Since their peak in 2007 GHG emissions in the USA have decreased more in absolute terms than in any other country. The results of this review suggest that approximately 40% of this decrease was caused by the replacement of coal with gas in generating plants, 30% by improvements in the efficiency of internal combustion engines and 30% by growth in low-carbon renewables. Another major contributor was the 2008-9 global recession, although its impact can’t reliably be quantified. Had economic growth continued at historic rates between 2007 and the present US GHG emissions would now be substantially higher than they are.
This review was prompted by an article in Time Magazine written by Michael Bloomberg, one of the world’s richest men and also one of its leading contributors to green causes, entitled “Where Washington Fails to Drive Progress, Cities Will Act”. Bloomberg’s conclusion that the decrease in US emissions was “driven by cities, businesses and citizens”struck me as complete nonsense, and I was going to write more about it but decided it wasn’t worth the effort. But his article did get me thinking about what really did cause US emissions to fall, and in this post I document the results of my efforts to find out.
We begin with emissions. Figure 1 shows total US greenhouse gas emissions by sector between 1990 and 2014 (2015 emissions are not available but would probably be slightly lower than 2014 emissions. The data are from the EPA’s US Greenhouse Gas Inventory Report):
Figure 1: Total US greenhouse gas emissions by sector expressed as CO2 equivalents, 1990-2014. Emissions from land use changes and from US territories are not included.
Note that Figure 1 shows total greenhouse gas emissions, which include CO2 along with methane, nitrous oxide and minor GHGs expressed as CO2 equivalents, from all sectors of the economy, not just the CO2 emissions from electricity generation that we usually get to see. According to these results the US electricity sector is in fact responsible for only about 30% of total US GHG emissions – a good illustration of the futility of concentrating on cutting electricity sector emissions while ignoring emissions from other sectors.
Between 2007 and 2014 total US annual GHG emissions fell from a peak of 7,370 to 6,826 million tonnes. This 544 million-tonne decrease is in the same range as the UK’s total annual GHG emissions – i.e. not peanuts. Figure 2, which plots emissions by sector, shows how the different sectors contributed to the decrease (data again from EPA).
Figure 2: Contributions to emissions decreases by sector, 1990-2014.
We see the following absolute changes in CO2 equivalent emissions between 2007, the peak emissions year, and 2014:
- Electricity: down 373 million tonnes
- Transportation: down 184 million tonnes
- Industry: down 48 million tonnes
- Agriculture: down 6 million tonnes
- Commercial & Residential: up 67 million tonnes
The emissions reductions were achieved almost entirely in the electricity sector with an assist from transportation. What caused them? Figure 3 shows generation from the electricity sector from 2007 through 2015 (data from EIA annual electricity reports ).
Figure 3: US Electricity sector generation by source, 2007-2015.
Hydro and nuclear have not changed significantly but coal and gas have, and there is a clear antiphase relationship between the two. Between 2007 and 2014 coal generation fell by 400TWh and gas generation rose by 250TWh, largely a result of a surplus of fracked gas displacing more expensive coal. Between 2007 and 2015 coal generation fell by 600TWh and gas generation rose by 500TWh. In terms of its impact on emissions this is roughly equivalent to replacing 250-300TWh of coal generation with low-carbon generation, which is sufficient to cut annual CO2 emissions by up to 300 million tonnes. However, generation from low-carbon renewable sources (dominantly wind) increased by 200TWh over the same period, indicating that the contribution of renewables to the emissions decrease was not that much lower than the coal-to-gas transition.
The cause of the decrease in emissions from the transportation sector is not immediately obvious. Transportation emissions are over 90% dependent on the consumption of gasoline and diesel fuel, and as shown in Figure 4 these fell during the 2008-9 recession. But along with total miles traveled they had recovered to around 2007 levels by 2015, so the emissions decrease can’t be attributed to lower fuel use or less driving. (Data from the 2017 EIA Petroleum & Other Liquids report and the Federal Highway Administration):
Figure 4: US deliveries of gasoline and diesel fuel and total miles traveled, 1984-2015
But after a little searching I came across Figure 5, reproduced from the EPA’s Light-Duty Automotive Technology, Carbon Dioxide Emissions, and Fuel Economy Trends 1975 Through 2016:
Figure 5: CO2 emissions/mile and fuel economy for US cars and trucks, 1975-2016
The left-hand graph shows how CO2 car and truck emissions, expressed in grams/mile, have decreased steadily since 2005. Between 2007 and 2016 combined car and truck emissions fell by almost 20%, from 430 to 350gm/mile, largely as a result of improvements in engine technology. This is more than enough to explain the approximate 10% decrease in transportation sector emissions, but the 20% number will be too high because the data are for vehicles purchased new and don’t allow for emissions from vehicles already on the road at the time (the average age of a car in the US is now estimated to be 11.5 years.)
Based on the results discussed above I came up with the following approximate distribution of percentage contributions to the US emissions decrease since 2007:
- Gas replacing coal in electricity generation: 40%
- Decrease in gm/mile vehicle CO2 emissions: 30%
- Growth in low-carbon renewables generation: 30%
The US government can take credit for roughly a third of the decrease. The 30% contribution from renewables is of course entirely a result of lavish government subsidies, and we can further assume that the anti-coal policies of the Obama administration would have had at least some impact on the decrease in coal usage and that the US Corporate Average Fuel Economy standards might have had a minor impact on gm/mile CO2 emissions, although there was no change in the standards that might explain the post-2005 decrease. The government also provided support to George Mitchell during the 1970s but despite claims to the contrary wasn’t responsible for the fracking boom. As one of Mitchell’s geologists put it: “George probably could have done it without the government. The government would not have done it without George.”
The wild card here, however, is the 2008-9 global recession. Figure 6 reproduces Figure 1 with a trend line drawn through the 1996-2007 data. Had this trend continued annual US GHG emissions would now be about 1 billion tonnes higher than they are. Does this mean that the 2008-9 recession was the basic cause of the emissions decrease?
Figure 6: Total US GHG emissions from Figure 2 with 1996-2007 trend line superimposed
Let’s look at a few more indicators. First US GDP since 1950 (Figure 7, data from the St. Louis fed). We see that previous recessions, such as the 1970, 1974-5 and even the 1981-82 “double dip” recession, had no lasting impact on the overall upward trend in GDP. The 2008-9 recession, however, did. Since 2009 US GDP has run $1-2 trillion/year below the projection of the trend line (which may explain the election of Donald Trump, but I digress):
Figure 7: US GDP, constant 2010 dollars, 1950-2014
Second is US primary energy consumption (Figure 8, data from the 2016 BP Statistical Review). The trend here is convex upwards, indicating a gradual decrease in the growth rate of total energy use with time. But like the GDP graph it shows a sustained decrease, in this case of about 150 million MTOE/year, relative to the trend line from 2009 onwards.
Figure 8: US primary energy consumption, 1985-2015
Third is US total electricity consumption. Since 2007 electricity consumption has flattened out and in 2015 was running over 500TWh/year below the projection of the trend line (Figure 9, data from the EIA electricity browser):
Figure 9: US total electricity consumption, 2001-15
Figures 6, 7, 8 and 9 demonstrate once again the close linkage between GDP, energy and emissions.
So did the 2008-9 recession cause the decrease in US emissions? If economic growth had continued at historic rates between 2007 and the present there’s no doubt that US emissions would now be higher than they are – according to Figure 6 about a billion tonnes higher. Accepting this number as a meaningful estimate is, however, difficult because it involves speculation. No one can be certain what would have happened if economic growth had continued unchecked during 2008-9. The world might, for example, have suffered an even deeper recession in 2009-10. The only thing we can be certain of is that if the goal is to reduce emissions an economic downturn is the best way to achieve it.