Matthew G Burgess et al 2021 Environ. Res. Lett. 16 014016
https://doi.org/10.1088/1748-9326/abcdd2

Scenarios used by the Intergovernmental Panel on Climate Change—IPCC—are enormously influential in climate science and policy. Recent studies have found that widely-used high emission scenarios project CO2 emission trajectories that are higher than recent observations and significantly higher than energy agency projections to 2040. 

This divergence is illustrated, for example, by this figure from Zeke Hausfather of the Breakthrough Institute. You can see that observed CO2 emissions—black—and International Energy Agency (IEA) projections—blue—are substantially lower than the IPCC “baseline scenario” range—grey—and the high-emission scenarios in dark brown and bright red.

The objective of our study was to understand why this divergence is happening, in order to inform better uses of scenarios in climate science and policy. We compared the projections of baseline scenarios used in the IPCC’s Fifth and upcoming Sixth Assessment Reports, to both observations going back to 2005 and energy agency outlooks from 2020-2040.

Our comparison focused on the Kaya identity factors. The Kaya identity breaks down CO2 emissions from energy into the product of population, GDP per capita, energy intensity (which is energy/GDP), and carbon intensity (which is CO2 emissions/energy). 

We found that most of the past divergence of observed CO2 emissions from baseline scenario projections can be explained by the scenarios projecting too much economic growth. Importantly, this pattern pre-dates the COVID-19 pandemic and is only partly explained by the Great Recession of 2008-2009. Indeed, as this figure shows, the largest divergences are in regions such as the Middle East and Africa and Latin America and the Caribbean, which were less affected by the Great Recession than the OECD. In the figure, the black line shows observed per-capita GDP growth rates; the colored lines show Shared Socioeconomic Pathway baseline scenario projections. These are the scenarios being used in the IPCC Sixth Assessment Report. The grey lines show projections of the COVID-19 recession and recovery made by the International Monetary Fund (IMF) in April 2020.  Clearly, the COVID-19 recession will only exacerbate the divergence.

The future of economic growth is uncertain, but we argue that past growth has fallen far enough behind the scenarios that it is unlikely to catch up before mid-century at least. We also argue that factors such as COVID-19’s effect on globalization, aging, debt, inequality, slower-than-expected economic convergence between poor and rich countries, future unexpected economic crises, and costs of climate damages and mitigation efforts are likely to make economic growth continue to be slower than projected by the IPCC baseline scenarios. But this is, of course, only a hypothesis.

Interestingly, the IEA’s economic projections have similar bases—and thus potentially similar errors—as the IPCC’s. So economic growth does not explain why the IEA projects lower CO2 emissions than IPCC baseline scenarios do. Instead, the primary reason is that IPCC baseline scenarios project much faster—and, we argue, unrealistic—rates of coal growth. For instance, as this figure shows, global coal use per capita has been roughly constant for the past hundred years. While the IEA projects that it will decline slightly by 2040, most IPCC baseline scenarios have it substantially increasing—in some cases by a factor of six or seven. Most energy analysts think this is unrealistic.

In closing, we offer several recommendations for climate research, two of which are: 1) distinguish exploratory research—which sometimes should use unlikely extreme scenarios—from policy-relevant scenarios, which focus on realism. 2) Update policy-relevant scenarios regularly as new observations become available, in the way that energy agencies do. Thank you and we hope you enjoy our paper.