incentives for the overall ecosystem.
#
And the last thing I would do going back to ideas for your hypothetical philanthropist
#
with 100 million dollars, like, you know, I mean, is that there are a lot of light touch
#
investments we can make that allow markets to function better.
#
Okay, I mean, even without being directly in the business of provision or insurance
#
So, part of the problem today is we have no transparent source of data on who the providers
#
So, if you're sitting in your city, you have no idea who's practicing around you, right.
#
And so, now, there are companies, obviously, information is valuable, right, so there are
#
companies that have tried to set up a business, say, I think in education, I was looking for
#
some schools and then you get these private companies that say school ratings, but then
#
the revenue model for them is either the school or the hospital is paying to get rated and
#
then you have the agency problem that you had that say credit ratings or the user is
#
paying to access that, which means it is restricted in the user base when this information is
#
fundamentally a public good.
#
So, as in public good in the sense that once you've created the marginal cost of spreading
#
So, if a government won't do this, which they should, but don't have the capacity of the
#
money to do, one incredibly leveraged thing philanthropy can do is just come and create
#
kind of open source information platforms like Amin that allows you to just crowd source
#
over time, think who are all the providers in a given market, right.
#
I mean, now, and then over time, you could have like a small cadre of dedicated staff
#
who will then go into field verification of each of these guys, right, like Amin.
#
So, make it a positive revelation game for people to want to be listed on the portal
#
as somebody who's providing services with a certain level of kind of qualifications
#
or basic details of what their practice is.
#
But I think we underestimate the value of just good information.
#
So, Jishnu has a series of very nice papers in Pakistan looking at information on school
#
I mean, and what they find is just providing better information on school quality without
#
any additional money, right, leads to huge downstream positive impacts over time because
#
it just, you know, it makes the market much more transparent and it forces you to react
#
to kind of what is possible and what you see in the market.
#
I mean, I have even more ideas, but I think these are enough to kind of, I think, you
#
know, clarifying some of these conceptual issues and hopefully providing some direction
#
of where we need to go.
#
These are stimulating ideas and I hope young entrepreneurs listening to this can take inspiration
#
I mean, why do you need a philanthropist necessarily throwing a hundred million dollars or the
#
evil dysfunctional state coming in and say, ah, we can do it.
#
We'll show Amit that we can also do some work.
#
You know, where is a private enterprise?
#
You know, there are so many areas to fix here.
#
These are the problems worth solving.
#
So kindly come in and try that is my message to all young entrepreneurs listening to this.
#
You are giving messages to health minister through my show, boss, what are you even thinking?
#
You know, but nevermind, you know, maybe, maybe the person who will be health minister
#
of India in 2035 is listening to this now.
#
So kindly, I hope your attention is good.
#
Let's, let's talk about the podcast will still be available.
#
I think, you know, I mean, one of the things I think you mentioned in your chat with Milan
#
Vaishnav, I think after the first episode with Patap is that, you know, the point in
#
some ways, yes, I think it makes sense to discuss health right now when we're in the
#
COVID moment, but it's almost more effective, I think, rather than making prognosis in the
#
fog of war to kind of focus on the first principles of just the larger structure.
#
And these things will then, you know, play through regardless.
#
Yeah, like my, my conceit was that I am making episodes for 30 years later.
#
So I don't want to go across space.
#
I want to also want to go across time.
#
And of course it is possible.
#
One reason I came up with that highfalutin idea was because I thought no one will listen
#
to me in the present time, but clearly that is not the case.
#
I think the idea to create timeless content is great.
#
And that's also why it's, it's so wonderful that you've come on the show twice, because
#
I think both of these episodes, education and healthcare are almost so useful in terms
#
of providing foundational understanding while we aim for the timeless, we also have to address
#
So let's, let's kind of get to COVID now.
#
Now on COVID, I've done episodes with economists of various types have done, you know, episodes
#
on it with relief workers who are out there feeding people with data journalists, with
#
epidemiologists, with scientists who are, you know, good at describing what the virus
#
And even though you don't call yourself a healthcare economist, you, you're probably
#
the closest to that definition with whom I guess Ajay also to a fair extent, because
#
you're both economists interested in who have looked deeply into healthcare.
#
So what's, what's your sense like, what could we have done differently, which is not obvious
#
Yes, I think I don't want to do Monday morning quarter backing because I think while there
#
is a space for accountability in our public discourse, I think the problem when you kind
#
of, and that's important, right?
#
This is not to say that there shouldn't be accountability for people who were asleep
#
But when in this moment, right, I mean, when you go down that road, you essentially create
#
Like I mean, and so I would like to kind of unapologetically focus on what do we do today
#
and going forward, right?
#
I mean, what has happened has happened, right?
#
So I think that really just three points, right?
#
I mean, which is the last year, we thought about ventilators this year, we are focusing
#
Like I mean, which is visceral because you see people dying and you kind of immediately
#
see that I need to get oxygen.
#
That's important, right?
#
But I think the most important message again, is what is the public health and what are
#
the things that the government should be doing that?
#
And again, optimizing the cost, see, one of, I remember reading somewhere that in, in a
#
war, any decision you make is going to have some casualties, okay, because there are only
#
So the point of leadership in a fog of war is to assess the risks and kind of be able
#
to take a reasonable decision that is informed by the best information available with a certain
#
amount of transparency so that the public continues to have confidence in the leadership,
#
I mean, and I think that is something which I will definitely say has been lacking and
#
consistency of communication and transparency to the public about where we are and where
#
we are going, I think is incredibly important.
#
So that being said, I would really just say three things, okay?
#
So there is obviously a piece of what we need to do that is just on the sheer logistics
#
of hospital beds or oxygen and stuff like that.
#
And even there, I think the important things are not so much getting somebody oxygen Twitter,
#
like you said, because like I was saying, that means somebody else is not getting it.
#
It's really intervening to reduce the frictions, right?
#
Is there excess oxygen somewhere and somebody needing it somewhere else?
#
Is that oxygen going to places where there is no trained operator to use it, right?
#
So we need kind of mindful interventions because there's complementarities across these different
#
So that I think is important, but I would say the single most important thing we need
#
See, the end game is vaccines, okay?
#
There's no question, but the vaccines are going to take a minimum of six to nine months
#
to roll out across the population, okay?
#
So what do we have evidence on today is the one thing that the country should just be
#
doing in a war footing is masking, okay?
#
Because we know a lot more about this disease in the past year.
#
We spent a lot of time cleaning surfaces.
#
We spent a lot of time doing other things.
#
What we now know is that the vast majority of transmission is airborne.
#
This is fundamentally an airborne disease.
#
And it is completely, you know, unglamorous, simple, but it works, okay?
#
And so the surgical masks, they just work.
#
Now, and again, this is a case where unfortunately, I think the public messaging was a little
#
mixed in the beginning because WHO in the beginning said masks are not needed, but that's
#
because they were trying to save masks for the healthcare workers.
#
So that was a case where I think the communication was actually wrong because it gave the sense
#
It is unambiguously clear that you need the masks, that they work, okay?
#
And we now have the supply elasticity of that is basically infinite.
#
You can get those masks at scale, okay?
#
So what we need is just a very simple public communication to the nation and maybe by chief
#
ministers, if not at the national level, to just say the following three things, right?
#
Which is saying that the lockdown is an incredibly costly last option.
#
The big cities are doing this because we have no choice for two weeks.
#
But the end game is going to come when everybody's vaccinated.
#
And between now and everybody being vaccinated, we need everybody to essentially take the
#
I would say the government should spend the money it takes to procure masks.
#
So my contemporary Mushfiq Mubarak, who's basically my batch mate, who's a professor
#
of economics at Yale from Bangladesh, in the past one year, he has put together a consortium
#
of engineers, medical experts from Stanford across universities, and they've really been
#
doing a lot of research on masking and both in terms of what is the most effective type
#
The surgical mask, what kind of filters, and they've been doing a lot of work on testing
#
different models of behavior change communication.
#
So they've done randomized control trials now in Bangladesh with different villages
#
doing different models, and they now have enough results from this.
#
I think that paper will be forthcoming and signed soon.
#
And they've done a series of kind of webinars, I think, in India.
#
But the steps are very, very clear, right?
#
Literally go door to door in every village, right?
#
I mean, procure four masks per family, like give them the mask.
#
And now, why does the giving the mask matter?
#
It matters because not only are you signaling to every citizen that this is important, that
#
the government is doing this, but you are also making it common knowledge that everybody
#
has gotten the mask, okay?
#
So it's no longer acceptable to be around without the mask because you know the government
#
has come and hand-delivered this at your door, right?
#
So they have procured a washable mask, and so this is washable, but a set of four means
#
you wear a different one like every week, you wash them and wear that.
#
Then there is this piece about you need an all of government approach on this, that it's
#
not just the health workers, but it's the gram, you know, it's the panchayat.
#
And in a way, the rural areas are even more important, right?
#
Because the health care infrastructure in the urban areas, at least you can do something
#
In the rural areas, you don't even have the health infrastructure.
#
So prevention is your only game in town, okay?
#
So you need an all hands on government approach saying between the health workers, the ASHA
#
workers, the Anganwadi workers, the ANMs, the village, all of the ward members, the
#
panchayatiraj officials, all hands on deck, right?
#
I mean, this is an all of government approach to saying, they're going to get you the masks
#
and this is going to happen.
#
Let the communication going in a standardized way, use your film stars, use your cricketers,
#
use whatever to just kind of reinforce again and again and again that there's the basics
#
And sometimes repetition is good that have the simple graphs that says this is fundamentally
#
So, you know, we obviously the epidemiologists need to worry a lot about variants and sequencing
#
But for the amjanta, that is irrelevant, right?
#
What is relevant is that this is an airborne disease and the way you kind of contain spread
#
is making universal masking till everybody's vaccinated.
#
And a simple way of saying this, you know, one thing I feel strongly about is that the
#
35,000 crore cost of universal vaccination, I mean, it's a rounding error, right?
#
Like I mean, compared to the cost of the disease or the cost of a lockdown, I think a nationwide
#
lockdown costs about 200,000 crores a week, okay?
#
And if vaccinating everybody costs 35,000 crores, like, you know, you should just write
#
a check from the government of India, not even ask a question and just get it done,
#
So similarly at a state level, if it's, I think our estimate on the masking right now
#
is that it will cost maybe 50 to 100 crores, like, you know, to do the procurement and
#
But the cost of a lockdown in an average size Indian state is about 2000 crores a week,
#
I mean, so, or sorry, 2000 crores a day, about somewhere between 1000 and 2000 crores depending
#
on the intensity of lockout.
#
So, and this is essentially, I feel so strongly because this is the core point of public health
#
versus curative health, right?
#
Which we've talked about throughout this class.
#
So sorry, I almost feel like I've made a class for this.
#
I'm perfectly happy to be a student in your class, no issues, yeah, we are all, me and
#
all my listeners, we are the biggest class you've ever had probably.
#
But no, I think, you know, but this is why I think it's useful to understand the first
#
principles and then be able to apply those first principles to the problem at hand, right?
#
The first principles is the ROI and spending on public health, dwarfs, like, you know,
#
so tomorrow, now we are already talking about needing to do cash transfers to kind of alleviate
#
the suffering from job loss.
#
Absolutely, that's correct.
#
But that palliative measure is going to cost you already 10 times, 15 times, 20 times,
#
what it will cost you to do universal, accelerate vaccinations in every possible way and get
#
the universal masking going for the next six to nine months, right?
#
I mean, so and think about it this way.
#
If there was an enemy at the border during wars, I remember my grandfather, like, you
#
know, my mother remembers growing up in Agarthala in 1971, like, you know, during that 71 war
#
and yeah, there would be these sirens, there would be blackouts, like, you know, when there's
#
kind of enemy air sorties coming in, everybody, every single person would turn their lights
#
Because you knew that one light on was kind of enough to give the whole game away, right?
#
I mean, and so, you know, but if you look at the statistics, pandemics kill, like, we've
#
lost more people in the 1918 to 1920 kind of, you know, Spanish flu in India than 10 times
#
more than World War One and World War Two put together, right?
#
So why are we not treating it with that level of urgency where the only contribution we're
#
asking every citizen to do is wear that mask and wear it at every single point when you're
#
outside, avoid the large gatherings and the government can kind of help and accelerate
#
this by procuring and door stop, door, door distributing four masks to every family and
#
just make it happen, right?
#
I mean, so that's the one thing which I feel very confident about saying that we just simply
#
I won't get into, you know, politics of vaccine pricing and procurement and all of that.
#
I think Ajay has talked about that, but, you know, I would just say regardless of what
#
formula we use, government of India, state governments, like, you know, just spend the
#
money it takes to get the vaccines done and kind of don't try to nickel and dime and get
#
poor patients to pay like 500,000 rupees, like, you know, it's kind of completely, completely
#
So you mentioned the term there, which kind of triggered a memory of a recent news headline,
#
which is you said panchayat and I thought of how in UP, these 700 teachers died because
#
they were sent on panchayat duty because the UP government refused to, you know, postpone
#
the elections, they send these teachers on government duty, they died.
#
In fact, a recent report from May 6 says it's more than a thousand by now, the death toll.
#
And there was a story of a woman who, you know, her husband couldn't go because he was
#
literally on his deathbed, he ended up dying, I think.
#
And she was asked by the government, why is your husband not coming for duty?
#
And she had to send a photograph of him on his hospital bed as proof, something of that
#
And that, and to me, that's murder, by the way, plain and simple, murder by the state.
#
And that kind of leads me to the thought that, of course, I agree with you, we should do
#
But what we're really talking about here is two things, information and messaging, that
#
this is a correct information we should get out there, but this is messaging which should
#
be done in the public interest.
#
What we instead see the kind of messaging that we see is actually spreading the wrong
#
kind of information, like coronal is good for you.
#
You know, that Ayurvedic thing, and that will cure you.
#
Or there will be a God man who will say that there's no problem with oxygen, you just need
#
to learn how to breathe, right, and you need to do these yoga poses and all that.
#
And for once, I'm moving away, by the way, you will note from criticizing the state to
#
criticizing society itself, because our society, honestly, we elites in the first class of
#
this plane and all that we might think that we read books, we read papers, in your case,
#
you write papers, and we are all okay.
#
But we are a very backward country.
#
Look at the things we believe.
#
It seems to me to be an insurmountable challenge at two levels.
#
And I don't mean to be negative and say that we shouldn't try at all.
#
And so in fact, I had not intended to talk about, say, Indian systems of health care
#
or like an alternative systems of health care.
#
But let me say, I think one important point here, right, which is, see, I think, if you
#
read Sanjay Bahru's recent book, see, a big part of the current dispensation in government
#
is something that reflects this deep sense of angst that the greatness of Indian history,
#
right, mean and traditions and cultures like, you know, are being undervalued by kind of,
#
And I think some of this is language, right, mean that the English speaking class has been,
#
you know, we have been incredibly remiss, I think, in kind of not getting scientific
#
content, right, mean out in local languages in a way that it's kind of much more accessible.
#
So, you know, there's, and in fact, when I started writing my column in HD, one of the
#
reasons I picked HD was I spent a lot of time trying to translate and write in Hindi, like
#
But it turned out too technical.
#
And like, I think they told me, hey, boss, you'll have to simplify this.
#
But that's a separate story.
#
But at least the intent was there.
#
But I think here's the point I want to make is that, see, modern clinical trials, okay.
#
So I do RCTs for the living.
#
I believe in RCTs and it came from medicine.
#
But that being said, there is one big blind spot, right, which is conditional on doing
#
We randomize the treatment and saying, did this treatment have an effect?
#
But the universe of treatments that get evaluated by RCTs is highly selective, right, you know,
#
so what determines what gets evaluated, correct?
#
So now, to me, the right way of kind of bridging, I would say the tradition and the strength
#
we have in our traditional system.
#
See, here is my view of our traditional systems is that a lot of it may be quackery, a lot
#
of it may be placebo, but just by natural selection, okay, the fact that something has
#
survived for 3,000, 4,000, 5,000 years means that there is likely to be some survivorship
#
That if people say take ginger or take turmeric or take honey, right, like, you know, that
#
is not kind of pharmaceutically done, but that has a bunch.
#
So my throat, after talking for us, what am I drinking?
#
I'm just drinking honey water, like, you know, with herbal tea, right?
#
But where I think we need to really be thoughtful about bridging kind of the greatness of our
#
traditions with the modern scientific standards is to say, instead of claiming that we had
#
all of these great things in plastic surgery in the back days, I can say, here is a body
#
of traditional knowledge that we can now invest in testing, like, you know, with more modern
#
scientific methods and then validate rather than waiting for Western pharmaceutical companies
#
to come and patent like name and like, you know, I mean, it's a good, right?
#
Like, you know, all these kind of things that we've been doing traditionally, you know,
#
so in that way, we've known it works.
#
So why leave it to the Westerners to come and bring modern science to our traditions
#
as opposed to kind of taking the traditions and meeting the science and kind of getting
#
the best of both worlds.
#
And the Chinese have been very good at this.
#
I forget the name of the scientists, but somebody who won a Nobel Prize, like, you know, I mean,
#
for medicine actually did exactly this, right, may take a bunch of traditional medicine,
#
And some of these will not be true.
#
The whole point of kind of the modern RCTs, some of these will, in fact, have been placebos
#
of folk wisdom, but some of them will turn out to be very effective.
#
I share your despair at kind of pop science, like, you know, and the marketing around it.
#
But I think it's important to recognize the source of cultural angst, right, I mean, of
#
where it comes from and then find a constructive way to channel that angst in a way that reflects
#
a modern scientific temperament.
#
Yeah, I mean, look, I have nothing but respect for the conservative view that if something
#
has worked over the centuries, I mean, there's a reason it's a survivorship bias, you know,
#
putting haldi in doodh when you're ill and all that, it's fine.
#
I have an issue with things like this coronal stuff or, you know, a yogic pose will solve
#
your oxygen problem, which to me is just offensive to the people who actually suffer from it.
#
I have a problem with those, but that's a separate issue.
#
I mean, two of the major problems that I have with, you know, our modern times have been
#
expressed by us in this episode, which is the incentives problem and the narratives
#
And neither of those is going to go away anytime soon.
#
So but before I let you go, let's turn away from healthcare for a moment.
#
What have you been reading recently outside of work?
#
Let's say what are the last three books which filled you with joy or which you thoroughly
#
enjoyed, which you can recommend to me and my listeners, except that, you know, kindly
#
don't put some random academic thing and say, no, no, I got joy from this and all that.
#
So here's the bad news and the good news.
#
So the bad news is I have not actually done much reading in the past year, year and a
#
half, because I've mainly been reading things around my book.
#
The good news is that my main source of new knowledge has been listening to podcasts,
#
including yours, right?
#
Like, you know, I mean, and so one of my routines here in this pandemic and again, San Diego,
#
it's interesting you mentioned Eric Topol, right?
#
Because he's here in San Diego and San Diego is just like a magical place to be in in some
#
Because you can just go out in long walks.
#
And so, yeah, my source of absorbing new knowledge in the pandemic has actually been primarily
#
So, yeah, so somebody else would ask me, I would say, go listen to a bunch of seen and
#
And I'm not saying this to flatter you because I've not done it in 180 at the time I did
#
the first one, but I have done as you can see from the connections I'm making.
#
What are your favorite episodes?
#
What are your favorite episodes?
#
I mean, obviously the one with Pratap was a classic, like, you know, I mean, that happened
#
like, you know, immediately, immediately after mine.
#
You know, I enjoyed the one with Vinay Sitapati a lot, though, I mean, I did think it kind
#
of, I would say, whitewashed a little bit, like, I mean, of kind of the larger narrative,
#
like, you know, of the rise of the RSS.
#
But I think overall, I really enjoyed hearing the bit about the combination of legal, journalistic
#
and scholarly training in that book, right?
#
And so I can see the lack of the journalistic training in my own writing, like, you know,
#
because I'll do the fact checking in the same rigor as an academic, but I haven't learned
#
the craft of storytelling in my writing nearly as much, right?
#
So that piece is a lot more work for me because, you know, I get the content, right?
#
But then I have to simplify, which is why I'm still saying I might do four podcasts
#
and say, hey, leave the book.
#
You're most welcome to, no, no, please write the book, but you're also welcome as a supplementary
#
effort, a supplementary teacher training, you can say.
#
But I enjoyed that in terms of the process of writing, I've enjoyed like your recent
#
scientific ones, like, you know, with, with, again, I keep calling him Bhalo Manush, but
#
You know, I think that was, I think that was fun.
#
I think, you know, just going back in, or Gautam, like, you know, hearing people talk
#
about the joy of science.
#
And I mean, I'm a nerd, I'm an academic at heart.
#
So as a way of kind of learning, you know, people who have been at the cutting edge of
#
other disciplines talking about their work, I think that's the part I've really enjoyed.
#
Well, I think many, many, many listeners of The Scene and The Unseen would actually say
#
that you are one of their favorite guests.
#
So once again, thank you so much for coming on the show.
#
And I hope our conversation keeps continuing over the months.
#
And good luck with your book.
#
And yeah, take good care of yourself as Amitabh Bachchan says on KBC, take good care of yourself.
#
Yes, I mean, my song is Shabba Kher, like, you know, Din Nisar Abizara Tori Angana, which
#
at the end of KBC, he used to say that he used to say a line which had Shabba Kher at
#
It was quite a flourish.
#
You know, actually, on our past episode, I think somebody had kind of commented there
#
saying, okay, this was a Bollywood-length effort, like, you know, why not put in a few
#
Would you like to sing?
#
So what I'm going to sing is I'm going to sing like a line for you, which reflects,
#
I think the writer's angst, but it's also, you know, one of my favorite.
#
So it's this line from Inkastidi and Mohafiz, right, which is, aaj ek harf ko phir dhoondhta
#
phirta hai khayal, right, which is the poet writing that I have this pot that is running
#
around, like, I mean, looking for the right word to land on, right?
#
Means, so when I kind of talk in high school, sometimes about Indian education systems
#
that obsess so much in science and math, I kind of keep emphasizing language because
#
the precision of your thinking is limited by the precision of your language, right?
#
I mean, and so that's kind of the dominant emotion as I write, right, because you spend
#
kind of sometimes hours grappling for kind of the right phrase, the right sentence to
#
But yeah, so I ticked that musical box, pisle baar Tamil mein baat kiyaan, iss baar gaana
#
Mohafiz, of course, was one of the rare films which Ismail Merchant directed himself instead
#
of his partner, James Ivory doing that.
#
Anything else you'd like to sing?
#
Of course, you have a good voice here.
#
So among the unknown things about me is back in high school, right?
#
I used to be like, you know, an inter-school Antakshari champion.
#
So I'm going to give you a letter and you've got to sing something with that, okay?
#
Ho jaayega, but show pe nahi yaar.
#
Nahi, abhi ek, ek, ek, ek do line gaado, I'll give you a letter, theek hai kya letter
#
Tumhare naam ka letter detao, ka.
#
Like, no, no, no, don't do, no, no, don't do this to me.
#
Come on, come on, this is, listeners will love it, ka, chalo.
#
Ka aur maa, I'll give you an option because both are your name only.
#
Maa say, you know, so many songs will start with mai.
#
No, no, no, there's no shortage of songs, like, you know, so in fact, see, here's the
#
tragedy of my Antakshari past also, my teammates, there were three of us, okay?
#
So the other two had much better voices, okay?
#
So I was the quantity guy, they were the quality people, okay?
#
So, so my, so what I would often do is essentially when the other team would start singing, I
#
would already know what is going to end in, so I will say, okay, ye gana gana hai and
#
pass it to my teammate who had the better voice for that particular song.
#
So, so I was kind of the, the, the intellectual kind of leader of the team.
#
I'm not letting you get away, bro.
#
I'm sorry, this episode is not airing, ka aur maa, I've given you an option, bro.
#
No, then, you know, if I want to get very melancholic on the ka, I would say like,
#
you know, kabhi khud pe, kabhi haalat pe, rona aaya, kabhi khud pe, and I say that
#
because, you know, the haalat have truly been like, you know, rona inducing, so I think
#
part of the problem was, yeah, most of my initial songs with ka and maa were happy songs
#
and I couldn't automatically bring myself to sing those, and I think the haalat have
#
been really hard because, you know, I had wanted to travel back and I'm not being able
#
to travel back and you kind of pray every day for safety of friends and family.
#
So yeah, maybe, maybe all survive and get through to the other side and then hopefully
#
build a better Indian state.
#
Well, the haalat may make us cry, but you gave us hope.
#
So thanks so much, Kartik.
#
If you enjoyed listening to this episode, do visit the show notes and dive into rabbit
#
You can follow Kartik on Twitter at kartik underscore econ.
#
You can follow me at Amit Varma, A-M-I-T-V-A-R-M-A.
#
You can browse past episodes of The Scene and the Unseen at sceneunseen.in.
#
Thank you for listening.
#
Did you enjoy this episode of The Scene and the Unseen?
#
If so, would you like to support the production of the show?
#
You can go over to sceneunseen.in slash support and contribute any amount you like to keep
#
this podcast alive and kicking.
