InsightsOctober 13, 2021

New approach to estimating an evidence-based social cost of carbon published in Nature

Study shows how emitting a ton of carbon today drives local changes in climate that impact energy use for populations in 24,378 regions, demonstrating unequal effects of warming

Energy consumption is viewed as one of the most climate-sensitive sectors of the global economy, with temperature considered a key determinant of the type and quantity of energy we consume. However, that relationship looks very different if you are not among the richest 40% of the world’s population.

A new study from the Climate Impact Lab, published today in Nature, reveals a deep divide between wealthy and poor populations when it comes to energy use, a divide that is likely to persist as the planet warms and cooling technologies like air conditioning become increasingly critical. The globally comprehensive analysis of energy consumption is part of the Climate Impact Lab’s effort to estimate how individual households, businesses, and industries around the world will be affected by climate change at a local level, with the goal of producing an improved estimate of the social costs of greenhouse gases for use in climate policy.

The study shows that—while there are vastly unequal effects among regions—globally, on net, emitting one additional ton of carbon dioxide (CO2) today results in a modest reduction in future energy spending, driven mainly by declines in the need for heating through fuels such as coal, natural gas, and oil, and the slow adoption of electricity in today’s developing world. Based on these findings, the study estimates a partial social cost of carbon for energy consumption valued between -$3 and -$1 per ton. This value represents the first peer-reviewed, empirically derived estimate of the global cost of climate change for a single sector of the economy, and the study is the first research article on the social cost of carbon ever published in Nature.

“Applying the Climate Impact Lab’s method of combining global data, big data analytics, and detailed climate models to estimate the costs of climate change—and the benefits of reducing emissions—enables the calculation of a full estimate of the social cost of carbon across many sectors of the global economy,” says Michael Greenstone, the Milton Friedman Distinguished Service Professor in Economics at the University of Chicago, and one of four co-directors of the Climate Impact Lab.

“Updating the social cost of carbon to the frontier of understanding is a key to designing evidence-based climate policies,” added Greenstone, who was a key architect of the social cost of carbon as Chief Economist of President Barack Obama’s Council of Economic Advisors.

The research contradicts previous estimates of the global economic damage caused by CO2 that suggested hotter temperatures would drive massive increases in energy spending by the end of the century. This includes the models that underlie the original 2010 social cost of carbon estimates, which were calibrated with data exclusively from wealthy countries.

To fill this knowledge gap, Climate Impact Lab researchers combined energy use data from 146 countries across four decades with historical weather observations, and then modeled the sensitivity of energy consumption to extreme temperatures. On hot days (average temperature 35°C/95°F) in wealthy locations, they find per-person electricity consumption soars, especially among the richest 10%. For the poorest 60%, extreme heat results in virtually no increase in electricity consumption, although extreme cold days (average temperatures 0°C/32°F) result in an uptick among all but the poorest 20% in consumption of fuels used for heating, such as coal, oil, and natural gas.

“Most prior analysis of the energy-related costs of climate change has focused on data that comes from the wealthiest parts of the world, without including evidence from developing world, where billions of people live in poverty and lack energy access,” says Solomon Hsiang, the Chancellor’s Professor of Public Policy at UC Berkeley, and co-director of the Climate Impact Lab. “The data shows that income plays an essential role in people’s ability to protect themselves from the impacts of warming.”

As income, population growth, and climate change evolve throughout this century, the analysis finds that most of the world will increase its electricity consumption while consuming less of the fuels commonly used for heating due to fewer cold days. Today’s middle-income regions, like parts of India, China, Indonesia, and Mexico, will likely benefit from expanded access to electricity. In contrast, lower-income, hot areas, such as much of sub-Saharan Africa, are projected to remain relatively poor through 2100 and therefore lack the means to protect themselves with cooling during increased heat.

“While some form of air-conditioning is present in 90% of homes in the United States, this is currently true for only 5% of homes in India. As climate change causes heatwaves to become more intense and frequent in future decades, the data shows that electricity to power cooling technology, like fans or air-conditioners, remains out of reach for more than half of the global population,” says Ashwin Rode, Director of Scientific Research at the Energy Policy Institute at the University of Chicago (EPIC) and a member of the Climate Impact Lab.

The Climate Impact Lab’s 2020 study of how climate change will impact human mortality, measuring the effects of warming with the same data-driven, granular approach, revealed similar patterns of vulnerability.

“Combined with our research on human health, we can map how the future impacts of climate change disproportionately fall on regions that are poorest and hottest today, exacerbating existing inequality,” says Trevor Houser, a partner at Rhodium Group and co-director of the Climate Impact Lab.

Notably, the Climate Impact Lab uses state-of-the-art statistical methods to produce a full range of estimates for 24,378 distinct regions. These include five projections for future income and population growth, two trajectories of future greenhouse gas emissions, eight scenarios of future energy prices, and simulations from 33 climate models, allowing for an assessment of the range of possible risk surrounding any particular projection.

“Every additional ton of greenhouse gas causes a measurable impact on society that can be quantified using this bottom-up, empirical approach,” says Robert Kopp, Professor of Earth and Planetary Sciences, Director of the Megalopolitan Coastal Transformation Hub, co-Director of the University Office of Climate Action, Rutgers University and co-director of the Climate Impact Lab. “Because emissions today stay in the atmosphere for hundreds of years, improving our understanding of local climate damages will be essential to taking the action we need to prepare for future risks.”

A more granular view of the findings is available at The Lab’s Impact Map shows estimated energy costs for three future time periods under two emissions scenarios, country-by-country around the globe and at the state level for the U.S.


The Climate Impact Lab is a collaboration of more than 30 researchers from the Global Policy Laboratory at the University of California, Berkeley, the Energy Policy Institute at the University of Chicago (EPIC), Rhodium Group, Rutgers Institute of Earth, Ocean, and Atmospheric Sciences at Rutgers University, the Bren School of Environmental Science and Management at University of California, Santa Barbara, the University of Delaware, Princeton University, and Fudan University. Together, they are linking state-of-the-art climate modeling, economic statistics, and big data analytics to build the world’s most comprehensive body of research quantifying the impacts of climate change around the globe. EPIC provides core financial and administrative support for the Lab. Learn more at