Rajat Kochhar on Agricultural Market Power and Farmer Adaptation to Climate Change

SHRUTI RAJAGOPALAN: Welcome to Ideas of India, a podcast where we examine academic ideas that can propel India forward. My name is Shruti Rajagopalan and this is the 2023 job market series where I speak with young scholars entering the academic job market about their latest research on India. 

Our last scholar in the series this year is Rajat Kocchar, a post-doctoral scholar at University of Chicago’s Energy and Environment Lab. He has a Ph.D. in Economics from the University of Southern California, and his research lies in the field of environmental economics, in particular, on the understanding the factors that incentivize adaptation to climate shocks. We discussed his paper, “Does Market Power in Local Agricultural Markets Hinder Farmer Climate Change Adaptation?” We talked about the impact of distortionary policies and regulations on farmer’s ability to cope with weather shocks in India, the agricultural produce market system, the choice of crop mix, and the effectiveness of water audits in the UK and much more. 

For a full transcript of this conversation including helpful links of all the references mentioned, click the link in the show notes or visit mercatus.org/podcasts. 

Hi, Rajat. Thanks so much for being here. This is such a pleasure.

RAJAT KOCHHAR: Thank you, Shruti.

RAJAGOPALAN: This is actually one of my favorite topics. Both in terms of climate change and the agricultural sector and market distortions, and everything crazy that’s happening in India. It’s also been in vogue a lot because India tried to actually get rid of the APMC laws, these are the Agricultural Produce Market Committee laws and the Mandis. Before we get into the broader political economy context of this, what you are looking at is something very, very specific. You’re looking at the effect of competition in the produce market that farmers face when they actually go to sell their produce and how that impacts how farmers are able to cope with negative weather shocks.

In India, we, as it is, have a broken system. There are problems when it comes to irrigation. There are issues when it comes to information about weather shocks and, also, how frequently they’ve been coming about given rising temperatures and climate change and so on. What you find is quite interesting that actually the market structure quite significantly impacts farmers’ ability to cope with a negative climate shock or effects of short-term climate change and so on. This effect is quite large. You find that they’re able to have almost 5% higher output for each additional day of extreme heat.

The way they’re able to do that is by increasing agricultural inputs, but this is only farmers who are selling their produce in a relatively more competitive system of agricultural Mandis. Farmers who are not selling in that competitive system are not able to cope with these weather shocks quite as well. Before we get into what this means and so on, can you first just walk us through what is this Agricultural Produce Market Committee or Mandi landscape in India? Then we can get into the rest of it.

APMC Laws and Market Competition on Farmers' Response to Climate Shocks

KOCHHAR: Thanks, Shruti. That was actually a great summary. The APMC or the Agricultural Produce Marketing Committee, these are basically the physical spaces that have been set up by the government of India where farmers are required to go and sell their produce. Basically, a farmer, let’s say, they harvest their crop and then they have to take this produce to these APMC Mandis. There’ll be intermediaries there who will bid for the commodity, and they’ll offer the farmers a specific price. The laws are somehow strange in the sense that these are state-specific laws.

What tends to happen is that the farmer, let’s say, if you’re looking at Punjab and Haryana, it’s not possible for a farmer to go from Punjab and sell their produce in a Mandi in Haryana. This is the kind of variation that I’m using in these APMC Mandis to answer questions related to climate shocks, but basically, the structure is that the farmer has to go and sell their produce to intermediaries in these physical locations that have been set up by the government. Except if it comes to fruits and vegetables, they’re more or less not allowed to sell their produce to private corporations. Therefore, they have to sell it to the specific Mandi, and they cannot cross state borders.

RAJAGOPALAN: Now that we have this general landscape, how does this impact competition and how much of that are you able to observe and measure? At what degree of granularity can you tell us about how competitive Mandi A in Region X is relative to Mandi B in Region Y?

KOCHHAR: Think of it this way. Let’s say, I’m a farmer today and I have two Mandis in my state. Now, if Mandi A is closer to me, I would prefer to go to Mandi A. The problem is that if Mandi B is extremely far off, then even though Mandi B exists in my particular state, it’s very hard for me to go there because, first of all, it will add to the transportation costs. The way the competition is measured, and this is actually taken from an Allen and Atkins Econometrica paper, where what they use is three things. There are three variables that go into the competition. For each Mandi, we calculate this competition measure and it’s dependent on three factors, like I said.

The first one is, how many Mandis are around me? If there are 20 Mandis around me, versus if there’s only 1 Mandi in my state, then, of course, the place with 20 Mandis will have more competition because a farmer has 20 extra options, 20 outside options. The second thing that we look at is the distance between these Mandis. Even if let’s say there are 20 Mandis, but all of them are 100 kilometers away from me, that’s a problem. The third thing that we look at is the size of these markets. I was very surprised to see, when I was actually researching on this, that the size of the Mandi is not uniform across India.

Some places have these large markets where farmers can come in, they have storage facilities, whereas in some cases you don’t have a storage facility, you just have a platform. Because you just have a platform, we assume that the Mandi is not that large. We take into account also the size of the Mandi, which we capture by the amount of trade that has happened in the last 20 years.

RAJAGOPALAN: This is similar to Shoumitro Chatterjee and looking at spatial composition and how that impacts market power within agricultural markets, right? Or are there more layers added on to that?

KOCHHAR: The only layer that was added on to Shoumitro Chatterjee’s paper was the size of this market. I also wanted to take into account that some markets are just bigger. The fact that a market is bigger, you probably have higher competition there because there are more intermediaries. I actually want to go to a bigger market, rather than a smaller market.

RAJAGOPALAN: Now that we understand that you can both figure out legally what the APMC laws are and now you can actually do that at the granular level, I wanted to ask you the same question about weather shocks. What do we mean by weather shocks or negative weather shocks when it comes to farmers? Aren’t these different for different kinds of produce and different kinds of farmers in different kinds of regions? What is a good way to think about this for what you are trying to measure?

KOCHHAR: There has been so many advancements in meteorological data recently. You have this extremely, spatially granular data on the temperature and precipitation. What we did in our paper was that we take these—usually, they’re 9 kilometers by 9 kilometers or 4 kilometers by 4 kilometers grids—and they give you the amount of time that was spent in a particular temperature bin within that grid. Then what you can calculate is that for a particular crop, let’s say, Rabi or a Kharif crop, you know exactly the number of hours that the particular crop spent from the day of sowing to the day of harvest in a particular temperature range.

What we did and how we define weather shocks here is that we basically classify the amount of hours spent for a crop from let’s say 0 to 15 degrees, then 5-degree temperature bins. We have 15 to 20, 20 to 25, 25 to 30, so on and so forth, up until 35 and higher. In our paper, we call a weather shock as temperatures that are above 35 degrees Celsius. This is actually, you can change these temperature bins. You can have 32 degrees Celsius, 35 degrees Celsius. We just tend to follow the Robin Burgess and Michael Greenstone paper where they look at mortality and that was 95 degrees Fahrenheit or 35 degrees Celsius. Exactly that’s what we are doing, but also, we can carry out robustness checks with specific different temperature ranges.

RAJAGOPALAN: When you talk about negative weather shocks, you’re really talking about heat, but does that also mean it’s interconnected to drought in some way or access to water? Is that part of the temperature measurement or is that something separate you need to look at separately?

KOCHHAR: I would say that drought is more a long-term phenomena. Basically, you have a drought that, like let’s say you have—we are looking at the number of hours that a particular crop spends in a temperature range. It’s possible that the drought is more a long-term phenomena where you look at the number of days where there was no precipitation or something. We are looking at very short-term weather shocks and drought is more like a long-term phenomena that we don’t really look at in this particular paper, but you’re right, that’s something that we can also look at.

RAJAGOPALAN: Now you find that farmers who are going to the more competitive Mandis actually end up dealing with these shocks better. Their output is actually better than farmers who are going to less competitive Mandis. The first question I have is how do the farmers in more competitive Mandis actually manage to accomplish that? Then I want to talk about what’s the underlying mechanism that is at play, making all of this work.

KOCHHAR: You’re absolutely right. What we tend to find is that if there is a farmer and they have access to more Mandis, then there is a climate shock, they basically don’t lose as much produce as a farmer does who’s not in a very competitive place. The effect is somewhere to the extent of 4.5% for every additional degree day of heat. Then we look at the mechanisms. The mechanisms was something very interesting. You could have a lot of potential mechanisms here, but the thing that we were most interested in was actually following from Shoumitro Chatterjee’s paper, and he looks at prices.

The mechanism here was that we were trying to look for—there is already a pre-existing difference in prices, which Shoumitro Chatterjee also finds. What we find is that this difference tends to get exacerbated. The logic is simple. In a climate shock, you would expect the supply falls. Because the supply falls, then let’s say the short-term demand is fixed or inelastic, you would expect that prices will rise. What we find is that these prices rise, but more in high competition areas. And the farmer knows this, the farmer, in anticipation of these higher prices, then they go to the changing their input usage, which ultimately helps them alleviate some of the negative impacts of climate shocks.

RAJAGOPALAN: Now that is directly going into how the farmers—[the] organization of their farming activity is taking place on each farm. If they expect that it’s going to be a more competitive market, and that’s going to actually give them access to better prices and so on, now they are more likely to take on additional risk of putting in more inputs in the expectation that they will likely get a better price. Do I understand that correctly?

KOCHHAR: Exactly.

RAJAGOPALAN: That’s the expectation.

KOCHHAR: That’s the expectation. It is an anticipation of this higher price that farmers are like, okay, I know that there has been a climate shock, but I also know that if I spend more money on, let’s say, irrigation, I call a truck and I ask them to water, something like that. If that tends to happen, they can spend that money because they know that they will be able to recoup at least some of that, because of the increase in prices that has happened due to the supply shock.

RAJAGOPALAN: In some sense, that’s not surprising. Farmers actually get better prices when there is a negative supply shock as long as they are not the farmer facing the negative supply shock, right?

That’s typically how that works out. The expectation is if there’s a negative weather shock, they think, “Oh, this year is going to be a bad crop year for some other farmers, but if I managed to actually put in additional inputs and save my produce or increase my produce, I will end up getting access to far better prices.” Is that about right?

KOCHHAR: That’s absolutely right. Another interesting aspect of this and that I want to study in future papers is basically how much of this markdown. It’s a monopsony power, and how much of this markdown is actually being passed on to the farmer by the intermediary? Is it possible that there has been an increase or decrease in markdowns caused by climate shocks? That’s an interesting question that I don’t study in this paper, but that is something of interest, I think, because it’s possible that climate shocks may increase or even decrease in equality. There is this possibility of a result going both ways, and that’s something that’s interesting to study as well.

RAJAGOPALAN: One of the main mechanisms through which the farmers are able to overcome this negative weather shock is increasing the inputs that they are adding to their production process. Now, I’m thinking about whether that’s almost a perverse trade-off in the sense that to overcome a short-term weather shock, you need to increase inputs, which may end up going into a space where you damage long-run climate change problems because you may be boring too much to tap into groundwater. For instance, or you might run a diesel generator a little too much and start polluting the environment. Or it could be an increased use of fertilizer, which will worsen your soil quality in the long run. Do you see this trade-off and is that trade-off being fundamentally caused by more competitive agricultural markets? 

KOCHHAR: No, that’s a great question. Honestly, I haven’t thought of this that well, but I know that there’s a paper, and you must be knowing that, the Chatterjee and Zaveri paper, where they basically see that groundwater extraction can be a problem in the long run. I think, yes, in the sense that when you have these farmers who are basically tapping into input usage, they can harm long-term adaptation. That’s actually pretty interesting because even in this paper, the first part is, basically, we find that there has been no long-run adaptation. Farmers are adapting in the short run, but it’s a problem that they have not been able to adapt in the long run.

Actually, this is not just relevant to India, but also in the US where Marshall Burke and Kyle Emerick actually find this, that there has been no long-run adaptation in the US. It is very much possible. I don’t know the mechanisms yet, but it’s possible that’s because they tend to overuse some of these inputs, which harm them in the long run.

RAJAGOPALAN: You know what, Shoumitro Chatterjee, Rohit Lamba, Esha Zaveri, what they find in that paper on groundwater tapping, if I remember it correctly, what they find is it is actually farm subsidies and minimum support prices. That is really what is distorting the incentives such that people actually tap in more and more into groundwater usage. It is not market power that is driving their result.

I guess what I’m asking you is, is market competition that driving force which is going to start impacting long-run climate change both decisions at the micro farmer level and larger agricultural level is in market competition that’s going to push that over the edge? Because what they find is the opposite. They are looking at non-market. They’re looking at less competitive, terrible government subsidies that actually lead people in Punjab and Haryana to export water-based grains and produce in areas which should not be producing that, but they do it because of the minimum support price.

KOCHHAR: No, you’re absolutely right. That’s a very interesting point, and I hadn’t thought of it that way. Another thing here: we also find evidence that there has been a change in crop mix. Now if you are actually changing the crop mix, it‘s not going to negatively impact at least your long-run adaptation through usage of higher inputs. That is one way to minimize the negative impact on long-run adaptation by changing the crop mix. For example, if the government feels that, oh, with higher market competition, farmers have an incentive to use more inputs, it‘s possible that they can actually incentivize them to change their crop mix.

We do find evidence of some of that. We use Herfindahl index to basically show that the crop mix increases with an increase in climate shocks in areas with higher market competition. That is something that can also be looked at. There are not just this strategy of using inputs, we also show that there is this other strategy where you can also change your crop mix.

RAJAGOPALAN: I guess the other thing that might be interesting to look at is, there‘s a set of inputs that can be used to increase the output and they are both complements and substitutes. Sometimes you can substitute fertilizer use for increased use of water or quality of seeds or something else. It would be interesting to see if during negative shocks, farmers are more likely to indulge or overuse those inputs, which are subsidized relative to those inputs that have rational market pricing. Then you can quite cleanly separate the difference between what is the market competition-driven effect versus what is the subsidy distortionary effect. I think that might be the next step question that I would ask.

KOCHHAR: No. I think that‘s super interesting, and that can easily be done because you know exactly the inputs which have been distorted by the entry of the government.

RAJAGOPALAN: Also, to what extent?

KOCHHAR: Also, to what extent. You can always check that, which are the inputs that have actually been used more. That way we don’t tend to blame market competition then.

RAJAGOPALAN: No, it’s not so much about blame, it’s more about what is driving the short-run versus long-run mechanisms. Why is it that farmers who are normally the—it’s a profession that’s gone on for a very long time, oftentimes it’s hereditary, it’s done at the family level, passed on from parents to children. Why is it that this group won’t think long-term when other people would think long-term? There’s something funky going on in the system, either being driven by the output markets or being driven by input markets making them not think—

KOCHHAR: In the long run. Exactly. It could be more or less to do with subsidies, but you’re right, that’s something that needs to be tested and seen as to how market competition could incentivize people to use inputs that are not good in the long run.

Farmer Dynamics and Long-Term Strategies

RAJAGOPALAN: One other thing I want to ask you is, how does this entire system work when there is a negative shock? Presumably, when there is a negative weather shock, it’s not a single farmer who’s affected by it, nor is a single intermediary affected by it. It’s that entire region of people who—and these are repeat interactions, it’s the same intermediaries or very similar group of intermediaries buying from very similar groups of farmers. Are there other things that you see going on within those networks that help with either over-extraction in the short run or help cope in the short run versus long run that is being driven by both ends, both the demand and the supply ends? Because this is such a repeat game among a relatively small number of players.

KOCHHAR: Yes, that’s a great question because when I presented this paper, a lot of people have said that, why is it the case? To put it in the political context, there were some farmers that actually were not happy with repealing the farm laws. Then you would expect as to why that is the case, that what is actually happening. Then you tend to look at this literature on relational contracts. You can expect that intermediaries and farmers have a good relationship because there is this repeat interaction.

Then it’s possible that the farmer is told by the intermediary that, listen, I won’t be able to offer you a higher price this time, but I can help you, for example, with the marriage of your daughter or your son later on, so I’ll give you a loan, I’ll pass you off a loan. In that sense, it’s I think important to understand how are these repeat interactions also maybe nullifying the negative impacts of the farm laws. That’s something that is yet to be studied. I haven’t seen papers that yet do that. That could also relate to what is happening with these increased climate shocks, and where is the excess prices that you get due to the supply shock passing onto?

Is it possible that farmers are now getting a bigger share or are intermediaries still keeping a bigger share and not passing it off because they’re helping them, let’s say, to get a loan or to help with the marriage of their children later on?

RAJAGOPALAN: Yes, that’s a much more specific question of producer and consumer surplus and how that split works in a repeat game. When we’re talking about these kinds of climate change adapting strategies, another not-so-short-run strategy is questions like cold storage. Better warehousing, better transportation because that is also a substitute for being trapped in the regional market. Whether your regional market is more competitive or whether it falls under this very tight and restrictive APMC licensing, the biggest problem with all of it is that it’s regional, and not in a sense national or global.

National and global, it’s once again a question of the kinds of inputs you have, the kind of cold storage you have, warehousing you have. What does that look like when you start thinking about weather shocks? Do those systems become more important in terms of coping that is in good weather years you’re able to smooth out hopefully for bad weather years and have a long-term adaptation strategy, or am I going too far? This is much more specific long-term adaptation strategies have really nothing to do with this kind of smoothing out of agricultural supply shocks more generally.

KOCHHAR: I think the best adaptation strategy here is conditional on you wanting to continue to be a farmer. You should be changing your crop mix or using these high-yielding variety of seeds, which can handle, let’s say, these climate shocks. The interesting part, and that is also I’m studying in a few other papers that I’m working on right now, is I think the best adaptation strategy for farmers, if they suffer from repeated climate shocks, one of the best strategies is basically to migrate out. The thing is then if you talk about, oh, yes cold storage will help, everything will help, but ultimately what’s going to happen is that if you want to increase your income, you have to migrate out.

I think one of the strategies that the government can potentially look at is how can it create more urban jobs, which can then help attract some of these surplus labor in the rural areas. An interesting part there, and I’m working with Clément Imbert on this is MNREGA. Now, the problem is this with MNREGA, the MNREGA problem is that if I have a program incentivized by the government for me to work, in case there is, let’s say, no jobs for me, I have lesser incentive to adapt by moving out because I know that even if there’s a climate shock that hits me, I have this MNREGA policy to fall back on, but that MNREGA policy is not like a long-run adaptation thing. How long will you continue with MNREGA or how long will you just allow people to come in and work and do unproductive work?

RAJAGOPALAN: There’s also limit on the number of days that NREGA can—

KOCHHAR: It’s 100 days or something. The interesting question that I wanted to study was that where this MNREGA is being implemented really well, do we see farmers adapting by not moving out and actually using more inputs because they know that, okay, even if the crop fails, I have this safe share option of going and working in MNREGA. Whereas in places where MNREGA is not that well implemented, it’s possible that farmers actually move out due to climate shocks. The problem with actually modeling adaptation in a model is that there are so many avenues that you can use to adapt.

RAJAGOPALAN: This has to be a general equilibrium question after. 

KOCHHAR: Exactly. In a paper, at least in a single paper, you can only look at a few adaptation mechanisms. You can, for example, move out, you can migrate out, you can basically—but, also then who migrates out? That’s an interesting point that I wanted to look at, which is, is it possible that I have four sons and I know that two of them can stay in the village and I can send two of them away? Household dynamics also play a part in who can migrate out and how you adapt.

Adaptation that way is a very, very interesting sphere of study that actually is just catching up with economists because most of the times we were just focusing on mitigation. How do we do carbon mitigation? Now the IPCC comes and says that, listen, there’s going to be climate shocks, there’s going to be climate change, there is some effects that cannot be reversed. Now how we adapt, is I think the more interesting part.

Challenges, Distortions, and Potential Transition Strategies in Indian Agriculture

RAJAGOPALAN: On this, I guess, you can study things very specifically one distortionary question at a time or effect at a time, or you can also imagine a different landscape. What if we dramatically re-envision agricultural sector?

We start liberalizing agriculture, whether in one question is, of course, of terrible inputs and distortionary inputs, but there are also other restrictions. Converting agricultural land use. There are questions of all MSPs and what people end up producing as opposed to what is better suited for that climate. There are questions of landholding size. A very large part of the increased input use is because of landholding size.

KOCHHAR: That’s the paper that I’m actually writing now with someone.

RAJAGOPALAN: Tell me more.

KOCHHAR: It’s just in a very, very initial phase. We are downloading, actually, this data from the agricultural census about landholding size and we are trying to understand does landholding size play an important role in whether farmers are able to mitigate some of the negative impacts. I’ll give you an example. In Bangladesh, one of my friends was working on this, who’s at Yale now. He sees that in areas where you have these climate shocks happening regularly, farmers with increased land holding size were actually able to adapt better, and the reason was through shrimp farming.

What tends to happen is for shrimp farming, you need a bigger land size. You let the seawater come in and then you have a shrimp business, but that’s only possible if you have a large land holding size. That is something that I was very interested in looking at in India as well. We are looking at that. I don’t have results right now to share with you, but that is something that I’m very interested in. You’re absolutely right. Landholding size plays a role. What the government distortions are doing, they play a role. Market competition plays a role. A lot of interesting things out there, yes.

RAJAGOPALAN: This is really a super fun area. I think there is a lot more thought to be given to short-run versus long-run adaptation strategies. Some thought on whether they are aligned in the same direction and not actually trading off against each other in some other way. 

I want to ask you the political economy of the APMC itself. This is very indirectly related to your paper, but you have a discussion in there on how the intermediaries are not only able to extract rents and curtail competition, curtail diversity in crops, all sorts of effects.

The other part is who becomes an APMC Mandi licensed trader in the first place? These are typically people who are highly influential in that area, who have some political connections, or oftentimes family members of the politicians end up becoming APMC traders and all. Can you tell me a little bit more about that landscape and how hard it is to transition out of that landscape?

KOCHHAR: Right. I think Shoumitro also refers to this in his paper. It was very interesting for me to see that the person who grants you a license—that’s another interesting thing; APMC laws also make sure that not everyone can become an intermediary.

RAJAGOPALAN: Yes.

KOCHHAR: You or I just cannot go and buy from the farmers in the APMC markets today. We need a license. The interesting question is that who gives you a license? It‘s basically the existing intermediaries who give you a license.

RAJAGOPALAN: It‘s a club?

KOCHHAR: Yes, exactly, it‘s a club and they have an incentive to not give you the license because, of course, the more the players, the more the competition, which will increase the price to these farmers and reduce the amount of rent that they‘re seeking. Yes, that was very surprising that it‘s basically the intermediaries themselves who decide who gets to trade in these markets. In that sense, yes. I think changing even the licensing rules and who can come in and actually buy the produce of the farmer can go a long way. Even if, for example, let‘s say removing the APMC structure is very hard. Just removing the licensing structure in that particular area. Again, the intermediaries may go up in arms against the government. I don‘t know how politically feasible these things are.

RAJAGOPALAN: This is an area where there is not one, but like six transitional gains traps all laid out one after the other, right?

KOCHHAR: Exactly. Also, what I study in the paper is what if it’s removing these state borders? How will that help? Another thing that you can study is adding more Mandis and see how that helps with long-run adaptation. You can also see, for example, if we improve the highways between different Mandis, and it’s possible that brings down the transportation costs, and so how that will help. A lot of different avenues that the government can use. It’s just a matter of, I think, time, as well as how politically feasible these things are.

RAJAGOPALAN: The system is fundamentally irrational. After reading your work, Shoumitro’s work, I am beginning to realize APMC is not the reason for all the flaws and problems in agriculture, but the system is so irrational, it literally has no reason to exist. I think that I’m convinced of.

KOCHHAR: I go into the history of it a little bit too in the paper. It’s basically started when the Britishers were trying to fleece us off cotton, and they wanted the mills in Manchester to use the cottons. They started it. It just tend to happen that we were so influenced by the Soviets of a centralized economy, that we thought that APMCs were the best way to go forward. Surprisingly, we haven’t built that many markets since 2001 as well. We stopped building markets, we continued with a centralized structure. Changing any of that will actually help, I think, to reduce the distortions in the market.

A Study of Incentives and Mechanisms in the UK

RAJAGOPALAN: Now I want to get into the other paper that you’re working on. This is quite interesting. It’s about water audits. Now, this is not in India, it’s done in the UK. You look at what is the effectiveness or the impact of conducting water audits as opposed to other kinds of mechanisms that one can think of to encourage water saving or water conservation and so on.

I’ll let you describe the paper, and then I want to talk to you about as an economist, my prior is that the price mechanism is fantastic at allocation. It may have lots of other flaws with it that people like to keep pointing out, but the one flaw it doesn’t have, especially in a market as thick as water is allocation. It rationalizes people’s use extremely well. I want to talk about how that system stacks up against other kinds of systems like audits or PR campaigns and social campaigns and so on. Let me first ask you about the paper itself.

KOCHHAR: This paper, it’s with Robert Metcalfe and Robert Han. What we do in this paper is that there was a field experiment that was done in the UK. We did this with Northumbrian Water Limited, this is a water utility in the north of the UK, and we wanted to see that can we incentivize people to join a water audit program? This water audit program was then meant to conserve water. The whole thing is because UK actually has a climate act that wants utilities to take actions to reduce water usage.

We designed seven different interventions, and most of these interventions were basically, we give you a letter and we ask you that, oh, can you please download this audit, which will help you save water? The audits were basically looking at different behavioral aspects. One of them was basically comparing your water usage to your neighbor. The other one was giving you a financial incentive. We also changed the size of the financial incentive. We give you £10 versus they give you £15. We also try to, for example, evoke environmental concerns in you. There’s an environmental aspect to it.

What we tend to find, it’s not surprising honestly. It’s basically that people tend to take up the audit when we give them money. Evoking the moral cost or altruism is good, but it’s not as effective as me giving you £10 or £15 to download the audit. Then we use IV strategy to basically show that it reduces water consumption. Just downloading the audit reduces water consumption by 17%, or actually, it’s somewhere close to 37 to 40 liters per day. This is the UK, so the household water consumption is not as high. It’s around 250 liters per day per household. It’s actually 17% to 20% reduction, which was pretty cool.

Then we also wanted to look at that, whether it is cost-effective from the government’s point of view to do that. We use a marginal value of public funds approach. Now, what this approach basically is this, how much is the government getting? How much bang for buck is the government getting? If government actually spends $1 on a particular project, how much benefit is it getting from that program? What we find is that these programs actually are not that effective for the government.

Specifically, the reason is that even though the water consumption reduces, the greenhouse gas benefits from reducing that water consumption are not as high as the costs of implementing this program. The costs include, for example, me sending you letters, me giving you financial incentives, me also spending time, filling out surveys, filling out audits, letting people come into my house to audit me. Stuff like that, I think that takes up a lot of cost, and the benefits are not that high. Now, coming to your question of price mechanisms, we know prices work, it’s just impossible to use that mechanism in a randomized control trial because first of all—

RAJAGOPALAN: I understand. My question is a little bit simpler. Clearly, a financial incentive is working on the audit side. I imagine that these people are rational individuals who respond to incentives, and if they just priced the water appropriately, then without this enormous machinery of water audits and things like that, you would get a similar effect, except now it’s being managed by pricing water better. 

KOCHHAR: I’ll give you a funny example. We are actually doing an experiment on the salience of prices in Westminster, Colorado. It’s a small town, and the thing is that recently they had a mayor recalled simply because he tried to increase the prices. The thing was that prices are very, very political issue, especially with relation to, let’s say, electricity and water utilities. When you try to increase prices, or let’s say charge differential prices, so let’s say I’m trying to do first, second, third-degree price discrimination there, it’s just very impossible to implement politically.

I think in a democracy, it’s very difficult to use prices as an instrument, specifically, and target individuals based on their water usage. Of course, you can charge higher usage people more money, but then, how do you identify those people? We are trying to run a few experiments there, but not with prices. It’s just very difficult to run these prices. You’re absolutely right. Prices are the best way. We are not following the first best, we are following the second best, simply because the first best can just not be used right now.

RAJAGOPALAN: You are finding that the second best is actually extremely costly. It’s got all these implementation problems. Frankly, the other thing I always fear with audits is if they will lose their long-run effectiveness. In the short run, things like PR campaigns, audits, all these interventions have effects. They usually disappear in the long run, whereas what we find with prices is the opposite. Prices actually strengthen and become more robust as allocation mechanisms over the long run relative to even the short run. Well, lots of people will grumble and write letters and try to make it go away and all of that stuff.

I know exactly what you’re saying about the political side of it and being difficult, but when I read papers as an economist, my first prior is, why aren’t we just doing the simple thing that we know works really well? Do you have that a reaction when you study these markets?

KOCHHAR: I always have a reaction about that, but then I also realize that in a democracy, it’s just not politically feasible to change prices and increase it. I’ll give you an example. I also have a paper on floods and the impact of floods on night lights. We look at night lights, we’re trying to understand how economic activity changes. One of the things that we looked at was the dynamics of recovery. For example, if night lights fall after a flood, how long do they take to come back? Now, again, this is by no means, a policy recommendation, but what we find is that authoritative regimes tend to do better when it comes to that because it’s just the case that there is centralization.

With all these water utilities, with all these electric utilities, there is always this case of, how can you control the externalities? How can you price-in the externalities? Authoritative regimes can do that better simply because there is centralization. Similarly, they can actually charge higher prices to high consumers because they don’t have these political considerations to think of.

RAJAGOPALAN: When you say authoritarian regimes, I just want to understand, what degree of authoritarian are we talking? Because when I look at the map of North Korea, there are no night lights. What is the degree of authoritarianism that is at play here?

KOCHHAR: There is this dataset that we use that goes from minus 10 to plus 10. Minus 10 representing extreme authoritarian regimes plus—

RAJAGOPALAN: Like for example, what are the countries we’re talking about? What are the regions we’re talking about?

KOCHHAR: When it comes to authoritarian regimes, most of them are actually going to be in Africa, North Korea, China. However, interestingly, we remove China, and then we still find the same effects. It’s not just China that is driving these results. It’s also the case that authoritative regimes tend to do better, also because I think they are scared of the fact that there’ll be upheaval, which could throw them off power. Again, it’s a mechanism we are not sure of yet. It’s just something in effect that we find that authoritative regimes the recovery dynamics are better.

RAJAGOPALAN: I have a follow-up question on that. Is it that authoritarian regimes have less of a budget constraint for these utility companies and can more easily pass it on to the taxpayer and so on, whereas democratic or more accountable regimes will actually need these utility companies to somewhat be fiscally viable? Is that what’s driving the result as opposed to the authoritarianism?

KOCHHAR: That is potentially possible. Like I said, this is a very macro-based argument that I gave you, authoritarian regime, but there could be a lot of mechanisms behind that. One of them, like you said, could simply be that I can pass on the cost to the taxpayer versus in a democracy it’s very difficult for me to do that. It’s just not feasibly possible for me to actually increase the price—

RAJAGOPALAN: You just control for loss-making or the extent of the loss-making of the utility company and then go from there. That’ll give you some sense of what is really at play here. This runs counter to the intuition that, for instance, Besley and Burgess find. Which is, in democratic countries with the transparent media and so on, you’re less likely to have famines. You have more responsive government towards these kinds of big shocks. It’s a counterintuitive result. That’s interesting to me what you’re finding.

KOCHHAR: Also, one of my friends, Eunjee Kwon who’s at the University of Cincinnati, what she finds is that it matters a lot when the shock occurs. Basically, let’s say there is a flood, and it occurs right before the elections are going to happen. You’ll basically see that the recovery dynamics are much faster as compared to, let’s say, the shock occurs after the election is over and the person is already in power.

RAJAGOPALAN: That’s the reason I have a sense that this might have less to do with democracy versus authoritarianism versus some other set of incentives and ability to pass on costs to some other players and so on. It’s certainly very interesting. I would love to read this.

KOCHHAR: There’s a lot of potential for the political economy of adaptation and disasters is, I think, something that’s very, very interesting. Also, of just carbon mitigation or even mitigation of excess water use. All these things matter a lot.

RAJAGOPALAN: No, this is fascinating. Thank you so much. For talking about your research and for sharing all this work. This was a pleasure.

KOCHHAR: Thank you so much, Shruti. I really enjoyed the conversation. Thank you.