Ed Dolan's Econ Blog

How Is Inequality Linked to Climate Change, and What to Do About It?

Progressives see climate change and economic inequality as two of the big problems of our time. As the global aid organization Oxfam points out in a recent media briefing paper, “Extreme Carbon Inequality,” the two are “inextricably linked.” But just what is the nature of the linkage? Does inequality cause climate change? Does climate change cause inequality? Is there an inherent tradeoff between mitigation of climate change and reduction in global inequality, or is there a way to address both problems at once? These questions deserve a closer look.

Are the rich responsible for climate change?

The principal message of the Oxfam study is that that the rich are disproportionately responsible for climate change. As evidence, it supplies the following chart showing “lifestyle carbon emissions” by income class of global population. The report defines lifestyle emissions as those that arise from consumption of goods and services, with emissions from producing those goods attributed to the country in which consumption takes place, even if they are produced elsewhere.

P151214-1The chart indicates that the poorest half of the global population is responsible for only 10 percent of total global emissions while nearly 50 percent can be attributed to the wealthiest 10 percent. The rich have average carbon footprints 11 times as high as the poorest half of the population, and 60 times as high as the poorest 10 percent.

I have no trouble with the proposition that wealthy consumers contribute more than proportionately to climate change, but to be fair, the Oxfam chart exaggerates that tendency, and in more than one way.

One problem is that “lifestyle emissions,” as defined by Oxfam, understate the role of agriculture in climate change. The Oxfam chart presumably captures emissions from food produced in the developing world and exported to wealthy countries, but most food produced in low-income countries is consumed locally, or exported from one developing country to another (for example, Latin American exports of soy products to China).

As detailed in a report from the respected environmental organization Worldwatch Institute, agriculture is responsible for a substantial share of climate change—two-thirds as much as transportation, and more than a third as much as the burning of fossil fuels for heat and energy production. According to Worldwatch, agricultural emissions are heavily concentrated in lower- and middle-income countries:

Emissions from manure on pasture in Asia, Africa, and South America together account for as much as 81 percent of global emissions from this source. These emissions from those three regions increased 42 percent on average between 1990 and 2010, reflecting an increase in range-based livestock populations; elsewhere, these emissions either decreased or stagnated.

Carbon dioxide emissions from cultivated organic soils account for some 14 percent of total agricultural emissions, with Asia contributing 54 percent of these emissions. Deforestation and clearing for agricultural land in many tropical South and Southeast Asian countries are a leading cause of these emissions. Asia is home to four out of the top five countries with the highest CO2 emissions from cultivated organic soils, with Indonesia contributing 279 million tons, Papua New Guinea 41 million tons, Malaysia 35 million tons, and Bangladesh 31 million tons.

A second limitation of the Oxfam chart is that it neglects sources of climate change other than CO2. Black carbon (BC), or soot, in everyday parlance, is an important example. According to a briefing from the Institute for Governance and Sustainable Development, black carbon is the second-largest contributor to climate change after carbon dioxide. The briefing gives this account of the geographic distribution of BC emissions:

Developed countries were once the primary source of BC emissions, but this began to change in the 1950’s with the adoption of pollution control technologies in those countries. Whereas the U.S. emits about 21% of the world’s CO2 , it emits 6.1% of the world’s soot . . .

Today, the majority of BC emissions are from developing countries, and this trend is expected to increase. The largest sources of BC are Asia, Latin America, and Africa. China and India account for 25-35% of global BC emissions. BC emissions from China doubled from 2000 to 2006.

Open burning of biomass; burning of wood and coal in traditional stoves for cooking and home heating; and use of low-tech diesel engines in transportation and industry are the biggest sources of black carbon emissions in developing countries.

Third, as the Oxfam briefing itself notes, consumer choice is not the sole determinant of the global pattern of emissions. Government policies, too, play a role:

While the richest citizens can and should contribute as individuals to cutting their own emissions through lifestyle changes, wherever they live, it would be wrong to conclude that they are solely responsible for solving the climate crisis, not least because their choices are often constrained by the decisions of their governments and the market forces they shape.

Government policies vary among wealthy countries. Some make an effort to shape market forces in a way that restrains carbon emissions, as the EU does with its carbon trading mechanism and many countries do with heavy taxation of motor fuels. Other governments, including the United States, prefer a command-and-control approach to carbon emissions while maintaining low energy prices that undermine those efforts at least in part. Even in the US, however, carbon emissions per dollar of GDP have fallen steadily.

The role of government policy is more complicated for the substantial share of lifestyle emissions that arise from consumption of manufactured goods exported to the developed world by China and other developing countries. Although the policies of developed countries are responsible to the extent that they encourage global trade, the technologies and market forces that shape the carbon footprint of production processes are largely the responsibility of the governments in the countries of origin.

Unfortunately, the governments of China, India, and many other third-world countries have consistently placed GDP growth above the environment. They have not been content with the natural comparative advantage that abundant labor and low wages give them. Instead, they have further reduced manufacturing costs by condoning low-tech, inefficient, and dirty production methods.

Even if, say, steel produced by a Chinese or Indian mill is exported to Europe or North America, the government of the country of origin cannot escape responsibility for the resulting pollution, or at least for the amount by which Chinese or Indian emissions per ton of steel exceed those from the cleaner mills of industrially advanced countries.

In short, although wealthy countries undoubtedly do contribute disproportionately to climate change, Oxfam’s focus on “lifestyle emissions” exaggerates the tendency. The Martini glass has a thicker stem and a deeper bowl than shown in the Oxfam chart. Overstating an already strong case only weakens it.

What is the direction of causation between inequality and climate change?

The Oxfam briefing not only places the responsibility for climate change disproportionately on wealthy countries; it also identifies climate change as a cause of economic inequality. It characterizes climate change as “a crisis that is driven by the greenhouse gas emissions of the ‘haves’ that hits the ‘have-nots’ the hardest.” This contention is supported by studies that suggest that poorer countries are likely to be more exposed to sea level rise and extreme weather events, and, at the same time, have fewer resources with which to adapt.

Some people have read the Oxfam briefing differently, however. For example, the website Common Dreams ran a commentary on it under the headline “Extreme Inequality Fueling Climate Change.” That wording seems to reverse cause and effect. In what sense could we construe inequality as a cause of climate change?

To answer that question, we need to know not just that carbon emissions rise when income rises, but by how much. If carbon emissions rise in exact proportion to income, then no matter how income is distributed, emissions don’t change. The technical term for that condition is unit income elasticity. If emissions increase more than in proportion to income (income elasticity greater than one), then more inequality means more emissions. If they rise less than in proportion to income (income elasticity less than one), then inequality decreases emissions.

As we learn from a technical report that accompanies their briefing paper, Oxfam researchers assume unit income elasticity, that is, they assume that each 1 percent increase in income is associated with a 1 percent increase in emissions. (They provide equations in which the elasticity variable can be changed by researchers who want to recalculate their headline results, but no alternative results are mentioned in the briefing paper itself, which is all that 99 percent of the public will see.)

The technical report cites some sources, but following up on them, I found that the rather large literature on the topic turns out to offer scant support for the unit elasticity assumption. Even the paper by Shoibal Chakravarty and associates that the Oxfam technical report cites in defense of its approach turns out to treat unit elasticity more as an upper bound than as the most likely value. It gives a likely range from 0.7 to 1.0. A thorough survey of the literature by Brant Liddle indicates an even broader range of estimates, very few of which are as high as 1.0.

What is more, Liddle’s research shows that the income elasticity of emissions decreases as income increases. He finds that the elasticity is significantly less than one for high-income OECD countries, but not for lower-income, non-OECD countries. Another nice bit of research, by Martin Ravallion of the World Bank, Mark Heilt of the EPA, and Jyotsna Jalant of the Indian Statistical Institute, explains why that is important.

Ravallion et al. agree that the income elasticity of emissions is less than one and decreases as income increases. Those findings, they note, imply that a simple redistribution of income from rich countries to poor ones would increase total carbon emissions. Specifically, they estimate that for each 1 percent of GDP transferred from the richest five countries to the poorest five countries, total emissions would increase by 0.49 percent.

That finding assumes that the income transfer has no impact on the relative internal distribution within either the rich or the poor countries. Any effort to design the transfer in a way that reduced in-country inequalities would make the tradeoff even worse. For example, if the donor countries taxed their wealthiest citizens and gave the money to the poorest of the poor in the recipient countries, total emissions would increase by considerably more than 0.49 percent for each 1 percent of income transferred.

The conclusion is inescapable: Inequality per se is not fueling climate change. The Oxfam study is right not to claim that it does. Quite the contrary: Reducing inequality with a pure redistribution of income, either within or between countries, would appear certain to increase emissions.

So, is there any way to fight climate change and inequality?

Up to this point, we have examined three propositions:

Proposition 1: Rich countries are disproportionately responsible for climate change. Even allowing for some exaggeration in the Oxfam briefing paper, this proposition appears to be true.

Proposition 2: Ongoing climate change will exacerbate inequality. The argument is that the adverse effects of climate change will be stronger in parts of the world where there are fewer resources to adapt to it. Drilling down seriously on either the science or the economics of this proposition will have to wait for a future post, but it seems plausible enough to accept for the sake of this discussion.

Proposition 3: Inequality causes climate change. This proposition appears to be false. The income elasticity of emissions appears to be substantially below one and to decrease as income increases. If that is the case, then any decrease in inequality either between or within nations, would, other things being equal, accelerate climate change.

All three of these propositions belong to the realm of positive economics. To reach any policy conclusions, we need to combine them with the normative proposition that people who cause harm to others should mitigate the harm when possible, and when not possible, make restitution to victims. Combined with Proposition 1, this implies that wealthy countries, and wealthy individuals wherever they live, should take a leading role in combatting climate change and its adverse consequences.

But, what can they do? Neither of the two most obvious policies would address both climate change and inequality.

  • Efforts by rich countries to reduce their own emissions could slow climate change, but they would not stop it. Even if wealthy countries achieved net zero emissions, doing so would only slow the increase of global inequality, not reverse it.
  • Policies that attack inequality directly through income transfers or conventional forms of foreign aid would not work, either. Other things being equal, any decrease in inequality would increase global net emissions and accelerate climate change.

Fortunately, there does seem to be one kind of policy that would simultaneously combat climate change and inequality, namely, policies that would speed the adoption in poor countries of emission reducing technologies, such as solar energy, improved agricultural practices, or cleaner cookstoves. If they include enough subsidies or financing to reduce the cost of clean energy to users below that of conventional alternatives like coal, such policies can simultaneously raise real incomes in beneficiary countries and reduce the carbon intensity of their GDP. They also satisfy the ethical obligation of the world’s wealthy, who disproportionately contribute to climate change, to compensate victims living low-income countries.

Such programs were very much on the agenda at the COP21 that recently took place in Paris. Article 9 of the final agreement includes the following language:

  1. Developed country Parties shall provide financial resources to assist developing country Parties with respect to both mitigation and adaptation in continuation of their existing obligations under the Convention.
  2. Other Parties are encouraged to provide or continue to provide such support voluntarily.
  3. As part of a global effort, developed country Parties should continue to take the lead in mobilizing climate finance from a wide variety of sources, instruments and channels, noting the significant role of public funds . . . Such mobilization of climate finance should represent a progression beyond previous efforts.
  4. The provision of scaled-up financial resources should aim to achieve a balance between adaptation and mitigation.

Some participants had hoped for more, including binding, quantitative, financing quotas for wealthy countries, but the agreed language is at least a step in the right direction. If the signatories to the COP21 agreement follow through on them, they just might make an impact both on global warming and on global inequality.








3 Responses to “How Is Inequality Linked to Climate Change, and What to Do About It?”

Tim GoreJanuary 14th, 2016 at 7:36 am

Hi Ed, thanks for your analysis of our work on carbon inequality, thought I'd make a few quick responses.

Agriculture emissions are not accounted for in the dataset we use on consumption emissions, from Glen Peters, which as you note only considers CO2, not other GHGs (including methane/nitrous oxide associated with agricultural production), although other carbon emissions associated with the food system (eg in transport, production of fertilisers) are I assume captured. Nor are black carbon. However, to the extent that emissions are exported from Southern to Southern countries, this is captured in the model and reflected in the data.

Certainly we are conscious of the importance of tackling the significant emissions from agriculture (see for example this recent report looking at the "hidden" emissions in the global supply chains of the world's food and beverage industry giants,…. You're right that inclusion of agriculture emissions would affect the results, although I wouldn't want to conjecture on exactly how – it's a complicated subject. We'll be publishing some new data on emissions footprints of major food commodities in March.

You suggest that governments in developing countries can't escape responsibility for emissions produced in their territories, even if the consumption of the end products takes place elsewhere. We agree. We are not suggesting in this paper that the UNFCCC move to a consumer-based accounting approach. But it is important to remember that while countries like India and China are major emitters in global terms, their emissions growth is not primarily being driven by the lifestyles of their citizens. In the often brutal politics of the UN climate negotiations, these are claims which are often used by developed countries as an excuse for inaction.

On the elasticity between emissions and income, you might be interested in this work by Lucas Chancel and Thomas Piketty, which follows an almost identical methodology –… . There you can see the results of various elasticity assumptions (they use 0.9 for their main results, which are remarkably consistent with ours). We both found that altering the elasticity assumption does not change the results much.

Finally, we are clear that these results are by no means perfect or definitive. The methodology has short-comings, which we note in the report and technical brief. But they point to important trends which we stand by, and which I think you also recognise.

Most importantly, we agree with your basic conclusion that the way to tackle inequality within and between countries while driving emissions down is through investment in sustainable renewable energy and sustainable agriculture in developing countries. It is for this reason that much of Oxfam's advocacy work at the UNFCCC for the many years building up to Paris has been focussed on those very finance provisions in the new Agreement you note.

Kind regards, Tim Gore, Oxfam, Extreme Carbon Inequality author

Ed Dolan EdDolanJanuary 14th, 2016 at 9:33 am

Thank you for your input. I greatly appreciate this kind of dialog, and I am glad to find that you, like I, see important areas where we agree.