Last data update: November 2021
The OECD Inter-Country Input-Output (ICIO) database, when combined with statistics on CO2 emissions from fuel combustion and other industry statistics, can be used to estimate demand-based carbon dioxide emissions. That is, the distribution across economies of final demand (household consumption and industry investment) for embodied carbon that has been emitted anywhere in the world, along global production chains.
The Trade in embodied CO2 (TECO2) database presents a set of indicators to reveal patterns of CO2 demand compared to CO2 production (via resident industry or household emissions). The aim is to provide policy makers with new insights into the environmental impacts of global production systems.
Indicators presented in TECO2 database include:
Data downloads: OECD.Stat
Note: In case of issue when downloading, right-click and select "save as".
The six largest producers and consumers countries of CO2 emissions in 2018 were China, United States, India, the Russia Federation, Japan and Germany; the European Union (EU27_2020) is also included for reference. While, since 1995, both production and consumption of emissions have fallen in Japan, Germany and the European Union, there has been a significant increase in China and India. China has the highest absolute emissions from both a demand and a production perspective. However, even though China’s per capita demand for CO2 emissions increased 3 times in 2018, when compared to 1995, US per capita demand in 2018 was still 2.8 times higher than that of China.
Total production and consumption based CO2 emissions by top 6 emitters
Source: OECD (2021), CO2 emissions embodied in international trade, http://oe.cd/io-co2.
The difference between production-based and demand-based carbon emissions is highlighted in the graph below. While the OECD countries, in total, are net-importers of embodied carbon (the solid blue line representing demand-based emissions is above the dashed blue line representing production-based emissions), non-OECD countries, as a whole, are net-exporters. Note that the shaded blue area (OECD net imports) and green area (non-OECD net exports) have the exact same size, i.e. OECD net-imports are non-OECD net-exports of embodied carbon. OECD net-imports grew from 1995 to reach a peak in 2006 and, ever since, have been gradually declining.
Total production and consumption based CO2 emitted by OECD and non-OECD countries
However, not all OECD countries are net-importers of carbon and similarly, not all non-OECD economies are net-exporters. For example, in the OECD area, Canada, Korea, Poland, Turkey, Czech Republic, the Netherlands and Denmark were net-exporters in 2018 while among non-OECD economies, Argentina and Brazil were net importers. Looking at the OECD and G20 countries, the average of the three largest demand-based CO2 emitters per capita (Australia, Saudi Arabia and the United States, 16.6 tonnes CO2) is nearly 8 times higher than that of the top three economies with lowest per capita emissions (Brazil, Indonesia and India, 2.0).
Per-capita production and consumption based emissions in selected countries
Demand-based emissions are calculated using IEA data on “CO2 emissions from fuel combustion” and the OECD Inter-Country Input-Output (ICIO) system (edition 2021). Using information from both, emission-intensities of production are calculated for each industry in each country. These intensities are then combined with the Leontief inverse of the ICIO system to get emission multipliers for final demand.
The carbon emissions associated with final demand FD_CO2 are calculated as follows:
where diag(EF) is the the diagonalised matrix form of vector EF of industry emissions per unit of production (i.e. emission factors) by country, of size KN (K*N) where K = number of industries and N = number of countries. The fuel purchased abroad by domestic airlines and marine water transport industries are now included from the 2019 edition of embodied CO2. A is the global intermediate coefficients matrix so that (I-A)-1 is the global Leontief inverse (both of size KN x KN), and Y is the global final demand matrix of size KN x N.
The rows in the result matrix CC, of size KN x N, represent the country and industry origins of emissions, while the columns represent the final demand countries where the emissions are consumed.
Total CO2 embodied in final demand by country is then estimated as follows:
where colsum(CC) is a 1 x N vector of total emissions embodied in the consumption and investment of final goods and services by domestic industries and households, while FNLC is 1 x N vector of direct emissions due to the combustion of utility and transport fuels by households. To ensure that all emissions are taken into account, these emissions need to be added to that of matrix CC. Similarly, to account for all production-based emissions, FNLC needs to be added to the rows of CC (distributed primarily to the manufacture of petroleum products and the electricity, gas and water supply industries).
Yamano N. and J. Guilhoto (2020) "CO2 emissions embodied in international trade and domestic final demand: methodology and results using the OECD Inter-Country Input-Output Database". OECD, Science, Technology and Innovation Working Papers, No. 2020/11, OECD Publishing, Paris, https://doi.org/10.1787/8f2963b8-en
Wiebe, K. S. and N. Yamano (2016), "Estimating CO2 Emissions Embodied in Final Demand and Trade Using the OECD ICIO 2015: Methodology and Results". OECD Science, Technology and Innovation Working Papers, No. 2016/5, OECD Publishing, Paris, https://doi.org/10.1787/5jlrcm216xkl-en
Any suggestions or queries can be sent to email@example.com, mentioning CO2 in the title of your message.
Please cite as: OECD (2021), Trade in embodied CO2 (TeCO2) Database.