U.S. Energy Information Administration (EIA) - Ap
U.S. Energy Information Administration logo
Skip to sub-navigation

As of July 1, 2022, we are continuing to restore our systems. The monthly data releases, including the Petroleum Supply Monthly, Natural Gas Monthly, and Electric Power Monthly, will be published next week. We will continue to post regular updates regarding the status of other data products.

Environment

U.S. Energy-Related Carbon Dioxide Emissions, 2020

Release Date: December 22, 2021  |  Next Release Date: December 2022   |   full report

Previous Issues:

Introduction

In 2020, the COVID-19 pandemic substantially affected the U.S. and world economy, energy markets, and energy-related carbon dioxide (CO2) emissions.

Total U.S. energy-related CO2 emissions decreased by 11% in 2020, or 570 million metric tons (MMmt) of CO2 relative to 2019. Both short-term and long-term factors contributed to this decrease:

  • The main factor in declining emissions in 2020 was a short-term reduction in energy demand because of the COVID-19 pandemic.
  • U.S. emissions related to petroleum product consumption, such as motor gasoline and jet fuel, decreased by 14% (330 MMmt) from the previous year. Consumption of both fuels decreased as a result of an increase in working from home and a decrease in travel demand during the COVID-19 pandemic.
  • A decrease in demand for heating fuels reduced CO2 emissions. The winter was warmer; the U.S. experienced 9% fewer population-weighted heating degree days in 2020 than in 2019. Demand for cooling remained mostly the same, with only a slight increase over the previous year.
  • Continuing changes in the fuel mix of electricity generation, exacerbated in 2020 by low natural gas prices, is the longer-term trend causing part of this decline. Emissions from coal generation declined by 19%, or 200 MMmt, which is about the same change from 2018 to 2019.

The combination of conditions that lowered energy-related CO2 emissions in the United States in 2020 relative to 2019 does not necessarily represent future trends, especially those related to the highly unusual economic and energy-related impacts created by the pandemic.

As a result of both short- and long-term factors, U.S. emissions decreased across all sectors:

  • The transportation sector had the largest decline, with emissions falling by 15% (290 MMmt).
  • Residential sector emissions declined in 2020 by 6% (57 MMmt). Although people stayed at home more often last year, the warmer-than-average winter temperatures resulted in lower-than-average heating demand, which led to an overall decrease in emissions.
  • Commercial sector emissions declined by 12% (100 MMmt) in 2020 due to a decline in commercial building activity brought about by lockdown restrictions and increased working from home.
  • Industrial CO2 emissions decreased by 8% (120 MMmt) following a decrease in U.S. industrial activity.

Overview

The 2020 decline in U.S. energy-related CO2 emissions was historic

In 2020, energy-related CO2 emissions declined to a level not seen since 1983. Much of this decline was related to the COVID-19 pandemic and the resulting economic downturn. Since peaking in 2007, emissions have declined in 9 of the past 13 years. The magnitude of the 2020 decline in emissions was bigger than all previous years on record, both in absolute and percentage terms.


 

U.S. energy-related CO2 emissions were 570 MMmt lower in 2020 than in 2019

Total energy-related CO2 emissions in 2020 were 11% lower than in 2019. Percentage changes in energy-related CO2 emissions can be broken down into changes in four factors:

  • Energy intensity (energy consumed per unit of GDP)
  • Carbon intensity (CO2 emissions per unit of energy)
  • Population
  • GDP per capita

These factors, when combined to estimate total energy-related CO2 emissions, are known as the Kaya identity.

Carbon intensity, energy intensity, and GDP per capita each fell by 4% in the United States between 2019 and 2020, mostly because of the impacts of the COVID-19 pandemic. The Kaya identity estimates that each of these factors decreased energy-related CO2 emissions by about 200 MMmt. The only factor that caused emissions to increase in 2020 was slight population growth, resulting in an additional 17 MMmt of CO2.

The 4% decrease in U.S. carbon intensity came largely from a decrease in the consumption of fuels with high carbon contents. Part of this change came from the continuing trend of natural gas and renewables displacing coal for electric power generation, both of which have lower or zero carbon content. Low natural gas prices supported this switch from coal use, and higher natural gas prices in 2021 have started to reverse this trend. Most of the change in carbon intensity, however, came from a decrease in consumption of high carbon content fuels in the transportation sector, namely motor gasoline and jet fuel.

U.S. GDP decreased on a per capita basis because of the economic impacts of the pandemic. U.S. energy intensity also decreased as declining energy consumption, associated with the pandemic, outpaced declines in GDP. Total energy consumption in 2020 declined by 7%, while GDP declined by 4%. The majority of the decline in energy consumption, about 58%, came from the transportation sector, which was most affected by the COVID-19 pandemic.

 

 

Fossil Fuels

The pandemic had different effects on fossil fuel consumption and its related carbon dioxide emissions

The largest absolute decline in U.S. fossil fuel CO2 emissions was associated with petroleum consumption, at 14% (329 MMmt). On a percentage basis, coal emissions declined at a higher rate—19% (203 MMmt). Natural gas emissions declined by 2% (37 MMmt).

The decline in U.S. emissions associated with coal consumption continued a trend from previous years. Since 2007, U.S. energy-related CO2 emissions from coal declined on average by 6% each year. Coal emissions’ 19% decline in 2020 surpassed the previous record decline of 15% set in 2019. The decline in 2020 was due to decreased demand for electricity as a result of the COVID-19 pandemic, as well as low natural gas prices. Natural gas prices for the electric sector fell by 17% in 2020, their lowest level since 1998, making natural gas a more competitive resource than coal for electricity generation.

The slight decrease in U.S. natural gas emissions in 2020 was a result of two counteracting forces. Natural gas emissions from the electric power sector increased, driven by low natural gas prices. However, demand for natural gas for heating decreased in 2020 relative to 2019 due to a warmer winter. Winter 2020 had a 9% decrease in total U.S. population-weighted heating degree days (HDDs)—a measure of heating demand. Together, the increase in demand for natural gas-fired electricity and the decrease in demand for natural gas heating lead to a relatively small decline in natural-gas related emissions.

Petroleum emissions had the largest decline among all fuels in the United States during 2020, mostly because of the impacts of the COVID-19 pandemic on domestic and international travel. Travel restrictions as well as an increase in working from home led to a sharp decrease in fuel consumption. In 2020, energy-related CO2 emissions from motor gasoline fell by 13%, and from jet fuel by 38%, reaching their lowest levels since 1991 and 1983, respectively. Part of this decline in petroleum emissions also came from the industrial sector, where petroleum emissions fell by 7% because of a decline in industrial activity.

 

Electricity

U.S. electric power sector CO2 emissions fell by 11% in 2020 (170 MMmt). This decline in emissions was due to changes in both electricity usage and fuel mix. Economic impacts of the COVID-19 pandemic, as well as relatively mild winter weather, led to an overall decrease in electricity usage in 2020; total generation declined by 3% (113 terawatthours).

Changes in the mix of fuels used to generate electricity were responsible for some reductions in U.S. emissions. A seven-year trend of decreasing coal-fired generation and increasing generation from lower-carbon fuel sources continued in 2020. Coal’s share of total electricity generation fell from 24% to 20%, and natural gas’s share increased from 37% to 39%. The generation share of zero-carbon power increased from 38% to 40%. Generation shares of all other sources remained relatively stable. This change in generation mix led to an 8% decrease in the carbon intensity of electricity, from 0.41 to 0.38 metric tons of CO2 per megawatthour.

 

 

End-Use Sectors

In 2020, U.S. energy-related CO2 emissions declined in all end-use sectors. These declines take into account both direct and indirect emissions from each sector. Direct emissions are each sector’s emissions from the direct consumption of fossil fuels, such as natural gas for heating or gasoline in cars. Indirect emissions are emissions from electricity generation, attributed to each end-use sector based on its share of total electricity consumption.

The U.S. transportation and commercial sectors experienced the largest declines in emissions, largely because of the economic impacts of the COVID-19 pandemic. Emissions from the residential and industrial sectors fell by slightly less.

 

Despite lockdown restrictions and increases in working from home associated with the COVID-19 pandemic, U.S. residential CO2 emissions fell by 6% (57 MMmt) relative to 2019. People spent more time at home, leading to a 2% increase in sales of electricity to the residential sector. However, this increase was offset by a decrease in heating demand, leading to a net decrease of 1% in residential energy demand.

Unlike in the residential sector, lockdown restrictions and working from home led to a significant decline in activity in U.S. commercial buildings, resulting in a 12% (100 MMmt) decrease in energy-related CO2 emissions from the commercial sector.

 

A further effect of the COVID-19 economic downturn was a decrease in industrial activity. U.S. manufacturing output fell by 7% in 2020, which led to an 8% decrease (120 MMmt) in industrial CO2 emissions. Output decreased across most industries; petroleum and coal products and primary metals decreased the most, each by 13%.

 

The U.S. transportation sector had the largest decrease in CO2 emissions (15%, or 293 MMmt), largely because of reduced travel due to the COVID-19 pandemic. Increases in people working from home, closure of public venues, and decreases in domestic and international travel led to a 13% decrease in emissions from motor gasoline, a 38% decrease in emissions from jet fuel, and an 8% decrease in emissions from diesel fuel.

 

Future Implications of the 2020 Decrease in Emissions

The combination of conditions in 2020 that lowered energy-related CO2 emissions in the United States relative to 2019 do not necessarily represent future trends, especially because of the highly unusual economic and energy-related impacts caused by the COVID-19 pandemic. The EIA products highlighted in this section contain the most recent data on short-term (2021 and 2022) and long-term (through 2050) energy-related CO2 emissions for the United States. Both the short-term forecast and long-term projections of energy-related CO2 emissions remain uncertain and rely on assumptions related to both the speed and nature of the economic recovery from the COVID-19 pandemic and any other longer-term behavioral changes, such as an increase in working from home.

EIA’s short-term forecast of U.S. energy-related CO2 emissions and key drivers is in our Short‐Term Energy Outlook (STEO). The STEO contains monthly emissions forecasts for the United States by fuel source through the next year (a two-year forecast) and is the timeliest source for our latest estimates on the effects of recent events on energy markets and energy-related CO2 emissions.

We publish our long-term U.S. emissions projections in the Annual Energy Outlook (AEO). The AEO provides annual projections of energy-related CO2 emissions by fuel source, sector, and end use, as well as projections of other elements of energy markets, through 2050.

Although not explicitly mentioned in this report, we also provide annual projections of international energy-related CO2 emissions through 2050 in our International Energy Outlook (IEO).

Our analysis of U.S. energy-related emissions in this report is based on data published in both the STEO and the Monthly Energy Review (MER).

Sector Contributions to 2020 Declines in U.S. Energy-Related CO2

When analyzing year-to-year changes in energy-related CO2 emissions, we calculate each sector’s share of the overall change in CO2 emissions. Annual changes in energy-related CO2 emissions in each sector are affected by changes in:

  • Electricity consumption levels
  • The fuel mix of electricity generation (which determines the carbon intensity of electricity consumed)
  • Primary energy consumption levels
  • The fuel mix of direct consumption of primary energy (which determines the carbon intensity of primary energy consumed)

Table 1 shows each end-use sector’s share of the total change in energy-related CO2 emissions for the U.S. economy in 2020. The table includes:

  • CO2 emissions resulting from the change in each sector’s electricity consumption, measured in British thermal units (Btu), from 2019 to 2020
  • CO2 emissions resulting from the change in the electricity generation fuel mix for electricity consumption and the resulting change in carbon intensity (CO2/Btu) of electricity sales to end-use sectors
  • CO2 emissions resulting from the change in direct primary energy consumption (Btu) by sector
  • CO2 emissions related to changes in carbon intensity (CO2/Btu) by sector
  • The change in CO2 emissions for each end-use sector based on the sum total of the changes for electricity and direct primary energy consumption
  • The overall change in CO2 emissions for all sectors from 2019 to 2020
Table 1. Sector contributions by electricity and primary energy changes to the total energy-related carbon dioxide emissions change
  Residential Commercial Industrial Transportation Total all sectors
Change in electricity-related CO2 emissions, 2019–20 -34 -73 -62 -1 -170
Change related to the carbon intensity of electricity-related CO2 emissions, 2019–20 -46 -40 -29 0 -115
Electricity-related CO2 with no change in carbon intensity, 2019–20 12 -34 -33 0 -55
Change in primary energy-related CO2 emissions, 2019–20 -23 -25 -58 -293 -399
Change related to the carbon intensity of primary energy-related CO2 emissions, 2019–20 0 -2 -18 -9 -29
Primary energy-related CO2 emissions with no change in carbon intensity, 2019–20 -23 -23 -40 -284 -370
Sum of the change in electricity and primary energy CO2 emissions, 2019–20 -57 -98 -120 -293 -569

Source: U.S. Energy Information Administration (EIA), Monthly Energy Review, October 2021, Tables 11.2–5, Carbon Dioxide Emissions from Energy Consumption by Sectors

Method for Including CO2 Emissions from Electricity Generated Outside of the Electric Power Sector

Not all electricity used in the United States is generated by the electric power sector. In particular, in the commercial and industrial sectors, coal, natural gas, petroleum and biomass are also used to generate power for use on site (accounting for 4% of total generation). Table 2 presents our analysis of CO2 emissions originating from outside of the electric power sector. We based the calculations for this analysis on our Monthly Energy Review (MER), Table 7.3c, Consumption of Selected Combustible Fuels for Electricity Generation: Commercial and Industrial Sectors (Subset of Table 7.3a). To perform this calculation, we used the following CO2 emissions factors:

  • Coal
    • 95.74 million metric tons per quadrillion Btu for the commercial sector
    • 95.59 million metric tons per quadrillion Btu for the industrial sector
  • Natural gas
    • 52.91 million metric tons per quadrillion Btu for both the commercial sector and the industrial sector
  • Petroleum
    • 74.15 million metric tons per quadrillion Btu for the commercial sector
    • 73.95 million metric tons per quadrillion Btu for the industrial sector

Emissions factors for coal and natural gas consumed in the United States are from the detailed factors spreadsheet available on our Environment page. We constructed petroleum factors manually by using each end-use sector’s consumption and emissions of distillate fuel oil and residual fuel oil.

We applied these factors to the amount of each fuel combusted (in Btu) to produce electricity in the commercial and industrial sectors. These calculations account for the changes in the carbon intensity (CO2 per kilowatthour) of electricity generated from all sources. Biomass is excluded from these emissions calculations because we assume biomass to be carbon-neutral.

Table 2. Carbon dioxide emissions from electricity generated outside of the electric power sector
  CO2 emissions from generation within the commercial sector (excludes CO2 emissions from the electric power sector) CO2 emissions from generation within the industrial sector (excludes CO2 emissions from the electric power sector) Total commercial and industrial CO2 emissions
  Coal Natural gas Petroleum Total Coal Natural gas Petroleum Total Total
2005 0.81 1.84 0.23 2.88 15.91 28.16 2.45 46.52 49.4
2006 0.73 1.88 0.13 2.75 15.61 29.14 1.92 46.67 49.42
2007 0.76 1.85 0.1 2.72 10.88 30.08 1.89 42.86 45.57
2008 0.81 1.82 0.06 2.69 10.82 28.26 1.36 40.45 43.14
2009 0.69 1.86 0.07 2.62 9.75 28.19 1.22 39.16 41.79
2010 0.68 2.14 0.07 2.88 16.97 30.06 0.88 47.91 50.79
2011 0.73 2.55 0.05 3.34 11.82 30.9 0.78 43.5 46.84
2012 0.63 3.42 0.11 4.15 9.56 34.35 1.71 45.62 49.78
2013 1.04 3.62 0.13 4.79 9.64 34.93 1.38 45.95 50.74
2014 0.41 3.93 0.17 4.52 9.52 34.07 0.92 44.52 49.03
2015 0.32 3.85 0.1 4.27 8.13 34.3 0.67 43.1 47.37
2016 0.21 2.55 0.04 2.8 6.08 29.33 0.6 36.01 38.82
2017 0.18 2.75 0.08 3 5.53 29.69 0.54 35.77 38.77
2018 0.16 2.89 0.1 3.15 5.02 31.05 0.49 36.56 39.71
2019 0.14 3.06 0.09 3.29 4.28 34 0.46 38.73 42.02
2020 0.12 2.85 0.09 3.06 3.57 33.02 0.39 36.98 40.04

Sources: U.S. Energy Information Administration, Monthly Energy Review, October 2021, Table 7.3c, Consumption of Selected Combustible Fuels for Electricity Generation: Commercial and Industrial Sectors (Subset of Table 7.3a) and Carbon dioxide Emissions Coefficients by Fuel

Terms used in this analysis

British thermal unit(s) (Btu):The quantity of heat required to raise the temperature of one pound of liquid water by 1°F at the temperature at which water has its greatest density (approximately 39°F).

Carbon intensity (economy): The amount of carbon by weight emitted per unit of economic activity—most commonly gross domestic product (GDP) (CO2 emissions/GDP). The carbon intensity of the economy is the product of the energy intensity of the economy and the carbon intensity of the energy supply. We currently express this value as the full weight of the CO2 emitted, rather than the weight of just carbon.

Carbon intensity (energy supply): The amount of carbon by weight emitted per unit of energy consumed (CO2 emissions/energy). A common measure of carbon intensity is the weight of CO2 per Btu of energy. When considering only one fossil fuel, the carbon intensity and the emissions coefficient are identical. When considering several fuels, carbon intensity is based on their combined emissions coefficients weighted by their energy consumption levels. We currently measure this value as the full weight of the CO2 emitted, rather than the weight of just carbon.

Cooling degree days (CDD): A measure of how warm a location is during a period of time relative to a base temperature of 65°F. CDD are used in energy analysis as an indicator of air-conditioning energy requirements or use. The measure is computed for each day by subtracting the base temperature (65°F) from the average of the day's high and low temperatures, and negative values are set equal to zero. Each day's CDD are added to create a CDD measure for a specific time period.

Energy intensity: A measure relating the output of an activity to its energy input. Energy intensity is most commonly applied to the economy as a whole, where we measure output as GDP and primary energy in Btu to allow for the addition of all energy forms (energy/GDP). On an economy‐wide level, energy intensity reflects both energy efficiency and the structure of the economy. Economies in the process of industrializing tend to have higher energy intensities than economies in their post‐industrial phase. On a smaller scale, for example, energy intensity can relate the amount of energy consumed in buildings to the amount of residential or commercial floorspace.

Gross domestic product (GDP): The total value of goods and services produced by labor and property located in the United States. As long as the labor and property are located in the United States, the supplier (that is, the workers, or for property, the owners) may be either U.S. residents or residents of foreign countries.

Heating degree days (HDD): A measure of how cold a location is during a period of time relative to a base temperature, most commonly 65°F. HDD are used in energy analysis as an indicator of space heating energy requirements or use. We compute this measure for each day by subtracting the average of the day's high and low temperatures from the base temperature (65°F), and negative values are set equal to zero. We sum each day's HDD to create an HDD measure for a specific time period.

You can find more definitions in our Glossary.

Methodology Used in this Analysis

With the exception of Figures 2 and 3, whose methodologies are described below, we publish the data in this report either as values in our MER or the result of calculations based on published values (for example, CO2 emissions per unit of energy consumed [MMmt CO2 per Btu]).

Methodology for Figure 2

Figure 2 (Trends in energy-related carbon dioxide emissions and key indicators) shows how several key factors of energy-related CO2 emissions, as determined by the Kaya identity, change over time and how these changes influence total energy-related CO2 emissions. These key factors are:

  • GDP per capita
  • Population
  • Energy intensity
  • Carbon intensity

Each of these factors is either directly published in the MER (for example, population) or is the result of taking the ratio of two published series (for example, GDP per capita). To put the factors’ growth or decline into context, we index them to 1990. In Figure 2, we divide the yearly value of each factor by its value in 1990 and multiply it by 100 to put it into percentage terms. Each point in the series represents the magnitude of each factor relative to 1990, expressed as a percentage.

Methodology for Figure 3

Figure 3 (Change in energy-related CO2 emissions by Kaya identity component) shows the change in total energy-related CO2 between 2019 and 2020 for the key emissions factors that make up the Kaya identity:

  • GDP per capita
  • Population
  • Energy intensity
  • Carbon intensity

The product of each component in the Kaya identity results in total energy-related CO2 emissions.

We calculate the changes in total CO2 emissions from each factor by taking the difference between total energy-related CO2 emissions in 2020 (that is, the product of each factor’s 2020 value) and a hypothetical total emissions value (that is, the product of each factor’s 2020 value except for the factor of interest, which uses its 2019 value). For example, the total change in energy-related CO2 emissions between 2020 and 2019 resulting from changes in GDP per capita is given by:

(GDPPC2020·Pop2020·EI2020·CI2020 )-(GDPPC2019 ·Pop2020·EI2020·CI2020)

where

GDPPC = GDP per capita;

Pop = population;

EI = energy intensity; and

CI = carbon intensity.

These values do not sum to the total change in emissions because of interactions among the changes in each component.

Table 3. Rates of change in Kaya identity components, 2019–2020
Parameter 2019–2020 percentage change
GDP per capita (GDP/population) -3.9%
Population 0.4%
Energy intensity (Btu/GDP) -3.9%
Carbon intensity (CO2/Btu) -4.1%
Source: Table created by the U.S. Energy Information Administration (EIA), based on data from EIA’s Energy Intensity and Carbon intensity; GDP per capita, U.S. Bureau of Economic Analysis and U.S. Census Bureau; and Population, U.S. Census Bureau

Methodology for Table 1

We divide total energy-related CO2 emissions for each end-use sector into two components: primary (or direct) emissions and indirect emissions. Primary emissions are CO2 emissions resulting from fossil fuel combustion in each sector (for example, natural gas used for home heating). Indirect emissions refers to emissions created by electricity generation, which we attribute to each end-use sector based on its share of total electricity consumption. Row 7 shows changes in total energy-related emissions for each end-use sector, row 4 shows changes in primary emissions, and row 1 shows changes in indirect emissions.

The first row in Table 1 represents the total change in electricity-related CO2 emissions by end-use sector between 2019 and 2020. We break down the change in electricity-related CO2 emissions into two components: the change in the carbon content of consumed electricity and the change in the total amount of electricity consumed. In other words, we express the total change in electricity-related CO2 emissions as

∆Elec.emissionss,y = ∆Carbon elec.(s,y) + ∆Demand elec.s,y        (1)

where

s = an end-use sector; and

y = the year.

We calculate the second term, the change in electricity-related emissions associated with changes in electricity demand, by multiplying the previous year’s electricity-related emissions in that sector by that sector’s change in electricity consumption

∆Demand elec.s,y = Elec.emissionss,y-1 · % change elec.con.s,y        (2)

This value represents the change in total electricity-related CO2 emissions in the sector, assuming that the electricity generation mix and carbon content are held constant. If this assumption is true, then the change in demand-related electricity CO2 emissions will be equal to the total change in electricity CO2 emissions. However, the electricity mix is usually dynamic over time. To account for this potential discrepancy, we define the change in emissions associated with the carbon content of electricity by reordering Equation 1 as

∆Carbon elec.s,y = ∆Elec.emissionss,y – ∆Demand elec.s,y        (3)

We calculate primary emissions for each end-use sector by subtracting indirect emissions from total emissions, or

Primary emissionss,y = Total emissionss,y – Elec.emissionss,y        (4)

We then break down these primary emissions into carbon content and demand components in much the same way as electricity-related emissions. The demand component of changes in primary emissions is expressed as

∆Demand primary energys,y = Primary emissionss,y-1 · % change primary con.s,y        (5)

and the change in emissions associated with the change in carbon content of primary energy is calculated as

∆Carbon primary energys,y = ∆Primary emissionss,y – ∆Demand primary energys,y        (6)