U.S. offshores 15% of its carbon emissions, not 20% as originally stated

Two weeks ago, I published a post that claimed that the U.S. had offshored just over 18% of its carbon dioxide (CO2) emissions. I was wrong – it’s only 15%. The problem was in how I calculated the CO2 emissions of other countries. Instead of using actual estimates of CO2 emissions (publicly available at the Energy Information Administration), I used market exchange rates and purchase power parity (PPP) exchange rates, and so added a significant source of error that made the percentage vary from 18% (for market rates) to 10% (for PPP rates).

I realized that there was a way to make the results independent of the currency exchange rate, and that’s how I generated the graphs below.

The figure below shows the 20 nations who’s CO2 emissions contribute the most to U.S. emissions, sorted by 2006 (the latest year for which EIA data was available). China is the overall largest contributor to U.S. emissions due their high emissions and large amount of exports to the U.S., and these 20 nations represent approximately 98% of all offshored CO2 emissions. The total CO2 that the U.S. offshored to other countries was over a billion metric tons.


The figure below shows the top ten countries (responsible for approximately 85% of all emissions offshored) by percentage, with the U.S. direct emissions shown as well. Notice that the U.S. direct emissions are falling while the emissions by other nations on behalf of the U.S. are rising.


This last figure shows the official carbon intensity (CO2 emitted per unit of economic production) as compared to the actual carbon intensity, as well as the total percentage of U.S. CO2 emissions produced in other nations. Note that the actual carbon intensity is significantly worse than the official carbon intensity, although not as bad as I had calculated two weeks ago.


Ultimately, the conclusion or the original post remains unchanged – the U.S. has offshored its carbon dioxide emission problem to the rest of the world, turning their economies into dumping grounds for our own air pollution in the process.


Relying on PPP and MER to convert imports and exports into CO2 emissions results in a range of calculated values that vary according to how the PPP and MER compare to each other. This problem was eliminated by normalizing the value of the imports from other countries against the GDP of that country. Similarly, the value of U.S. exports to that country were normalized against U.S. GDP. Then the percentage of imports and exports were multiplied by total emissions in order to determine the CO2 value of the imports and exports. Finally, a balance of CO2 emissions due to trade was determined via simple subtraction.

All calculations were performed in current dollars, not “real” dollars (2000). However, since the calculations were normalized, this doesn’t matter to the accuracy of the calculations.

U.S. carbon intensity was calculated using the actual CO2 emissions (including emissions offshored on behalf of the U.S. economy) divide by the U.S. GDP. GDP was converted from current dollars to “real” dollars for direct comparison against official EIA carbon intensity figures.

To see the calculations yourself, click here (zipped .xls file).

7 replies »

  1. There’s a better than 50% chance the dollar will go up further against everything except the yen during the next year. It’s a lot worse everywhere else, and bunch of other reasons are keeping the dollar afloat.