Economic Perspectives on Air Pollution: Interdisciplinary Global Environmental Health Collaborations


KENNETH LEE: So please join
me in welcoming Dr. Jina. He will be speaking on the
global mortality consequences of climate change, accounting
for adaptation costs and values. AMIR JINA: OK. So this is a little bit of
a change to the program. But I thought in keeping
with the health discussion, I would talk about the health. So it’s a change in two ways. It’s a change from the
program in that I’m going to talk about health
consequences instead of labor consequences. And also it’s a change
from the program broadly because I’m going to be
talking about climate change, rather than air
pollution specifically. Now of course– this
came up yesterday. CO2 is still a pollutant. And so we can think of this
broadly about air pollution. But it’s not the
direct effect of CO2 that is the thing which is
going to be harming our health, but rather the indirect
effects of CO2. And I think it still fits
in with the broad view of the conference in thinking
of bringing in health work into environmental-related
research. Of course, in the
panel later, we’re also– this is part of a
larger project where we’re trying to look at the
impact of climate change on numerous outcomes. And one of those is labor. So I’m happy to talk
about labor effects. There’s also quite a lot
of research, some of which I contributed in a small way
to, about the labor or even educational effects or cognitive
effects of air pollution from an economic
point of view, that I can discuss during the Q&A,
that I’d be happy to talk about. But now I’m going to talk
a little bit about what we call the Climate Impact Lab. So as background, this
started roughly five years ago when a group of donors
in the US decided that they wanted to
try and understand what the risks to
the United States were due to climate change. This was something which
climate change impacts. It had received quite a
lot of study but not often from an economic point of view. And so we stepped in to
do that focused on the US. And we’ve expanded it globally
over the past five years or so. There’s now a group of 30 of us. And so Michael Greenstone,
who’s the director of EPIC in the BFI, is
one of the collaborators on this project. And we’ve been
slowly expanding what we’re doing over the course
of the last five years. What we’re aiming to do is
to try and match research from economics with
research in climate and other environmental
research to try and understand what the economic
costs of climate change are on many different outcomes. But always found this
on a strong foundation– as strong as we can– of data
and rigorous economic research, which to date hasn’t really
been the way in which this question of the
impacts of climate change has been approached. So just as some background,
and there’s some jargon that I might use, this
is what the projections under climate change look like. The blue lines here are– so I look like I’ll do
karaoke for the day. So the co-benefits of
this could be enormous. In fact the benefits could be
much higher in a monetary sense than what I’m
going to show here, which is just these indirect
effects of CO2 on health. And part of the reason
that we do this– there’s a set of
models where people try to understand what the
costs of climate change are. They have some empirical
evidence supporting what the impacts might be. This is just a histogram
showing the number of studies that are used to
calibrate those models which are used in policy all
over the world to calculate the costs of climate change. So each part of the
bars here correspond to the number of
academic papers that are used to inform what’s
going on in those models. We did the lit review
a couple of years ago and found that, in the years
after those models stopped updating themselves,
there have been hundreds– if not thousands–
of papers giving us more evidence on what the
impacts of climate change are. And so we’re trying
to make sure we update this in a certain way. These are not just
drawn from economics. They’re drawn from epidemiology. They’re drawn from ecology, from
all kinds of other disciplines. And so this is quite a big
interdisciplinary field. For the discussion that
I’m going to show now, I’m focusing on health. I mentioned we had this
research on the United States. This is a paper we
published last year. And this breaks down what
we call a damage function. This is damage to the
US economy and percent of US GDP as a function of
global surface temperature. So as temperatures warm,
there’s increasing damage– the top of this line. We break that down into
different outcomes. And this big blue wedge is
the costs due to mortality. So the health costs that we
analyzed for this make up the bulk of the costs
of climate change that we observed in the US. So as we started
expanding this globally, we thought the best place– the first to start– was to
look at the mortality effects again, thinking that they
would be the largest costs. I guess I’m restless and
I have to walk around during my presentations. This starts with
collecting a lot of data. So a similar problem to
what I mentioned yesterday. A lot of the research
that had been done before was
trying to say, here is the effect in
the United States, or in a couple of
countries in Europe. And then extrapolating
that all over the world, assuming that the effects that
we would see in the United States are the same as effects
we would see everywhere else. So one of the first
things we have to do is to break that
dominance of the US or certain parts
of northern Europe in being the people that
we represent when we talk about these health impacts. So we went around
and we collected as much high-resolution
health data as we could. Usually these are
monthly-frequency mortality data at very high
spatial scales. All these little lines in here
are the different spatial units that we calculate. And we actually have
mortality data here for 56% of the world population
covering a few decades. So it’s one of the larger
high-resolution mortality datasets which exists. The reason we got up
to 50% is because we have both China and
India in here, which accounts for a lot of that. And then we run
what is a little bit of a standard
statistical model here, which is for each
location– again, similar to yesterday–
for each location using those as a
control for itself. So in a hot year versus a cold
year, what is the mortality change, knowing that those
fluctuations in temperature are random? We’re able to identify
this causal effect. And we develop
something like this. So this is our dose
response function here for temperature and mortality. Meaning that on days with
cold daily temperature, mortality rates are elevated. So our days with colder
temperature, mortality rates are elevated, compared to
a day at this average here of 20 degrees Celsius. And on days of hot
temperatures, mortality again is elevated– this
characteristic U-shaped response. So cold days lead to deaths,
hot days lead to deaths. This is not the same
for every person. This actually differs a
lot as a function of ages. So for those aged under five,
there’s almost no effect. A very small effect possibly
for aged between five and 64. And then for ages 65
plus, a very large effect. So a lot of the mortality
burden of temperatures happening among the over-65s. So again, this
U-shaped response, but much greater for
the older age groups. OK. One of the other things
we want to understand– so then we move away
from public health, and we put on a bit
of an economics hat, and say, so we have this
potentially physiological response to why
temperatures are harming us. There’s direct. Some of the reasons that it
could be leading to mortality are direct causes
from heat stroke, but also other exacerbation
of other conditions. And one of the
reasons why you see a lot of this in
the age 65s and over is that they are
people with weakened cardiovascular systems. And so extra stress on that
as they try to cool down can be exacerbating potential
conditions they have. We control for what we would
call mortality displacement. So a very old person who
was already very sick, and might have died
within a couple of days, exposed to a heat wave, and
so they die two days earlier. We control for that
so we’re making sure we’re in this
type of result. We’re only looking at
those deaths that would not have happened otherwise. So this is not
just moving people around a few days in time. But we want to put
on our economics hat and think, in the US we
know that this response is very different than
it would be in India. In fact, Michael Greenstone has
a paper showing exactly that. So people are clearly adapting. There’s an income
component of that. As people get
wealthier, they all get air conditioning so they’re
able to protect themselves. But also, as they get more
exposed to different higher temperatures, they are
required to invest more in protecting themselves. So what does
adaptation look like? We have this U-shaped response. Mortality is elevated on
cold days and on hot days. What would this look like in the
data if people were adapting? It would look a bit like this. This curve would get
flatter and flatter until there’s no relationship
between temperature and mortality. Even the United States– one of the richest
countries in the world– we don’t see this flat line. Meaning that in
order to adapt fully, it must be extremely costly. This might mean we live
inside an enclosed, completely air-conditioned bubble,
and that’s the only way to escape these effects. That’s too costly
for people to do. And so we make some trade-off of
accepting some level of damage, given the price we have to pay. So do we see that
effect in the data if we divide up the world
into deciles of temperature. So the coolest 10% and
10% hotter and hotter, up to the hottest 10% of
regions within the world. The world here’s divided up into
many different small regions. And look at just what’s
happening on the hot end– actually at 35 degrees Celsius– so the level of this
curve at 35 degrees. For the people in colder places,
we see a very large response. If they experience
a hot day, they’re going to have a very high
incidence of mortality due to that. People are not very much
adapted in these colder places. As we go to warmer
parts of the world, that response gets
lower and lower. The reason that it
doesn’t go down to zero is because, as we get more
into the hotter tropics, the effective income starts
to become more important. So people in this band
tend to be a lot richer than people in this band. So this elevated
mortality again. But you do see in the data
that people are adapting. You can ignore
everything on this slide. It’s just to remind
me to say how we’re going to deal with this
issue that people are adapting. Effectively what we do is, we
look at that U-shaped response to temperature,
but we work out how it varies, based on
how rich people are, and how frequently they
experience hot days, like heat waves and other things. And you would imagine
that people who are richer have a lower U shape. And people that are in hotter
places have a lower U shape. We look at how this
varies in the data– that’s what this is explaining. And that gives us
an amazing ability. We had that map where there
were still lots of areas where we had no data. But we can measure average
temperatures, or the number of hot days, or the
number of heat waves, as well as incomes
in those places. Knowing those two things,
we can go from that map with lots of gaps in it to– I’ll skip just
that for a moment– to a map where we
have filled out what everybody’s response
to temperatures are. So here, this is
just plotting what the effect of
really hot days are. Well, we can do this for every
one of those temperatures. So we have that
U-shaped response for each of these different
regions in the world. So the world divided up into
25,000 different regions, 25,000 different
Us, even in places where we don’t have data. I can talk about how we test
that out of sample a bit if you want. So instead of thinking
of climate change as that average change– that red line that I
showed at the start– we instead think of
it a little bit more as what’s going to happen
not just to the average, but to the whole
distribution of temperatures. Because what that
U shape implies is that the really hot
temperatures hurt us but the average
moving around doesn’t make too much of a difference. So this is what the distribution
of daily average temperatures looks like. So there’s a lot of people
experiencing averages around 28, 29 degrees
Celsius in the world. This is population weighted
for the whole world. This is what climate
change is going to do under that
business-as-usual scenario. And this is the change. So what we see here is that
we experience a lot fewer days here in the 25 to 30 degree
range, and a lot of them move to this increase– to these higher temperatures,
which we’re seeing from those U shapes are particularly
bad for our health. So is what the world
looks like in 2020. Redder here is worse. This means you’re
more sensitive to, in this case,
hotter temperatures. So these lighter places
are less sensitive– some richer places and
some warm places in this. The lighter places
are less sensitive, the redder places are more
sensitive to temperature. As we project this
into the future, knowing that we have some
projections of how incomes are going to grow and temperatures
are going to change, we can observe people adapting. And so this is in
2050, 2080, and 2100. The sensitivity by and large
gets less across the world. So people are
definitely adapting to higher temperatures. OK. Again, you could ignore most of
the things that it says here. That shape– the U
shape coming down, this adaptation that
we’ve talked about– we can observe that in the data. But remember I said that that
U shape doesn’t completely go away, even in rich countries. That would imply that
adaptation is not free. So what we model in the data
is the benefits of adaptation but not the costs of adaptation. So we’re only seeing
one side of the picture. In order to be able to
lower that U shape down, we as a society need
to pay some costs, whether that’s me personally
buying my air conditioning, whether it’s my government
investing in better health care infrastructure, it’s
some kind of change, which is costing me money. So if we only look
at the benefits, we’re overstating
how well we’re going to do as the world heats up. So how do we back out
what those costs are? So there’s some benefits. And we measure
how the adaptation happens but it’s not free. So we use an appeal
to economics here, this idea of
revealed preference. And essentially
what we observe is that we can tell by the level
that people have adapted to, we can infer what the
costs of adaptation are. Because if the damages
to health were higher than those adaptation costs, you
would pay the adaptation cost to avoid those damages. If the damages to
health were lower than those adaptation costs, you
don’t pay the adaptation costs. And we can see people reveal
the level that they’ve adapted. So they’re actually
revealing to us, through this economics argument,
what the costs of adaptation are. So we do that for
everywhere in the world. And so we back out not only
what the benefits of adaptation are, but also what
the costs are. There was a question yesterday
when talking about the AQLI that maybe the
housing in China might be different in other places. This would be one
way in which you can look at this variation
in the effect of adaptation to air pollution, in
the case yesterday, by trying to build up this
knowledge around the world. It’s extremely data
intensive so it’s definitely not an easy thing to do. But you could start to see
the effect of other factors on mitigating a bad effect
versus a milder effect. OK. We project this out
into the future. This is time on the x-axis. This is change to death rates. If people did not adapt, we
would see an enormous change to death rates. The death rate in the US– and in fact, in India, as
well– is about 1,000 people per 100,000. So this is about a 15%
change to death rates if people aren’t
allowed to adapt. If people get richer,
this drops a lot. I’m an economist in front
of all of you telling you that money is good. So it seems that that’s
borne out by the data. If we then look at the benefits
of climate adaptation– so this is just, I get richer
and I buy an air conditioner, or I buy a better
house, or I live in a place with
better infrastructure and I’m less exposed. This is, I wake
up in the morning and I realize it’s
gotten hotter. And so I need to make some
change to my environment. Whether it’s then I make
the decision to buy an air conditioner or
something else, it’s something I would not have
done just because I got richer. It’s something that
I’m doing because I’m facing a different environment. But that benefit gets us down
to very low mortality rates. So people are adapting
quite a lot here. It’s still positive,
and it’s still costly. And in fact, a lot
of these deaths are happening in poorer
parts of the world. So there’s a big
inequality effect going on here that
is masked by just looking at this global number. And if we pull back in
those adaptation costs, this goes back up much higher. So the amount that we have to
pay to get to this low amount is– this is expressed
in terms of deaths. But it’s quite a large cost. OK. Of course, I’m just
presenting a mean there. What we do here is we run this
massive computer simulation to project these
into the future. So we have a lot of
uncertainty around this. But essentially what
we’d be able to say is, here’s the risk associated
with this large change in temperatures that
we’d experience. And there’s a lot of uncertainty
around this, largely driven by uncertainty in
what we think is going to happen in the future. So in connecting this
back to air pollution, the uncertainty in temperature
modeling is very large. The uncertainty in air pollution
modeling into the future is even larger,
because as CO2 gets released around the world,
that mixes everywhere around the planet. But it’s coming from
individual factories and individual sources which
are producing air pollution. The spatial configuration of
those air pollution sources is something that is
very hard to predict. So it’s much easier to
predict what the population exposure to higher
temperatures is than it will be to predict what
the population exposure to air pollution is in the future. It matters where those
factories get built, where those power plants get built. So if you’re thinking about
this in an air pollution– the co-benefits of climate
change mitigation on air pollution– trying to do this exercise
becomes very, very difficult in terms of knowing where we’re
going to build in the future. This is what this
looks like in a map. Red is worse here. This is damage. These are monetized and some
value associated with them. And you see that some of the
exposures in the northern parts or the very southern parts of
the world are quite modest, or even they look beneficial. But this vast band of hotter
and poorer countries– this is by 2090 under
business as usual– experience large damages
due to climate change. So what we’re masking in
looking at those averages is that there’s a big
effect here on global health inequality effectively. OK. That image that I showed
at the start, which showed that mortality was a
big part of the cost in the US, we can draw a very
similar kind of image– the relationship between GDP
damages here, and temperature changes on the x-axis. So as the world heats
up, what does this look like in an economic cost? So this is for that
medium mitigation. This is for this higher,
this business-as-usual track. And effectively what this
says is, as the world heats up to somewhere around four
or five degrees, which is what we might be expecting
under this business as usual, that will be somewhere
close to a 15% share of GDP lost per year just on
those mortality costs. Remember, I already said to you
that in particular, mortality, but broad health costs of
air pollution are probably far greater than this. So you can imagine this is an
enormously high cost to pay. We fit a line to
that and we look at the share of direct costs
versus adaptation costs. And it seems that a
lot of that damage, a lot of the cost to
society, is coming from what we will pay to
adapt to those temperatures. So the adaptation is
really accounting for a lot of the health care costs. So that’s essentially
what we do. We’re doing this for a lot of
other things beyond health. So energy use, agriculture,
labor productivity. And we’re eventually
also going to try and build this a little bit
more out into air pollution. We get this dataset that
covers over half the world’s population. So all the deaths recorded for
half the world’s population. We develop this
effort to try and find what the costs of adaptation
are, which I think is a new thing to do. The costs would be
much larger if we ignore that people are adapting
and that people can adapt. But the adaptation costs
account for a lot of the damages that we experience. The final thing here, which
I think connects to one of the later talks,
we can estimate what the social cost
of this is, because we know how many tons
of CO2 are going to lead to a certain
temperature increase. We can work out
the cost to society of one ton being emitted
due to what we found here. And we find in the range of
$20 to $40 per ton emitted. That’s the net
present value costs. I can explain that later. For comparison, some
of those other models, which we currently
base policy on– so one of the main and
most sophisticated ones– they give a cost for
mortality of less than $1.50. So using data here for
countries that aren’t just, as that one does– I think 12 cities in Europe,
which covers about 1% of the world’s population. But using a much higher
cost of mortality. So I think the
take-home points, we can identify what these
adaptation costs are. And thinking about this in
terms of the co-benefits you get from air pollution. It’s not just this cost
but all the extra benefits, in terms of reducing the
exposure that people will have. And thank you very much. KENNETH LEE: Our next speaker
is Doctor Sola Olopade, who is the Professor of
Medicine and the Director of International Programs at
the Pritzker School of Medicine, and Clinical Director of
the Center for Global Health at the University of Chicago. SOLA OLOPADE: I
mean, this is what makes this kind of
setting exciting to me. Because I’m a physician,
but we’re interested, through interdisciplinary
approach to some of the challenges
around air pollution, to look for solutions. And as physicians
and academics, we’re doing a lot of randomized
controlled interventions. And if we do those
interventions, come up with solutions, if we
don’t have colleagues who can help us
implement and scale up, then the results and outcomes
of our randomized controlled trials just stay there. So this is why I’m
actually excited, why I look forward to hanging
out with some of our economics and some of our
policy people, so that some of the
outcome of our research can go beyond the clinic. So for those who were
not here yesterday, I presented the result of
a randomized controlled intervention that we did in
Nigeria that transitioned pregnant women who are
cooking with kerosene and firewood to ethanol. And I’m just going to put
two slides from yesterday’s presentation especially
for the benefit of those who weren’t here. During this study, the
women were actually given free stoves,
which cost about $65. That’s a lot of money. And free ethanol
during this study. And because of the benefits
of the intervention from the ethics perspective
and being responsible, we actually gave the
same stove to the women who were in the kerosene
and firewood control group. But the observation
on this slide was actually very reassuring
to us because about 81% of the participants,
after the study, when we stopped giving them ethanol,
were using their own money to buy ethanol,
because they perceived it was beneficial to their
health, regardless of the cost. And about 50% of them use
the ethanol exclusively, as we disclosed yesterday. Sometimes in cooking
specific foods, people need to have different
stoves that they use. So as much as we want
to be purist in terms of what we expect
people to do, sometimes they do what we
call stove stacking. Where they know if you want
to make tea or a quick meal, you use this one. If you want to do a big
party, you use on there. So it’s very difficult
to actually get people to adopt this stove. And what was very intriguing
for us after this study was, we were delivering
about 10 liters of ethanol to these
women every two weeks. And the end of the
study, the expectation was that they will
show up and they will buy 10 liters of ethanol. But people didn’t have money
to buy 10 liters of ethanol. So initially, even though they
wanted to adopt the stove, they were going
back to kerosene, because they could buy
1 liter of kerosene and use it to cook,
instead of 10 liters. So when that came to
our attention, then we went in the direction
of ensuring they could get 1 liter of ethanol. So when you normalize the
cost between kerosene, they started buying
ethanol because, based on their observation
during the study, they preferred to use ethanol. So these are some
of the elements that those of us
who are physicians don’t always think about when
we’re doing our clinical work. But when you are there, doing
social and behavioral work to ensure that people actually
adopt what you want them to do, you really need to think
about behavior change. So learning from our
social scientists, that’s what’s been
most intriguing to me. And at the end of
this studies, Shell– as I mentioned yesterday–
they’re actually going to be doing a 2,500-stove pilot study. This as a commercial
pilot, unrelated to what we did in the RCT, which is
really quite fascinating. But before Shell
decided to do this, they sought permission
to actually talk to some of our women
who are participating in the randomized
controlled study, if they could go talk to them. And so these reasons
why the women liked and adopted the ethanol
stove was not part of the RCT. This was what Shell was able
to elicit from the women and which influenced their
decision to actually do the commercial pilot. The pots and pans,
according to the women, who had no more black. And most people know,
especially women, when you have clean
cooking utensils, it makes you feel really good. They didn’t have any burns. There were no burns
associated with cooking. I know even when
people use LPG, which is supposed to be one
of the gold standards, people are still
worried that it’s going to blow up or explode. And we’ve seen those
all over the world. And it was easy to use– all of the reasons that are
in there on this slide, which was very important to factor in
in any effort to actually get people to scale up the
distribution and availability of such stoves while cooking. So in transitioning
from a tier 4 stove– how many people know
about the tier and stoves? OK. Maybe I shouldn’t assume. OK. For people who don’t know, if
you’re talking about a tier 4 stoves, you’re
talking about fuel that bun efficiently,
that don’t emit too much, and that are efficient. And the fuels that
fall into that category are if you’re using electricity,
if you’re using LPG, and if you’re using ethanol. If you look at
the energy ladder, the ones that are in tier 2
will be your improved biomass, the acolytes, and all of
those non-tier-4 stoves. So in looking at transition
into the best stoves out there with minimal
production of pollutants, there’s significant
direct health benefits. Our studies show that,
as well as other people who have demonstrated this. There’s a drastic reduction
in emission of pollutants that contribute to climate change. And I’m hoping that
the panel, I get to ask in there a few questions, too. Because the CO2 part
of industrialization is very, very important. But as people cook with
firewood and biomass, it’s important to
also understand that one of the pollutants
from the incomplete combustion is black carbon. And black carbon is
pretty much like suet. It stays up in the air,
absorbs solar energy, and is deposited on
the mountain top. So it’s a major contributor to
some of the immediate climate forcing that we see,
and some of the melting of some of the glaciers,
and stuff like that. But the good thing
about black carbon is the half life of black
carbon is about 3 weeks, whereas CO2 that’s emitted
from industrialization, the half life is 200 years. So if we really want to think
about practical solutions that may benefit all of us,
why not do something about black carbon, that has
a half life of just 3 weeks? Or even particulate
matter that goes up there, that if you have a
good torrential rain, it takes it away. So the bottom line is
trying to figure out the activities and
attributes of poor people who cut down trees for
their cooking needs. If we dare focus on their
needs and their benefits, we may actually all
benefit from lowering the temperature a little
bit and slowing down the environmental
degradation that we see. There would also be a
reduction in deforestation. Because if you look at
ethanol, it’s a non-fossil fuel and does not require
cutting down any trees. The same thing goes with people
who actually use bio-gas. It doesn’t involve
cutting down any trees, yet it’s very, very energy– at least environmental friendly. And I don’t know how
to put a cost on this. Young people after
school are usually the ones that are involved
with scavenging for firewood, especially some of the girls. Look at the time wasted away
from doing their homework. And look at the impact
of some of the dangers from being raped
while looking for it. So I don’t know how to
put an economic cost to that benefit from that. So what we thought we would do,
as part of the commercial pilot study in Lagos that was
being done by Shell Nigeria, is to build a
partnership around it. And SNEPCo– all
I said is Shell. Unikem is the largest producer
of ethanol in Nigeria. And Forte Oil is a gas
company, where they have a lot of gas stations. And building a team around that
is actually quite important. Because if you want people
to have ready access to the canisters where
they can get the ethanol, it makes for easy access
to the cooking fuel. And the plan was
that the gas stations will sell some of these stoves
that would be discounted by Shell as a commercial pilot. The interest around
consumer uptake, the financial
feasibility of this, and to actually be in a position
to develop a supply chain. So that people like the
stove, they look for it, they can find it readily. And that it’s safe. And to also see how
that could actually inform some of the policies
around cooking, not only in Nigeria, but it’s an
effort that’s generalizable. I know that some
of our colleagues, since we published
our RCT in Bangladesh, they’ve been harassing
me in terms of, how do they get the stoves? How do they start
producing ethanol? So it’s an issue that’s
generalizable to anywhere you have poor people, whether
it’s in India or in Africa. So the idea of this– before selling 2,500
stoves, we’re academics– to do a small pilot
to see whether, if you introduce the same stove
and fuel in a different market, in a different setting outside a
randomized control study, where we were giving these
stoves to people for free, and the ethanol for free. Of course, they would love it. To go to a totally
new place and see whether people who have
never seen this stove would actually like it. You just don’t go to
the market and start selling all of these stoves. So the idea was to assess
the consumer preferences, to look at usage practices,
and taking advantage of what we know how
to do with respect to the objective monitoring. Put stove use monitors
on all of these stoves, so that it’s not people
just telling you, we like this stove, even
though they’re not using it. And as I show you
some of the results, you understand why it’s
important to do that, and not just rely on
people telling you that they love the stove. So these involved doing
household interviews, especially with a person who
was making economic decisions in their household. To look at market
surveys and also look for a way to do promotions
and measure stove performance in the field to see how much
reduction in CO2 emission could actually come from
introducing this study. And this was actually funded by
the African Development Bank. So the idea was to have an
experimental home of about 30 that are close to
these gas stations. And these two components
are not completed. So I’m just going to give
you the initial results from the first study. This is just to show you
the location geographically in Nigeria where the homes were
located very close to the gas station. So that if you needed a canister
exchange, you are close to it. And this is what the
stoves look like here. This is a two burner on the
left, and a single burner. The important technological
advantage about this stove is that, this canister
that you see here, you can put over a liter of
ethanol into this canister. You turn it over. It doesn’t spill. OK? So that’s why, when you use
it, there is really no fire. Within three years of
conducting the RCT, we didn’t really see
any single burn or fire. So the technology in that
is actually quite unique. And when you look at the
average use of the stoves in this small cohort
of 30, you understand why we have stove use monitors. You can see that kerosene
was still the most used fuel in their house. This is the CleanCook
stove, and this is LPG. OK? And you can see that kerosene–
which is a little cheaper than LPG or CleanCook, the ethanol– was the most used. Even though when you ask people,
they were telling us, oh, yeah. We like the ethanol. We’re using it. And if you look at the average
number of minutes used, you can see again that
it’s the kerosene, more than even the LPG
or the CleanCook stoves. And if you look at the
average cooking events per day by location– there
are three locations. What I forgot to
tell you was that we looked at three
locations that are of different
socioeconomic status to see whether being
able to afford the cost would allow you to do more. You can see that kerosene
in the different groups was the most used, regardless
of socioeconomic status. Because it’s perceived to be
good and readily available– more than the LPG
and the ethanol. Again, consistently,
we saw that kerosene was the stove most used. And if you look at collectively
throughout the study duration, the first area here– this is
CleanCook, this is kerosene, this is LPG. So consistently, you could see
that there is a lot of stove stacking related to this. We also did an estimate of
the CO2 emissions associated with the different fuels to
see climate benefit from it. And I think in India,
and even in China, there is an ongoing effort to blend
methanol with ethanol to take away all the impurities. It makes the combination
even burn more efficiently with less emission. And you can see that,
compared to just the baseline, the intervention,
and the pure ethanol, you can see how much
reduction in CO2 emission you could
get from pure adoption of the ethanol and the
methanol, looking at them. If you’re looking
at black carbon equivalent, the same thing. So there seems to be significant
benefit from the environment, not only in terms of cost. And if you looked at what people
liked about their baseline fuel before, you can see the spread. And what they thought
about the CleanCook stove. It produced less
smoke, cooks fast, keeps kitchen
clean, looks modern. This is a different
group of people about 200 miles away from
where we did our RCT. So it seemed as if that was
at least consistent in terms of likability of the stove. And there were some
of the challenges, in that they didn’t know how,
the fuel didn’t last long. Because once you
put it in there, you couldn’t see or
shake it to see whether. So that is actually
very informative. Maybe it was just evaporating. Because what happened then,
as a business approach, it was to actually
put a cap on it, so that when you’re
not using it, you’ll put a cap on it so that
it doesn’t really evaporate. In times of fuel use and
being able to afford this, I’m sure you guys
will really like this in terms of willingness to pay. We also tracked how much
people were spending in terms of canister exchanges. We knew exactly how many
canisters people really needed to be able to
cook for their household. It turned out they were
using about a third. So we knew from the
sales information also that there was a
lot of stove stacking. And the data from
the sums used also showed us that they were not. And when the fuel canisters
were not available, they either moved to the
kerosene or the LPG stove. So this is also
good data to have before going out with a full. And then lastly, we did
conduct a willingness to pay for the stove, where all
the houses that were involved participated. And this stove actually
cost about $66, in terms of factory price. But the households
were told that they could pay a discounted
price of 19,000 naira, which is about $52. And if they declined
that offer, then they were invited to make up
to 3 bids for the stove, with any bid over
15,000 being acceptable. So in essence, which
was $42 to about $66. So really discounted. And 30% of the homes purchased
on average use spent about $42 on this stove, based on
using this stove for 3 weeks. That’s a lot of money
that people shelled out. And it goes towards showing
the likeability and potential success of introducing
this stove, when you start rolling out the 2,500 stoves. So in summary, it appeared as
if the ethanol and methanol appeared to be a likeable
and affordable household fuel, as demonstrated by the
use pattern and the willingness to pay out of pocket
for this fuel. And that’s why
the stove stacking and the incomplete
adoption which– no matter setting that you go
to, even people who have money, they use different stove
for different meals that they want to cook. And the findings are
a little limited, because we really didn’t
get the opportunity to get into the marketing
phase, where you have fliers, public service announcements,
to actually make the case for the
availability of these stoves. And the canister
refill data also appears to show a
cost-effective and accurate way to measure adoption when you
looked at the stove use pattern data, and you related that
to the purchase of ethanol. So this made us feel comfortable
that a commercial pilot could actually be successful
when deployed in a mega city like Lagos,
with about 17 million people. So this is really
a good experience for someone like me,
who usually doesn’t think in terms of the
economics, and policy, and the social science. But if we really want later
the randomized control intervention, that those of
us who are medical people do all the time, it would be
very, very important for us to understand how to do
a implementation science and looking at the
economics of implementing some of the
strategies that we’ve shown to be very effective. Thank you for your attention. KENNETH LEE: And final
speaker for the session is Dr. Santosh Harish. SANTOSH HARISH: So one
of the first things that I want to talk about. Ken showed this in his
presentation yesterday. This is the the
air-quality-life index, too, that researchers from
EPIC have developed to essentially translate
pollution levels into life expectancy estimates. And so one of the things that
I want to briefly mention, because I think this is a
point worth reiterating, is that air pollution
in India is certainly not a purely winter problem. It is not a purely
Delhi problem. It is not a purely
urban problem. And so what you would
find, for example, is that across the
Indo-Gangetic belt– in fact, starting from Punjab in Pakistan
all the way down to Bihar and parts of West Bengal– you have very high
levels of air pollution almost across the
board, urban and rural. So that’s something
to keep in mind when we’re talking
about policy solutions, and what the different
sources are, and so forth. Because it’s easy,
especially in Delhi, I think, to get
caught up in things like crop burning, or
vehicles, and lose sight of the bigger picture a bit. It’s a more complex
problem in some ways. So I’ll now briefly talk
about some of the challenges we have had when it
comes to enforcement of pollution regulation. So I think one of the first
things to point out here is that we do have a set of
fairly powerful legislations in place. We’ve also had a fairly
activist judiciary that has over time built
a fairly large volume of jurisprudence in how
to interpret these laws. And in many ways, in
the ’70s and ’80s, we were early starters globally. We were one of the first,
if not the first country to have a dedicated ministry
of environment and forests. But it’s not enough. There must also be enforced,
which is often difficult. Now as all of you are aware,
the difficulty in enforcement partly stems from the complexity
of air pollution itself. Fine particulate stem cells
are not a single chemical substance. They are a cocktail of
various things coming from various different sources. The composition of these sources
differs from place to place. When it comes to
regulation and enforcement, these sources typically
are under the jurisdictions of different levels
of government– center and state, municipalities. The pollution control
board takes care of a certain set of things. The transport
department takes care of a certain set of things. And so forth. So there is this
larger complexity to the enforcement,
which mirrors the complexity of
the pollution itself. Sola mentioned this
in his presentation. This particular study
came earlier this year. It was coordinated by the
Health Ethics Institute. And what this graph
basically shows are the number of
deaths attributable to different sources. And they have chosen a few
sources to highlight here. I want to take a
minute to just go over what I think are a
couple of salient points. First, the single
largest source, in terms of its
burden of disease, is residential biomass burning. Professor Olopade talked
about cook stove choices and cooking fuels
in his presentation. And worth thinking
about cook-stoves from a purely pollution
mitigation standpoint. Not just indoor
air pollution, but from in terms of its
contribution to outdoor air pollution, as well. We have a range of
studies which estimate that the contribution
of households burning firewood biomass
in their kitchens contributes a quarter to a
third of ambient air pollution. So it’s not just restricted
to within the houses, even in terms of the
contribution to the ambient. It’s the single largest
source out there. Now coming close
if you add up are power plants and heavy
industries burning coal. And in fact, even the fourth
category of anthropogenic dust includes, among other things,
things like fugitive emissions from plats from industries. So the formal regulatory
apparatus in India– the pollution control boards– have primarily focused
on the industries. Which makes sense
because these are a relatively limited finite
number of large point sources of pollution. And therefore much of
regulation has focused on how to mitigate those. And open burning,
which to a large extent are the stubble burning
that we keep reading about, also has a very large
burden of disease. So when we look at enforcement
of pollution regulation in the industries,
the compliance seems to be very low. And what complicates
matters a little bit is we don’t precisely
know how low it is for a couple of reasons. The measurement of emissions
often tends to be weak. And I’ll briefly
talk about that. Ken has already talked about
the conflicts of interest and that corruption associated
with measuring emissions from industry, as well. So that has a
large role to play. In addition, the measurements
and enforcement actions taken by the regulatory
boards are typically opaque to the public. And additionally, even
within the pollution control boards, what you would find
are that these are basically strewn over various files. Different regional offices
handle different industries, and so forth. So even within the
pollution control boards, there isn’t necessarily a
very clear understanding of what the
compliance rates are, what enforcement
actions have been taken in a systematic,
documented fashion. So this graph comes
from work that we’ve been doing with the Maharashtra
pollution control board. So in collaboration
with them, we basically consolidated the
various files and so on that were there, digitized
the results. And what we find
is that in excess of 50% of the manual
samples that they have are above the regulatory norms. We find similar
results from surveys that we conducted with the
central pollution control board in Gujarat, Maharashtra,
and Tamil Nadu. Basically very
high noncompliance rates among industries. One major constraint
in enforcement is that the regulators have very
limited tools available to them in responding to non-compliance. As per the Air
Act, noncompliance is a criminal offense. What the pollution
control boards can do when they find that an
industry is polluting above and beyond what
they are allowed to is to essentially
give a “show cause” notice, which in some ways
is a polite correspondence. They can threaten to
shut industries down. Sometimes they do, in
fact, shut industries down. They can mandate certain
improvements and processes, certain equipment that they
need to install, and so forth. Or they can take
them to court, which in general, most pollution
control boards are sort of loathe to do. Because these tend
to take a while. They are very
expensive, in terms of the time requirements from
the pollution control board staff. So the equipment
mandates, for example, are relatively easy to enforce. You can observe that a
certain piece of equipment has been installed or not. But these need not
necessarily translate into reductions in emissions. Because while you can
observe the installation, it’s harder to enforce
the correct design, the regular maintenance, the
regular use of these equipment. So there is this [INAUDIBLE]
which gets installed there. And it may not
necessarily translate into reduced emissions,
which is what we found. So what this
particular graph shows is the distribution
of emissions. The red line you see
are the standards that they were supposed
to comply with. And the columns you’ve
got here are basically different combinations of air
pollution control equipment that have been installed. As you can see
for the most part, everybody is above what
they’re allowed to be emitting, despite having these pollution
control equipment, which in theory, should ensure
that they are in fact very compliant. They should in fact be
emitting at a fraction of what they’re allowed to if
the equipment were designed correctly. So what can we do here? Over the last several
years especially, we have been witnessing
this dramatic change in the amount of pollution
data that’s being produced, and the computing
power that we’ve got to consume this
information effectively. We also have lessons from what
other countries have done. India isn’t the
first country to be dealing with industrial
pollution or non-compliance. So we have plenty
of opportunities to learn from the
experiences elsewhere to make regulation more
effective and efficient. So the next five slides,
I’ll borrow from the work that we’ve been doing
with these boards to strengthen monitoring, to
improve transparency, and then conclude with a couple
of ideas about the use of market-based instruments
for enforcement. So the first finding and
recommendation we have is to improve
emissions monitoring by better aligning the
incentives of the stakeholders. So Ken has already talked about
the third-party audit study and what we found. So basically, by ensuring that
these conflicts of interest do not hinder the truthful
reporting of emissions, there are fairly large gains
that we can expect in terms of, certainly, what’s being
reported as emissions, but also compliance. So in this particular
instance for example, the experiment found that,
by the pollution control board randomly allocating
auditors to the industries, and by paying them
through a central pool, and essentially
detaching the payment structure between the
industries and the auditors, we found that false
reporting reduces by 80% and average pollution
from industries reduces by 0.2
standard deviations. So not only is the reporting
more truthful as a result, industries start
emitting less CO2. The second point is
that traditionally, we have relied on manual
inspections for compliance. Now manual inspections
basically involves a team– either the pollution control
board, or accredited labs, and so on– essentially sending
somebody who climbs a stack, takes a physical sample, follows
certain protocols to do so, takes this to a lab,
and analyzes the sample. Now, these are fairly
time intensive activities. What we found is that typically,
for a given stack, a given chimney, this process
takes nearly a day. For regulatory boards that
are already understaffed and have various pressures,
what this results is that manual inspections
are restricted to once, maybe twice a year. Which is not nearly
the kind of granularity with which you would
expect these emissions measurements get done. Because it’s easy
to be on your best behavior when you know
somebody is coming, and so on, for a day. But what happens
364 days a year? So the future
certainly lies in what are called continuous emissions
monitoring systems, where essentially there’s a little
sensor that is installed in the chimneys of
these plants, which is relaying the data
to a central server essentially every minute. So this would be a paradigm
shift, as far as the visibility that the regulators have on
the emissions of the industries are concerned. But then, for the data to be
useful at all for enforcement, we still need to be wary of
the conflicts of interest and so on. These devices need
to be calibrated, so that those readings
correspond correctly with actual emissions. And it’s rather easy
to fudge the data. So it requires a significant
amount of vigilance from the regulatory boards
for this information to be meaningful at all. So we’ve been working
with the pollution control boards in Gujarat and Odisha to
essentially use the emissions data from the industries, and
categorize these industries into from one to five stars,
depending on their compliance. And this information’s out
there in the public domain on this website– MPCB.info, accessible
to anybody. And we believe that
this is something which could be fairly powerful
in holding both the polluters as well as the regulators
more accountable to being compliant to the norms. A month back, in September, the
Odisha State Pollution Control Board launched a similar
transparency program, which in fact uses the data
coming from these countries’ emissions monitors. So it gets updated more often. It’s based on more
granular data. And we’re certainly hoping that
other states do get on board and are interested in launching
similar programs of their own. But these basically talk
about better measurement and greater transparency. In some ways, the question
on the inflexibility of instruments available to
the regulator still remains. In the absence of flexible
proportionate regulatory instruments that the pollution
control boards can use, as I said, they are
restricted to these very blunt instruments, things
like disconnection or basically sending
letters of various kinds. As a result, the most
egregious polluters– people who are 10 times, 5 times
the norms allowed for them– do get shut down. For the most part, polluters
who are non-compliant are let off with a
slap on the wrist. One of the urgent reforms that
we need in regulation in India is the use of things
like pollution charges– basically monitoring instruments
that allow the pollution control boards to
systematically, transparently, and proportionately charge
industries for non-compliance. All these other
instruments would remain or could be used
in different ways. But it allows for
regular enforcement. And finally, from various
programs across the world, like the Acid Rain
program in the US, we’ve learned that what are
called emissions trading schemes or
cap-and-trade leverages the heterogeneity in the
costs of reducing pollution from among different industries
to basically all of the group to be compliant, while
reducing the costs on average for everybody. And there is immense potential
to try out instruments like this in the Indian
setting, as well. So these estimates are for
the Surat industrial cluster, based on inputs we
collected in our survey with the Central
Pollution Control Board. And we found that the
emission levels that would exist if all
the industries were to meet the
concentration standards could be met with a
cap and trade program, with a reduction
40% of the costs. And there are similar estimates
elsewhere in the literature, too. So there’s immense potential
for using instruments like this. So I’ll conclude with that and I
look forward to your questions. Thank you. KENNETH LEE: I would like to,
first of all, thank all of you for taking the time to speak. This is a pretty varied group
of interdisciplinary experts looking at the question
of air pollution from different angles. And so I’m going to summarize
the main takeaways, or at least how I have strung together
these different presentations. And then ask you a very
general question, OK? So what I learned from these
presentations is first, from Dr. Somanathan’s work,
is that there is clearly no rationale for further
investments in coal. But when we look at that
profile of generation plants, these are long-run investments. And if you look at
how much coal has been added to the portfolio
over the past decade, we’re stuck with this
for the medium run. And then when I listen to
Santosh’s presentation, we learn that there are a
variety of policies or programs that can be implemented today on
these same types of plants that could actually reduce
the emissions that are coming from them. And these would include
things like breaking the conflicts of interest
in monitoring, improving data collection, information
disclosure, monetary charges. And so those two
presentations for me fall into the
mitigation side– how what are we doing to
actually reduce air pollution at the major sources. From Dr. Jina’s presentation,
it was about climate change, but similar to climate change,
when you think about how people respond to levels
of pollution, there will be high costs
of adaptation. And I’m not sure
that this is well understood in the context of air
pollution– that it’s measured. And I’m not even sure that we
know what the best ways are to adapt to air pollution. With temperatures, A-Cs
are fantastic inventions. But with air pollution,
do air purifiers– is that the main thing? Or is it wearing masks? How do we adapt? And from Dr. Sola’s
presentation, for me, there was a connection
between what Amir was talking about about adaptation. And this fact that, even if
you can identify technology that is useful or
valuable, there needs to be a
consumer market that is built that enables
people to access that, or enables people to
understand that that’s there. And certainly that
is the case in Delhi, with things like air
purifiers and masks. So that’s why I actually
learned a lot from this panel. And so thank you. The question that I
have, that I would like for us all to discuss
is much more general, because it is a very
interdisciplinary group. And it’s simply this. What do you think
needs to happen in Delhi and in other
cities in northern India in the next year, the next five
years, and the next 10 years to tackle some of these
big issues with measurement of air pollution,
mitigation of emissions, and defense or adaptation? So I’ll start with you again. AMIR JINA: So
specifically about Delhi. Well, actually to sidestep
the question briefly and talk about something you’d
mentioned before, and what the potential
adaptations or some of the adaptation
costs would be. I mean, we all know
that you can here spend on an air filter, which
you wouldn’t be spending money on if the air
quality was better. I don’t have any HEPA air
filters in my apartment. That’s not money
I’ve had to spend. But there’s other research
showing, obviously, a lot of it based in the US,
that actually there’s significant avoidance behavior
when it comes to air pollution. And so, as probably everyone
here has done at some point, there’s days when you
don’t go out, where you would have gone otherwise. There’s something
either productive– whether it’s work– that you have not done
because there was high air pollution, or something that had
amenity value– some enjoyable activity which you didn’t do. So something that’s
costing you in terms of quality of life, which are
costs which I just don’t think are very well understood at all. So all of those
behaviors are things which we could think of
as costs of adaptation. And one that was raised was
that induced brain drain that may be happening. What is happening to people who
are either not coming to Delhi or leaving Delhi who
have the choice to do so? And what is that doing, more
broadly, to the economy? We could also think of
that as effectively– I mean, it’s a damage. But there’s kind of
an adaptation cost there to what we’re losing. And I feel like perhaps
there’s better people able to answer on
this panel about what Delhi should be doing in
the next couple of years. But in the thinking
of the near term, where some of the
air pollution is going to be completely
unavoidable, it’s not that we can shift
away from most of the polluting sources over the course
of the next year, or maybe in the next
couple of years. It’s in trying to change
behavior sufficiently that it’s limiting those health costs,
which are probably the largest costs that will be
associated with that, even though there’s labor
effects and other things. But for example, as
has been pointed out by many of the speakers
either in their presentations or over lunch or dinner, when
you go to Beijing or Shanghai, you will see everybody
wearing a mask and taking some
kind of protection. Here in Delhi, you
don’t notice that level of awareness of the negative
costs of air pollution. And so I think
one of the things, just speaking near-term, is
trying to have enough awareness to those personal costs. Removing this idea
that the air pollution is going to be bad anyway,
who cares that somehow we shift people off that
reference point of, it’s always bad
this time of year. It could be worse. And maybe I’ll do something when
it’s twice as bad next week. But in this week, it’s
worse than most cities in the world at the moment. But shifting that
view to something where people can start taking
some of those protections. I think in the near
term that’s one of the most important
things as we develop the policies to shift
away and improve the environmental quality. I think I’ll pass this. SOLA OLOPADE: In
the last two days we’ve learned a lot
about the efforts being made in India to enact
policies that are very good and directed towards
correcting or reducing some of the pollutants that’s
causing a lot of the health problems all around India. I think there is enough
sophistication around in India with respect to monitoring. So in terms of
environmental monitoring, I don’t think that is an issue. And as Santosh just
indicated, there’s a need to manage the conflict
of interest and the corruption around deployment of the
monitoring abilities that exist in India to
directly monitor the emissions in real time. Because if you are
in Delhi, and you open the newspaper every day,
you can see the regional AQI. So there is that ability to
actually do the monitoring. It appears from listening
to the Member of Parliament, the Doctor [INAUDIBLE]
and also Santosh– there seems to be a lot
of progressive policies on the books. The enforcement is
where things fall short. And the Indian
government would have to decide what to do with that. With respect to
adaptation, I know we’ve been focusing
on adaptation mostly around the increase
in temperature. But when you look at the
impact of climate change, it’s not only the increase
in temperatures that we see. There are low-lying
areas that are going to experience a lot of floods. You’re going to have
deaths, as Amir said, related to the heat strokes. I don’t know how we factor in
the cost of these raging fires that we are increasingly
having a hard time controlling. I don’t think the cost of
investment in air conditioning or filters should
be where we start. Because as the temperature
continues to go higher, we need to actually invest in
investigating the materials that we use for
building our homes. Because repeatedly, people were
asking questions yesterday– why are we just focusing
on particulate matter? What about the VOCs? What about the other POPs that
have been produced indoors? And if you look at
some of the materials that we use for building
some of these homes, they need to be tested
and re-evaluated for the higher temperatures. Otherwise you might
be in an environment like this, where
it’s not everywhere that you’re going to
have the air conditioner. You might get a lot of gas in
from all of these materials that one really tested
to operate safely in such a high environment,
high temperatures. So I will be very interested
in hearing a lot from Amir about the mitigation costs,
based on the projections that you guys did. And being able to
exponentially project costs related to lives in Africa. Really, when I was
looking at the map, there was really
no data collected from the whole of Africa. Yet we are making projections
based on cultural practices and differences among other. It may not translate
directly the way we want it. And I’ll be interested in
knowing what other adaptation costs were put into the
formula that you guys used to make all of the projections. But I think India
is doing a good job in trying to mitigate this. But it’s ground zero for
most of the pollution. The enforcement of the policies
has to take the front burner. Otherwise nothing
is going to change. And I hope that, as we come
back next year and the few years to come, that we’ll
keep hearing good things about the enforcement of these
pollution-mitigation policies in India. SANTOSH HARISH: So
in some ways I’d want to change the framing
of this a little bit once again, because I think
it’s problematic to think of it purely from a
city’s perspective. Indeed, the North Indian
air pollution problem is really the North
Indian regional air shade. And when it comes
to the solutions, I think one of the things
we definitely need to do is strengthen regulatory
capacity where it already exists, as opposed
to diluting efforts in that direction, looking
for new innovative solutions to sources, dispersals of
pollution, which are harder, in some ways, to target. So among the things that
we can do in the next 5, 10 years, I think
certainly, is strengthen the regulatory capacity in
terms of reducing pollution from the big industries, from
the power plants, and so on. And otherwise, follow all of
the above approach, pretty much, with the polluting sources
where we do have some capacity. For example, with vehicles– moving the fleet towards things
like the Bharat-VI norms. Phasing out older
vehicles, and so on. I think that’s a
reasonable approach. The programs like the
PMUY, which was our program to get BPL households
to shift to LPG, I think are a very important pollution
mitigation intervention. And not just clean household
energy intervention. And so that’s something
I think that needs to be aggressively followed–
ensure that people are essentially buying the
second, third, fourth LPG cylinder then, and
continue to use them. Crop burning, I think,
is another major area. But I know this is something
that Soma has talked about and written quite a bit. So I defer to him
when it comes to that. Yeah. E SOMANATHAN: Yeah. I completely agree with
Santosh that we shouldn’t think of this as a city
problem, because it’s really a regional problem. And that said, I
think that all the measures that Santosh talked
about– better monitoring, having instrument, allowing
fines, rather than these all or nothing thing
of either you basically do effectively nothing, or
you shut the industry down. This is a huge problem. And all those things
have to be done. But I think I’d like
to step back and ask a bigger question, which
is, why is it that so many of these measures
which are obvious. I mean, it doesn’t need
a PhD to figure out that, if you really want something
in between closure and nothing, right? Why is it that none of
these things have changed? Right? It’s not that we’ve had air
pollution now for a long time. We’ve had water pollution
which is much more severe for much longer. And yet none of these
institutional changes– none of these kind of
fairly obvious measures have been taken. Right? And I think that one has
to ask that question. Because otherwise,
I think we are going to be tinkering at
the margins, basically. We have a huge problem. And then you’re going to
have just very small epsilon changes at the margins. I think there are
some exceptions. Like, I agree the AGLR
program is a big deal. And I think that that is
the beginning of making a very big difference. Because we know that
the household sources are the biggest single source. So you have to deal
with that problem. And there are some
other areas where I think the regulation has
moved quite significantly. For example, I think on the
fuel standards for refineries, and on the standards
for new vehicles, which have been shifted forward
from what was 2024 to 2020. This, I think, signals a
new political attention being given to
environmental issues, and air pollution in particular,
which we didn’t have before. Because after all,
we are at least 10 or 15 years behind Europe. And Europe is behind
California with regard to vehicle regulation. So eventually, though,
slowly the political process is catching up. But I think that
looking forward, I don’t think this political
process ever will catch up. And there is a very
good reason for that, and it relates to a question
that somebody asked just now. We are all focusing
on particulate matter. What about all the
other pollutants? The reason is really that
all of these problems are highly technical,
highly complex. It’s highly expensive to
measure the pollutants. It’s highly expensive to
know what they do to people. So it takes a huge amount
of institutional resources just to understand the problem. By the time that problem is
understood by some scientists sitting in a lab, right? And then it has to make its
way through a million workshops like this before it’s
out there in the media. And then the politicians feel
that they have to respond. By that time, 30
years have passed, and all the people who are
affected are already dead– quite prematurely. So what I’m saying is, I
don’t think that this process is really going to do the job. If we leave it to
this, it’s just going to carry on like this. And we’re going to go on
having this same workshop 30 years from now. It will be a different
set of faces. Ken may still be here. (LAUGHING) Santosh
may still be here. Some of you will still be here. I won’t. But I think that we need to
think about it a little more here at the long-term level. And I feel the reason here is
because pollution problems are such a highly complex
and technical issue. I don’t think having the
regulator directly controlled by the political executive,
as we have in India, is a good idea. I think this is one
of those issues where decision making
should be delegated to a technical authority
which is autonomous. Because precisely for this
reason– that the incentives to do something about it,
the benefits are so unclear, not easily understood, that
there isn’t political pressure to do something. The costs, on the
other hand, are falling on Volkswagen, and Ford,
and Maruti Suzuki, and so on. They understand perfectly well. And so they are
going to go all out to stop that from happening. And so that’s why it doesn’t. So we understand this
dynamic very well. And if you want
to break that, you have to have an
autonomous regulator who is scientifically competent. So I really think we
need a change in the law. Unless we have a
change in the law, we’re not gonna get beyond
tinkering at the margins. We need new legislation. And that legislation has to
say that the regulator is going to be somebody who’s
appointed for a fixed four-year or five-year term, can
only be removed by impeachment. They’re independent. They have full
autonomy for hiring. They have some
fixed budget, which is some percentage of GDP. Take a percentage which is sort
of common among the countries that actually do serious
environmental regulation. Take that and say,
OK, that’s what they’re going to get as
some levy from all industry. Give it, and then let them
hire the scientific people. Write down, what is
our technical capacity? Is there even one
serious toxicologist? One epidemiologist? One environmental economist
sitting inside the regulator? No. Not even one, right? And we need hundreds
of these people. Because this is a huge country. So we need that. We need, I think, that autonomy. And we need also a requirement. It’s not enough
to have autonomy, because then you can
have a regulator run through it, right? You need a requirement
that the regulator has to do scientific studies to
justify any kind of regulation. The regulator must have the
power to set pollution fees. It’s not enough to
have regulation. So Santosh talked about
improving compliance with regulation. But it’s always
in a context where there is some physical
limit on some pollutant. But that’s not going to be
efficient most of the time. Most of the time we don’t
want to set a physical limit on a pollutant. We want to put a charge on it. Because then you can reach large
parts of the pollution problem that you can’t reach. You can put a charge on
nitrogen as fertilizer, because that creates a
huge pollution problem. That’s a non point source. Every little fee, you’re
going to put that. You cannot regulate that. Right? So you’ve got to have the
flexibility to have fees. You have to have a requirement
for scientific cost-benefit analysis, that every
fee and every regulation has to be justified by studies. Regulator must be obliged
to publish all data. If we get a law like
that, and I don’t think we’re going to get
it anytime soon. But we are certainly not
going to get it unless we realize that we need it. That’s the first step. People like us have to realize
that we need such a law. That if we don’t
get a law like this, we are going to make
very little progress. I think all of us need to
start campaigning for a law like this– for an
independent regulator, with a requirement for
transparency, publication, data, scientific studies. We need to start
this process now. And maybe in 10 years
we’ll get a law like this. That’s a good end.

Leave a Reply

Your email address will not be published. Required fields are marked *