Declining U.S. carbon dioxide emissions
March 24, 2013
Declining U.S. carbon dioxide emissionsEmissions of carbon dioxide from fossil fuel consumption in the United States have fallen remarkably since 2008, with recent levels the lowest since 1995. Here I comment on some of the factors behind this.
Carbon dioxide emissions from U.S. consumption of fossil fuels for 12 months ending at indicated date, Jan 1975 to Nov 2012, in millions of metric tons. Data source: EIA Monthly Energy Review, Feb 2013The policy response to growing atmospheric carbon dioxide concentrations that is favored by most economists is a tax on carbon emissions. If we could calculate in dollar terms the economic damage associated with an additional ton of carbon emissions, a tax in exactly that amount would lead society to reduce emissions by the optimal amount in the least costly manner. Estimates of that dollar cost differ greatly across academic studies, but a modal estimate might be $25 a ton, which would correspond to about 25 cents a gallon of gasoline. A relatively high estimate from the range of academic studies would be in the neighborhood of $100 per ton of CO2, or about $1.00 per gallon of gasoline.But anyone who buys gasoline knows that 25 cents a gallon is only half the size of the run-ups we saw in the spring of each of the previous 3 years. Longer term, the $2/gallon increase since 2003 is about twice the size of the upper range of proposed carbon taxes. As far as gasoline is concerned, the U.S. did not adopt a carbon tax, but the bottom line for consumers was a change much bigger than most of the policy proposals envisioned.
Average weekly U.S. retail gasoline price in dollars per gallon. Data source: EIAAnd it's worth remembering why that happened-- we didn't have a choice. Global field production of crude oil (excluding natural gas liquids, which are not used as transportation fuel) stagnated at about 74 million barrels/day between 2005 and 2008. It is up a couple of million barrels since then, but more than 100% of this increase has been consumed by China alone, forcing the U.S. and other countries to reduce our oil consumption.
Global crude oil field production, millions of barrels per day. Data source: EIALooking at the breakdown of U.S. carbon emissions by sources, reduced consumption of petroleum made a prominent contribution to the total, as seen in the turquoise line on the graph below.
Carbon dioxide emissions from U.S. consumption of petroleum, coal, and natural gas. Data source: EIA Monthly Energy Review, Feb 2012But rising oil prices cannot account for the concomitant decline in carbon dioxide emissions from coal. And here I want to argue that at least one reason for declining use of coal was that we did have a choice, and chose something better. The U.S. has been quite successful at increasing natural gas production in recent years.
Data source: EIA.Rising gas production helped lower prices, which has helped make gas more competitive relative to coal.
Data source: EIA.Generating electricity from natural gas emits about half the carbon dioxide as coal. Hence we enjoyed a serendipitous, unplanned decline in carbon emissions from a development that ended up benefiting consumers rather than forcing them into painful adjustments.This is not to claim that the problem is solving itself. An even bigger factor in the reduction in fuel use was likely the economic recession itself. The Economic Report of the President estimated that 52% of the decline in U.S. CO2 emissions could be attributed to the economic recession, 40% to fuel switching, and 8% to improved energy efficiency. As the economy recovers, the EIA is expecting U.S. emissions to resume their historical climb if there are no new policy actions.
Source: Economic Report of the President.But there is one important lesson from the recent U.S. experience that's worth emphasizing: these outcomes may ultimately end up being driven by events that are bigger than any policy initiatives we might contemplate.UPDATE (3/25/2013 at 4:11 PDT): Reader benamery21 correctly notes a problem with a graph that appeared in the original version of this post, which has since been removed.Posted by James Hamilton at March 24, 2013 12:33 PM