Oct 24, 2024
Intro: Assistant Professor Tan Soo Jie Sheng primarily uses micro econometrics techniques combined with economic modelling to conduct empirical research in the areas of environment, health, and development. Recently, he and his co-authors published an article titled, “Using Cost–Benefit Analyses to Identify Key Opportunities in Demand-Side Mitigation.”
Their study examines how different environmentally-friendly behaviours, such as switching to public transportation or using reusable bags, impact people's well-being in Beijing, China. They assess both the financial and non-financial effects of these behaviours using cost-benefit analysis. The findings highlight which actions are beneficial or detrimental to individuals and society and identify opportunities for targeted policy interventions.
David Austin: Thank you so much for joining us, Dr. Tan. Could you start by just giving me a little bit of background of what question were you trying to answer when you began this study, when you were formulating it?
Tan Soo Jie Sheng: Thank you David. I guess basically came down to the observation that, there are a lot of things that, individuals can do to reduce their carbon footprint. But then the question is, why is it that a lot of people are not doing them? Is it lack knowledge or is it, some other reasons? As an economist, my hypothesis is that it could be that a lot of these behaviours are detrimental to people's welfare because it actually costs resources to change behaviours.
And I wanted to, study or, examine if this is indeed the case, that by adopting this kind of low carbon lifestyle, it would actually reduce people's welfare? So that was the question that I was trying to answer.
David Austin: Is that why you decided to choose a cost benefit analysis as your primary methodology?
Tan Soo Jie Sheng: Because I'm trained as an economist I've been trained to think in terms of monetary terms. And so, in that sense cost benefit analysis seems to be the most natural approach to take in this case.
David Austin: Can you tell us what does demand side mitigation actually mean? And what motivated you to study demand side mitigation, and why specifically in Beijing?
Tan Soo Jie Sheng: Demand side mitigation is actually an often overlooked part of climate mitigation.
For a long time, policies have been targeted at firms, at factories, trying to get producers to change their behaviours, or to change their production techniques, or to change the kind of energy that they use from high carbon fossil fuel type to renewables. But demand side focuses more on the consumer's part, more on the individual's part because a lot of our behaviours actually emits greenhouse gas directly or indirectly. So, demand side mitigation, basically focuses on what consumers, what individuals can do to change their behaviours so that we can reduce our carbon footprint.
And the idea behind this research actually came about when I was having a conversation with an NUS colleague. He's an ecologist. So you know, he was telling me, demand side mitigation has been studied quite extensively in the ecological or biological space. But what they did back then was mostly focused on the emissions mitigation potential, meaning, how effective is this particular behaviour in reducing your footprint?
There was this really famous study back, I think published in 2017, and one of their headline findings were: to really reduce our carbon footprint, the most effective thing you can do is to have one less kid, is to have one less child. So, when this colleague told me about this finding, back in my head I was thinking, I really don't think people are not having kids because they care about their carbon footprint. It could be because it's more welfare enhancing for them to not have kids. So that's got me started thinking, perhaps a lot of this demand side mitigation behaviours, what it boils down to is really about our welfare or how our welfare will be affected before we decide whether we want to engage in that.
And so that's what, let me down this rabbit hole of this study.
David Austin: And then you took 12 different behaviours, right? 12 different things that could be anything from switching from private vehicles to public transportation, maybe switching from a combustion engine vehicle to a battery electric vehicle, or also things like just turning off the lights in every room, switching from an animal protein diet to tofu or to a plant-based meat, or even things like switching from plastic bags to reusable bags for shopping. And then you looked at what you call the pecuniary components and the non-pecuniary components.
And the pecuniary are the, basically the economic part.
Tan Soo Jie Sheng: Yes. The monetary part.
David Austin: Then for the non-pecuniary components you talk about things like time, psychological, discomfort and attention costs. Can you explain a little bit about how you conceptualised those, and then how did you put a monetary value to them? Give us a little bit of a framework of how you viewed those costs.
Tan Soo Jie Sheng: Absolutely. Now the one thing about cost benefit analysis is that everything needs to be expressed in a common unit. And the unit we use is the dollar sign. A lot of costs, a lot of benefits have dollar signs, but a lot of other costs and other benefits do not have dollar signs, but they are still important, so this is what we term as non-pecuniary.
Time is one of them because in some of these behaviours, for instance, switching from private to public transport, sure, it will save you a lot of money. You don't have to own a car anymore, you don't have to pay for fuel or parking or whatnot. Obviously, the cost of public transportation, at least monetary wise is much, much smaller than private transportation. But the trade-off is, you're going to spend a lot of time travelling. And this time is a cost, it's a resource, right, for the individual.
So, then the question is how do we value something like time? Fortunately, in environmental economics, we have actually devised a lot of methods over the years to monetise this seemingly impossible-to-monetise attributes. So, for instance, time we'll use things like know wage rate. Time equals Money. How much money you can make if you are productive instead of spending time on, on the public transport. Wage rate is a very natural substitute to value time.
Some other things that's more difficult to value, for example, attention costs. Say the attention costs of turning off the lights when you leave a room. Not a lot of intuitive substitutes. So, in that particular instance, what we did was we ask ourselves what is the technological substitute for that? So fortunately, there are some devices that will automatically sense whether there's someone in the room, and if there's not, the light will turn off by itself.
So we use the cost of those technological substitutes as a proxy for attention cost. Because the idea is if people absolutely do not want to pay attention to that, then they can always use those technological substitutes to adopt this behaviour as well. So essentially, it's a very creative mix of trying to find appropriate substitutes that can cater to these non-pecuniary costs or benefits that we are thinking about.
David Austin: And what about discomfort? What did you use for that? You must have been very creative for that one.
Tan Soo Jie Sheng: I think that was mostly for public transportation. The idea of switching from private to public transportation has a pretty long history of research.
So fortunately, we were able to find some studies that were conducted in some other settings. Not in China, not in Beijing, but in some other countries. And then we extrapolate the monetary value that they found to our setting. And how those others got their monetary value were basically simply by running a survey, by asking people, “Hey, how much are you willing to accept if I were to switch you from a private transportation to public transportation?”
So, this is what in the literature we call it their stated preference. So essentially, if there's no creative substitutes, we just ask people directly the question, how much are you willing to pay, or how much are you willing to accept for this trade off?
David Austin: I see. And then for everyone, there's some point where the discomfort becomes too great for them to accept, but that differs from person to person, is that right?
Tan Soo Jie Sheng: Exactly, so it’s about getting the average, yeah.
David Austin: Speaking of survey, part of this study included a survey of Beijing residents. What did that involve and was that challenging in any way? Were people willing to participate and do you think you got good responses from the residents?
Tan Soo Jie Sheng: Good question. So interestingly, actually surveys got really easy in recent years. In the past, when we need to conduct surveys, we actually need to go down to the location itself, knock on doors and all that stuff. But now, with internet is so much more convenient. There are a lot of companies who have panels of respondents that's willing to answer surveys for a very small fee. The trade-off is these respondents are often young, highly educated, high income, so they are not very representative of the population that we need.
And in this case, because we need not just respondents from Beijing, but we need respondents to be broadly representative of the population in Beijing. So, it actually took us some time to be able to draw out the respondents that can fit the overall population distribution of Beijing in a better way.
In most internet surveys, we started off with a pretty low incentive in this case, after some time we realised that we are not going to get the population that we want them. And after that, we increase the incentive several more times so that we can finally get the people that we really want to survey to make sure there is overall representative of Beijing. But all in all, I would still say that conducting a survey over the internet is way easier in this case, compared to the past where we had to be there physically.
David Austin: Let's move on to some of the findings then. What do you feel are the key findings of the study in your opinion?
Tan Soo Jie Sheng: The first key finding is that most demand side mitigation measures are detrimental to individuals' welfare.
That's surprising to me. I kind of expected it because, if it wasn't, I wouldn't be conducting this study to begin with. People would already be changing their behaviours pretty easily. So, I had a hunch that would happen.
Tan Soo Jie Sheng: What was surprising to me was that even if we account for the cost of carbon, a lot of behaviours are still detrimental to welfare at a societal level. That part was surprising. What it means is the emissions mitigation potential of a lot of these behaviours are not that great, and because they're not that great, that means that, even if we take a broader view and look at it from a societal perspective, it's still not clear if government should really persist, at least on the broad scale, on demand side mitigation.
David Austin: Okay. I read that part in the report and I just, I wasn't quite sure I understood.
Tan Soo Jie Sheng: Okay. Let me try to break it down. Okay. So, the main gist of this paper is to look at the cost benefit, or the welfare implications from an individual viewpoint. And in economics, we tend to assume that individuals don't really care about environmental outcomes.
And it makes sense from an economics viewpoint because a lot of these environmental outcomes are what we call public goods. And so, from the individual level, there's really no reason for us to care about public goods. And this is especially true in the case of climate change because a lot of these behaviours, what they will result in is a lower level of CO2.
So, CO2 in itself is not a direct pollutant to people. Second, it's not a local pollutant, it's a global pollutant. Or rather a global atmospheric air or whatever you call it. So, in that case the assumption that, individuals do not factor the emissions reduction or mitigation potential into the cost benefit ledger makes sense. But from a societal viewpoint, governments would care about environmental outcomes because that's what governments are designed to do. That's what public institutions are designed to do. They're designed to take into account of all these public goods and bads. When we refer to societal welfare, it means that we are taking the individual welfare plus the benefits of reducing greenhouse gas emissions.
Greenhouse gas emissions have a global price and not a local price. Again, because greenhouse gas is a global pollutant, not a local one.
David Austin: Okay. I think I understand that. Now the study also found that eight out of 12 of the pro climate behaviours reduced individual welfare. What does that really mean to the individual as far as having a lower welfare from these behaviours?
Tan Soo Jie Sheng: When we break it down, we see that actually a lot of these behaviours have this characteristic where adopting these low carbon behaviours would increase welfare at a pecuniary level, but it's the non-pecuniary part that's really driving down the overall welfare implications. So that means that we cannot just look to the monetary savings if we want to justify low carbon behaviours or pro climate behaviours.
We have to look into the non-pecuniary parts, the example I used was switching from private to public transport, you'll save money, but you also incur a lot of discomfort costs, a lot of time costs that will actually drive up the cost of this behaviour.
David Austin: The study identified some behaviours that could be thought of as like low hanging fruits for policy makers. Which are those and what could policy makers do to take advantage of that?
Tan Soo Jie Sheng: So how we define low hanging fruits basically behaviours that has a positive welfare implications at the individual level. And that has high emissions mitigation potential. Okay, so these are what we think governments can more easily focus on because they are already beneficial to welfare at the individual level, then it really shouldn't take that much to convince people to switch their behaviours.
So, the three that we have identified are switching from internal combustion engine to battery electric vehicles, switching from domestic flights to high speed rail, so this contextual to Beijing, and switching from meat or animal protein to tofu.
David Austin: Based on that, what would you recommend to policymakers? What would they do with that information?
Tan Soo Jie Sheng: We are aware that by simply saying, “look, these behaviours are good, focus on them” is not sufficient information for policymaker to act on.
So that's why, in this study, we also incorporated this thing called a tornado analysis. Basically, it's just a fancy way of saying that, since each behaviours have different sets of costs and benefits, meaning each behaviour has a constituents of cost and benefits, we can undertake more analysis to see which of these costs and benefits are most influential in affecting the welfare of these behaviours.
For instance, switching from internal combustion engine to electric vehicle, we found that the price of the vehicles is the most influential in convincing people to switch. So, meaning policies that can drive a wedge between these two sets of prices would have a big impact. Interestingly, we found that energy prices actually do not play as much of a role, despite anecdotes about how the pump prices is causing people to switch to electric vehicles and whatnot.
But this is not what we found in this study. We found that whether it is the electricity price, or gasoline price, it really doesn't affect the overall cost benefit or the overall welfare figures that much. It's really the price of the vehicles. And similarly, by using this tornado analysis we also looked at some of the other factors that affects the other behaviours as well.
David Austin: Do you have any insight into interventions in Beijing by the local authorities that had been tried regarding any of these 12 behaviours?
Tan Soo Jie Sheng: They have done two things.
First is a lot of persuasion, or as we call them, moral-suasion in Singapore's context. So basically, just preaching to the population. And in fact, this is where we got this list of 12 behaviours because, the different branches of local governments in Beijing actually have advocated for these behaviours in different ways.
So, we compiled all these behaviours that they had advocated for. And then, we chose the top 12 that were most frequently mentioned. So moral-suasion is one, basically just telling people about it. Another thing that I think they are doing right now they're currently embarking on is, giving people a chance to, record or to tally up their behaviours in an app, for instance. But this only works for certain behaviours that can be readily collected and verified in an app. For instance, certain brands of electric vehicles are linked to their app, and every time you charge a vehicle, you'll be recorded as you are charging an EV, and then you'll receive some kind of credit from the Beijing government, which you can then exchange for vouchers or discounts or whatnot. In Singapore's context, it's pretty similar to the health app that we have where we get points for steps or activity. So, it's similar to that concept.
David Austin: Yeah. I'm wearing that watch on my wrist right now for the steps. Is there any way that policymakers can address the psychological and time costs associated with pro climate behaviours?
Tan Soo Jie Sheng: So, time cost obviously really depends on the behaviour that we're talking about. Some requires a lot more intervention than the others. Like for instance, China is peculiar in the sense that they have a very extensive network of high speed rail.
So, in that case, to reduce the time taken to do domestic travels within the country is a lot easier in China compared to a place like the United States where, you know, the local rail infrastructure has not been built up since the 1900s. China has already invested a lot of money, so it's much easier for them to reduce the time cost of domestic travels.
Similarly, for changing from private to public transport, that requires another set of interventions for instance, investing more in infrastructure, increasing number of buses or trains or frequencies or allowing them more leeway in terms of giving them right of way during congested rush hours. So, all this will help for time costs, but again, it really depends on the behaviours that we are talking about. Some are more visible than others.
In terms of psychological, it depends on the nature of it. Again, if you're talking about switching from private to public transport, we have to look more deeply into the context of the place. In certain countries they reduce psychological costs of traveling by having gender specific cabins so that will help some people feel more comfortable taking public transport.
And say, switching your diet from animal protein to a plant-based protein diet. So that I would say it takes a lot more time, because there has to be educational base, that we have to start from a very young age if we want to shape people's preferences in that manner.
David Austin: Moving on to some of the dietary changes, and you found that animal protein to tofu was the only behaviour in your study that enhances social welfare after accounting for emissions mitigation. And the study said primarily because tofu is familiar and widely consumed protein in China. Does this have any special relevance to Singapore, or is the dietary culture here just too different?
Tan Soo Jie Sheng: A large portion of population is Chinese and I know we do consume tofu, as part of a regular diet. So, I would say that it wouldn't take too much to convince people, maybe not give up meat totally, but you know, at least substitute a portion of your meat with tofu, and still, get your dietary requirements met. But I would say we can actually generalise or extrapolate this finding to other culture because, that's something that I'm trying to work on now, which is, you see a lot of cultures have traditional plant based protein.
For instance the Malays, they have tempeh, which is a form of tofu, but fermented, which is also a good source of protein. The Indians, they have lentils and chickpeas, which is their source of protein, vegetarian protein. And in the Latin Americas, they eat a lot of beans and rice, which is again, a source of protein.
So, my take is, rather than reinvent the wheel with, the modern plant-based protein that is supposed to mimic meats texture and behaviour but cost a lot more than regular meat, why not we look more closely into what people have been consuming for the past 1,000 or 2,000 years, and try to borrow and try to lean on that?
And so, this is why, in our findings we saw that we were able to switch tofu, to being socially welfare enhancing. And the main reason is because the cost of tofu is way lower than plant-based meat. At an individual level, the detrimental impact on welfare was not that big to begin with because of the price of tofu. And once we include the societal benefits, it's actually not that difficult to switch it over. And I suspect this is the case for other traditional types of plant-based protein.
Yeah. So anyway, to answer your question, yes, I do think that there's a lot of relevance to Singapore not just for the Chinese population, but for the Malays and the Indians as well.
David Austin: Just in general, what was the most surprising finding for you personally as you were going through this process?
Tan Soo Jie Sheng: As I mentioned, the most surprising was definitely that even after we included the cost of carbon, that a lot of these behaviours are still detrimental to society.
And why I say that is because there are many price of carbon that can be used. For instance, in the US, the US EPA, they now use a price of carbon, I don't know, maybe around 70 or 80 USD per tonne. But in our study, we set a very high benchmark. We use a price of 185 USD per tonne of carbon, so this is way higher.
Again, for context, Singapore's carbon tax is around 25 Singapore dollars per tonne, but we're using 185 USD. So initially I thought that with such a high price for carbon, we would find that if we include the price of carbon in all these behaviours, we should see all of them being positive.
At least even if they are negative from an individual's perspective, they should be positive from a societal perspective. But so, the surprising part is even after using such a high price of carbon, we still find that a lot of these behaviours are detrimental to society.
So that means that even if governments want to justify these behaviours from a societal viewpoint, it's not something they should propagate on the very broad level. Which means that we should still take a very targeted approach and look at what are the behaviours that's really beneficial at the individual level and that has high emissions mitigation potential to target rather than focus on all low carbon behaviours under the sun.
David Austin: Looking forward, are there any other behaviours or areas that you believe should be analysed in future studies using a similar cost benefit analysis?
Tan Soo Jie Sheng: For any cities, countries or whatnot that wants to advocate this type of demand mitigation behaviours. My recommendation obviously is to use our method, use our framework before, you know, embarking on this whole propagation campaign . Having said that, I'm also conducting some other studies on trying to better understand the welfare implications of these behaviours across different segments of society.
Because what we had done in this paper is to basically look at it from the perspective of an average Beijing resident. But we would think that different segments of the population, say, low income, high income, different education levels, some of these behaviours may have different welfare implications for different segments of society.
So that's another thing we are interested in looking at in the next project, which is to narrow it deeper to look at the welfare implications on different segments of society and see if we can devise even more targeted interventions for different segments of people.
David Austin: Right, so it's really targeting the right people that's the key, right?
Tan Soo Jie Sheng: The right people and the right behaviours.
David Austin: Key to getting the best outcomes as far as climate mitigation based on these behaviours. So that's really what you're pushing for as far as getting the idea out there to policy makers.
Are there anything else that you see as next steps in research for demand side mitigation? Anything on the horizon that you think would be useful for everyone to know?
Tan Soo Jie Sheng: Yeah. So right now, I'm still trying to look more deeply into the distinction between beliefs and behaviours.
So, the idea is that believing in something may not always lead to a change in behaviour. In the literature, we call this the attitude behaviour gap. So I'm trying to see if this gap has widened or narrowed, along with how, you know, our knowledge of climate change has changed over the years. So, this is another thing that I'm trying to work on, essentially, they all revolve around the same idea of trying to better understand people's behaviour.
David Austin: Very good. Thank you so much for spending time with us and helping explain the study to us.
Tan Soo Jie Sheng: Thank you so much for giving me this platform.