← The CGD Podcast
The CGD Podcast
Who Pollutes Most? Surprises in a New US Database - Kevin Ummel
00:24:18
Pollution has no respect for party lines. In the US, Republican and Democratic districts may differ in many ways, but when it comes to the carbon emissions heating our planet, the differences are much smaller than you might expect. This is one of the...
Loading summary
Transcript36 lines
- [00:16]
A
Welcome to the Global Prosperity wonkcast. I'm Lawrence MacDonald, vacationing here at the center for Global Development. As some of my loyal listeners may know, I recently left CGD to lead the communications work at the World Resources Institute. I'm delighted to be back for today's wonkcast with Kevin Ummel, talking about a hugely exciting paper that I've been waiting for for a while. It's just been published. Who Pollutes A Household Level Database of America's Greenhouse Gas Footprint. Kevin, welcome to the show.
- [00:50]
B
Hi Lawrence.
- [00:51]
A
I remember when you first told me about this, you believed that you had identified databases that were going to enable you to give the median household emissions down to the level of congressional districts or counties and then look across the whole country and identify some key trends or some simple facts about the nature of who pollutes the most. Is that in fact what you were able to do?
- [01:25]
B
Yeah, I think we can say the answer is yes with some qualifications and reservations. What's new here is that when we think about greenhouse gas emissions that are causing all this climate change causing problems for certainly for developing countries down the road, most often we talk about production based emissions. So how much pollution are the power plants and factories in China producing or in the United States producing and what's really of interest? If you're interested in ethical questions around climate change or you're interested in questions of carbon taxes, which really interests me, then you're much more interested in what we call consumption based emissions. So how much greenhouse gas is my personal consumption producing regardless of where the factory or power plant is located? So this study is estimating those consumption based emissions across the United States. And as you suggest, I've shown it's possible to do that at very high spatial resolution by combining data from a number of different sources.
- [02:36]
A
And when you say consumption, I think a lot of people will think of things like gasoline and the oil or natural gas that heats my house. You've also been able to go much broader than that into things like food, air, miles, consumption that might not happen right in the house, but is related to the income and the lifestyles of the people that are in the database.
- [03:02]
B
Exactly. And I think this is a big source of confusion when we think about our personal emissions in particular, like you said, we tend to focus on what in the economics literature we call direct emissions directly emissions resulting from the consumption of energy. Electricity to light your home, natural gas to cook and heat the home, gasoline in your car. And those are very carbon intensive activities. But as this work shows, those Direct emissions only account on average for about a third of the typical person's consumption based footprint. And the other 2/3 is coming from things like food and air travel, clothing, hotels, all the other types of consumptive activities we engage in in that use fossil fuels in the production process. But we don't usually think about them in that way.
- [03:55]
A
Whenever I think about this idea, which I think I learned from you and David Wheeler of Embodied Carbon, I think of this lovely little cast iron patio table that sits out in the deck in my backyard that came from China and the fact that the mining of that iron ore and then the smelting of the steel or the cast iron in the shipping, all of that produced carbon. And now I've got a lovely patio table at a fairly modest cost. But I'm actually the consumer or the producer if you will. I'm the reason that that carbon got emitted into the atmosphere, right?
- [04:28]
B
We can trace every kilogram of, at least of anthropogenic, I.e. human caused emissions. We can trace it back to a consumptive choice on the part of an individual, a household, or in some cases governments. Interestingly, when we do that, when we do that sort of consumption based accounting, then we find really interesting differences across countries. So in the US as this study shows, the average per person household emissions is about 22 tons a year. If we go to a place like China, we have estimates as low as 4 to 5 tons per year. If we go to India, we're talking less than 2. So looking at it from a consumption based standpoint I think gets us closer to information that is really meaningful, certainly in an ethical way.
- [05:18]
A
I want to give some of the headline numbers here real quick and then go into how you compiled them. We'll take a break and then come back and look in more detail at the findings and finally end up with some discussion of the implications for something that both you and I are preoccupied with, which is a carbon emission fee or tax and dividend approach. But I think the headline numbers will give people a taste of why this paper is potentially so interesting. For people interested in climate policy in the United states, the top 10% of U.S. polluters are responsible for 25% of the country's greenhouse gas footprint. No big surprise there. Higher income people emit more. Again, the highest greenhouse gas footprints are found in America's suburbs. And we'll dive into that a little bit more towards the end of the show. Finally, and maybe most surprising, residents of Republican held congressional districts have slightly higher average greenhouse gas footprints than those in Democratic districts. But the difference is small and I'm gonna let our listeners chew over the implications of that while you and I leave behind these headline numbers for a minute and describe for me and my listeners how you went about creating these estimates. I've read the paper. A lot of it's very technical. Basically it's a mashup of databases is the simple answ. Maybe you can give us a little more insight into the sausage making process.
- [06:42]
B
Sure. So the basic idea here is, as I said, we can trace the production of greenhouse gases back to consumption choices on the part of households. The easiest way to do that is to look at how households spend their money. So there is one particularly large household survey conducted in the US on a rolling basis called the Consumer Expenditure Survey, and that tracks how households across America are spending their money. How much are they spending on clothing, how much are they spending on beef and rice and all the other things we consume. And another large household survey that I brought into this is the American Community Survey, which is run by the Census Bureau. And it is a massive survey. Over a five year period, about 6 million households participate, about 5% of the entire population. And that database includes all sorts of information about household demographics, the size of people's homes, where they live, what type of fuels they use, how they heat their homes. And by combining these two data sets in a really, I think, pretty novel way, and bringing in other data from the environmental sciences, life cycle analyses, et cetera, were able to take how people spend their money and translate that into an estimate of how much pollution they are producing.
- [08:03]
A
Is this the sort of thing where you put it all together and send it up to the cloud and come back three days later and it's been churning away at the numbers. How many computations are involved in a problem like this?
- [08:16]
B
A lot. I didn't have to use the cloud, but there were some. Some of the model fitting procedures did require me to have all four cores on my computer running. We're going to get really nerdy here. All four cores running basically for a 24 hour period.
- [08:31]
A
I feel like we should step aside here for a minute and tell people that this is not the first time you've been involved in a massive data exercise. The Carbon Monitoring for Action database at the center for Global Development has the estimated emissions of all 60,000 power plants in the world. And however many firms, 6,000 firms that own them and their locations. Kevin, is that right?
- [08:53]
B
Yeah, even more than that, there's 15 or 20,000 firms. I think there's been a trend in my research, which is find pieces of data that people haven't thought to put together before and put them together and see what new information you can generate.
- [09:05]
A
This sounds a lot like David Wheeler, a mentor to both of us.
- [09:09]
B
Yeah, that makes sense.
- [09:12]
A
So why don't we take the break then? And when we come back, I want to dive into some of these findings and then invite you to go out on a limb a little bit in terms of the potential political implications. This is the Global Prosperity Onecast. From the center for Global development, I'm Lawrence MacDonald. My guest today is Kevin Ummel. We're discussing his new paper, who Pollutes A Household Level Database of America's Greenhouse Gas Footprint. We'll be back in a moment. Welcome back to the Global Prosperity wonkcast. I'm Lawrence MacDonald. I'm speaking with Kevin Ummel. He is the author of a new center for Global Development paper, who Pollutes A Household Level Database of America's Greenhouse Gas Footprint. Kevin, you explained how you put these estimates together. Give me a sense of some of the most surprising findings that came out of this work.
- [10:34]
B
Well, as you mentioned earlier, we see this considerable inequality in greenhouse gas emissions in the paper. I call it pollution inequality to sort of juxtapose it with income inequality. Now, the inequality of pollution in the US Is not as great as income inequality, but it's still pretty significant. So you mentioned top 10% of US polluters responsible for about 25% of the entire footprint of the country. The flip side of that is that the least polluting 40% of the population, largely people in lower income groups, are responsible for only 20% of the total. So there's a significant inequality driven by income. There's also a lot of differences driven by geography. Much of the research, previous research, has focused on different regions in the country. The idea being that differences in climate or the nature of the electricity supply would drive these differences across space. But I find that really the more significant geographic differences in terms of per person footprints has to do with population density. So are you living in the suburbs where we see the highest footprints, or are you living in a rural area or in a high density urban area which tend to have lower footprints?
- [11:57]
A
I remember early on in this work that you had identified two of the suburban counties of Washington, D.C. montgomery county and Fairfax county as being among the dirtiest in the country in terms of carbon emissions. Did that turn out to be true?
- [12:15]
B
Well, I'm imagining that at least some of our listeners are located and live in those counties. But yes, that's true. The paper doesn't go into revealing specific results for counties and zip codes because I didn't want that to be the focus of the paper. But yeah, it is true. There's a map in the paper that allows you to look with some detail and you will see in the Washington D.C. area, a core of low footprints which are predominantly the high density and oftentimes lower income communities of Washington D.C. and surrounding that, where the much wealthier classes live in those counties you mentioned are some of the deepest red, not politically red just means high footprints. In the map, those counties, high income, relatively sparsely populated suburban communities, lots of gasoline, lots of heating at large homes, really, really big footprints.
- [13:12]
A
So these are the big houses, big cars, long commutes, big yards, lots of international travel, and probably also the people who want to reduce the number of food miles that their food has traveled because they're concerned about climate change and they're trying to reduce their carbon footprint by shopping at the local farmer's market.
- [13:36]
B
Well, so that brings up a really interesting point which goes to this political difference. So that group that you're describing, David Brooks, the New York Times, has sort of called this group the bourgeois bohemians or the Bobos. You and I are both cultural Bobos, Lawrence, I know you well enough. Bobos are this class of, of relatively wealthy, well educated, basically elites whose lifestyles attempt this very tenuous marriage of sort of social and environmental liberalism with relatively high consumption, high income lifestyles. And if you look in the paper, there's a really interesting graph that shows all 435 congressional districts, the average footprint in each of them and their political affiliation. And when you look at the Democratic districts, you see this incredible spread. You see some Democratic districts have very high footprints and some have very low footprints. And what you're seeing there is this incredible diversity in the Democratic Party. We have these low income, urban core communities with very low footprints. And then there are the Bobo communities, which are exactly the people you're describing who care about climate change, care about where their food comes from, but by nature of having high education, high income, they also consume a lot and have very, very high footprint. So there's a bit of a scale schizophrenia within the Democratic Party when it comes to this issue.
- [14:59]
A
And the red districts tend to be more tightly clustered, not consuming significantly more on average than the blue districts, but somewhat less diversity.
- [15:11]
B
Yeah, I think Republican districts, certainly in terms of their footprint, probably in terms of other factors as well, tend to be more homogeneous, less range in those across those districts.
- [15:23]
A
Let's Shift. Then we've sort of begun to talk about the politics. In looking at the carbon emissions of the Democratic and Republican districts, what do you see as the biggest implication of your work for the potential of a change in the politics that would lead us to putting a price on carbon in the United States?
- [15:46]
B
That's a great question. I think one of the most important findings in that respect is this very, very small difference between Republican and Democratic congressional districts. Voters, people living in those districts are basically polluting at similar rates. There's variation, but on average they're similar. What this means is that if we were to put a carbon tax in place to a large degree, both Republicans and Democrats would be exposed to the cost of that tax roughly equal. And what this means is that if we were to put a tax in place, it is not the case right off the bat that members of one party would be disadvantaged relative to another.
- [16:29]
A
Then how do we explain this incredible political divergence? I mean, I think this is a fascinating, it's even a shocking finding because if you look on the surface of American politics, one party is, you know, either in denial about climate change or certainly opposed to any efforts to do anything about it. And it's easy to think, oh, well, these are the coal belt states. They're heavily dependent on coal based energy and so their interests are not well aligned with action on climate. But you're telling me that at the household level, the emissions is not so different.
- [17:06]
B
Yeah, and I say in the paper, you know, the political rhetoric, the difference in political rhetoric is far greater than the difference in the environmental reality here. And I think one reason that we have, despite this, that we haven't seen political traction on this issue. I actually think the research community is partly to blame here, which is to say that we have spent a lot of time researching and publishing and creating findings about the macroeconomic effects of the carbon tax. What does it do to gdp? What does it do to employment, what does it do to efficiency? But we have not spent nearly as much, paid nearly as much attention to what is the effect in particular areas on particular types of people in particular districts. And in other words, if we had a way of putting together a policy package, a combination of a carbon tax and a group of tax cuts, for example, and we could show to members of Congress, look, this is the policy package. Here's what it means for your district. 70% of households in your district are going to see an immediate financial benefit under this policy. Well, that sort of policy, it's possible for someone in a very red district to get behind potentially if we're showing an immediate financial benefit for voters in that district, but we have not done that yet. This database though introduces the data that we would need to do that type of analysis.
- [18:29]
A
You're finding that the high income groups pollute a lot more than the lower income groups also suggests to me the possibility of some cross subsidies. Because if energy prices go up for the higher income people, it's a relatively modest dent to their overall income even though they're consuming more energy and emitting more. Whereas for low income people, even though they're emitting less, it's presumably a bigger hit to their total household budget. So one could imagine a tax and rebate scheme or a fee and rebate scheme that would be progressive in that more of the revenue that is rebated could go to the lower income households to ensure that they come out ahead at the end of the day. And the higher income households, frankly, they're doing well enough, they're a small enough piece of the electorate that if and many of them live in these liberal districts like Fairfax and Montgomery, where they're eager for action on climate change, presumably they could be persuaded to go along with a certain amount of cross subsidization to poor households. Am I imagining this?
- [19:40]
B
No, frankly, I think this is the holy grail of carbon tax policy and politics. Right. I describe to me the politics of carbon taxes is like the politics of the Israeli Palestinian conflict. Rational people know where this is going to end up. It's going to end up with a two state solution, Israel and Palestine. Carbon taxes are going to end up with some sort of a carbon tax swap where we implement a carbon tax and we reduce labor capital taxes. On the other hand, I mean this makes complete sense. Democrats should support it for the reasons you mentioned in that it can be made relatively progressive. Republicans should support it because it is eliminating free riding, eliminating the freeloaders who are polluting and not paying the costs. It's not a tax on success, as Republicans used like to talk about higher taxes on high income individuals. This is not a tax on success, this is a tax on polluters.
- [20:34]
A
And in fact you could roll back some of those taxes on success. You could cut the taxes on income, on the Social Security, the self employment tax. There's quite a number, there's a long list of taxes that many of us would like to see reduced.
- [20:49]
B
Exactly. And the rhetoric should be why are we taxing things we like, like income and not taxing things that all of us would agree we want less of like pollution. So there's very good practical reasons to do this. It's the mechanics of the politics that's difficult. And I think if we had the research in front of us to, like I said, go to members of Congress and show them, look, here's the policy package, here's the proposal, here's the consequences. That would be two things. One, people who are inclined to support this issue but feel politically hemmed in, I think a lot of moderate Republicans would be in that group would suddenly have the data behind them to go out and say, I support this policy not because I care about the environment, which might not be a winner in their district, but because of what it will do for households. And the second thing it would do is it would provide ammunition to opponents. So in competitive districts, for example, if I'm a Republican and there's a Democrat who's running against me and that Democrat is saying we need to do a carbon tax policy because it's going to benefit households, it's going to be difficult even for the deepest red Republican to oppose that policy under those circumstances.
- [22:02]
A
Kevin, this is wonderful work. I want to ride on your coattails a little bit here. I know you've read my piece because you were generous in your comments. I will soon be publishing here the Sudden Rise of carbon taxes 2010-2030 a future history. And for those who are interested in how Kevin's wonderful research could feed into a broader political scenario about the rise of carbon taxes, I have, together with Jing Tsao, a non resident fellow here at cgd, imagined a world in which, as you say, Kevin, we, we all know where this is going to come out. The question is, can it come out there soon enough to make a difference? And I've tried to imagine a world in which it would. So, so much for my self promotional advertisement. Kevin, I love this piece. I think it deserves to be very widely read and I'm sure it's going to be. Thank you for doing it and congratulations.
- [22:58]
B
Thanks Flores. It's always good talking with you.
- [23:00]
A
This has been the Global Prosperity Wonkast from the center for Global Development. My guest today was Kevin Ummel and, and we've been discussing his wonderful new paper who Pollutes A Household Level Database of America's Greenhouse Gas Footprint. It's available now on the center for Global Development website. I think anybody who is interested in climate policy and in particular in breaking the deadlock on US Climate policy should read it right away, tweet it, blog it, Facebook it, send it to your friends. It's a marvelous piece of work with I think potentially big political implications. You can find the Wonkast online on itunes and on Stitcher. Just search for wonkcast or CGD and sign up to hear a new interview every week. Until next time, I'm Lawrence MacDonald. Thanks for listening.
- [24:03]
B
Sam.