Inconvenient Data – 1
In his NY Times column “Use Energy, Get Rich and Save the Planet” published on April 20, 2009, John Tierney says that “…the United States and other Western countries seem to be near the top of a Kuznets curve for carbon emissions and ready to start the happy downward slope. The amount of carbon emitted by the average American has remained fairly flat for the past couple of decades…” He attributes this trend to our increasing affluence: “…As [our] wealth grows, people consume more energy, but they move to more efficient and cleaner sources…”
Is Mr. Tierney correct? This blog takes a look at historical US carbon emission data to refute his claims.
Predictions of future levels of CO2 emissions may be seriously flawed if they are based only on trends of CO2 emissions per capita. Any scientifically-based analysis should include the combined effects of consumer behavior, technology evolution and population growth. This is actually what the IPAT equation attempts to do. But, the IPAT equation and Mr. Tierney’s claims about it will be the subject of another blog.
Figure 1 plots the US CO2 emissions per capita (in tons of CO2 per person per year) from 1949 until 2007 using the historical data published by the US Department of Energy (DOE) . At first glance, the DOE data seem to support Mr. Tierney’s claim. Indeed, the per capita emissions have been hovering around 20 metric tons per person for the past 25 years and may be on a downward path for the past few years.
First, the per capita emissions must decrease more rapidly that what is shown in Figure 1 to compensate for population growth. Contrast, for example, the recent trends of Figure 1 with the 25% increase in US population over the last 20 years.
But, the DOE data do not support even the simple claim that the per capita CO2 emissions are leveling off. A different picture for the per capita emissions appears from a more careful look at the detailed DOE data that separately track CO2 emissions from the major sectors of the economy (transportation, residential, commercial, industrial) and the primary energy sources (coal, petroleum, natural gas etc.)
Figure 2 clearly shows that the CO2 emissions per capita from the transportation, residential, and commercial sectors of the US economy continue to increase at a fairly constant rate for the past 25-30 years. Only the emissions from the industrial sector have shown a significant decline since the mid 1990’s.
An analysis of the DOE data for the industrial sector shows that the amount of CO2 emitted per unit of energy consumed by the US industries has remained fairly constant. Therefore, the recent drop in overall industrial CO2 emissions is (to a significant extent) the result of the movement towards offshore manufacturing that shifts CO2 emissions away from the US.
Unlike sulfur dioxide, however, CO2 is a global pollutant and we must consider the emissions resulting from the manufacturing of the goods necessary to support our ever-more-affluent lifestyle, even when these emissions actually occurred in China or other offshore countries. Since we don’t have reliable data to estimate these contributions to the emissions of the US industrial sector, we will focus our attention on the combined per capita CO2 emissions from the transportation, residential and commercial (TRC) sectors. In any case, these three sectors account for three quarters of our total CO2 emissions.
Figure 3 plots the normalized total emissions from the three TRC sectors vs. the normalized per capita GDP in chained 2000 dollars with 1949 as our reference year. The combined TRC emissions show a rapid increase with GDP for the period 1949-1973 and a slower but steady increase from 1983 until 2007 (see Figure 3). A linear trendline gives a good fit for the 1983-2007 data, while an exponential trendline provides an even better fit for the 1949-1973 data. Both trendlines shown in Figure 3 have a correlation coefficient larger than 0.9.
Figure 3 gives no indication that “.. we are near the top of a Kuznets curve for carbon emissions and ready to start the happy downward slope…” CO2 emissions from the transportation, residential and commercial sectors of the US economy have been on the same upward slope since the early 1980’s, despite the fact that our GDP per capita has increased by more than 60% in the same period and is now more than $38,000.
But, what about the following statement attributed to J. Ausubel of the Rockfeller Institute: “…Over the past century, nothing has drastically altered the long-term trends in the way Americans produce or use energy — not the Great Depression, not the world wars, not the energy crisis of the 1970s…”
Figure 3 clearly shows a rather disruptive change in the pattern of CO2 emissions as the US went through the energy crises of the 1970s and early 1980s. Figure 4 also shows that the US energy consumption per capita (in GJ per person) exhibits similar trends with a sharp (exponential-like) increase in the 1949-1973 period and a slower (linear) increase in the 1983-2007 period. These trends are more obvious if we consider the combined energy consumption for the transportation, residential and commercial sectors (see above).