Spotlight Cancún: Why Do U.S. States Emissions Vary So Widely?

Elizabeth A. Stanton, Guest Blogger
Another in a series from the Triple Crisis Blog and the Real Climate Economics Blog on the Cancún Climate Summit. The following is based on a recent E3 Network paper, “Why Do U.S. States Emissions Vary So Widely?” by Elizabeth A. Stanton, Frank Ackerman, and Kristen A. Sheeran.

Much of the U.S. resistance to ambitious global efforts to reduce carbon dioxide emissions reflects a fear common amongst Americans that high emissions are necessary to maintain high standards of living. While the examples of Belgium, Demark, Germany, Ireland, Japan, and the United Kingdom – countries where emissions per capita are roughly one-half of the U.S. average – demonstrate that lower per capita emissions are consistent with high living standards, many Americans remain unconvinced that the same standard of living can be produced with varying emissions levels. This is true despite the fact that some of the best evidence for this can be found within the U.S.: individual states vary only modestly in average incomes, but have widely differing per capita emissions.

If U.S. per capita carbon dioxide (CO2) emissions were equal to those of its most populous state, California, global CO2 emissions would fall by 8 percent. If, instead, U.S. per capita emissions equaled those of Texas, the state with the second-largest population, global emissions would increase by 7 percent. What makes Californians’ emissions so different from those of Texans, and from U.S. average emissions? And are the factors that explain these differences amenable to replication as policy solutions?

In our recent paper, we analyze the variation in energy-related per capita CO2 emissions among U.S. states, seeking to explain why some states have much lower per capita emissions than others. Our analysis begins with reported data on emissions for the 50 U.S. states and the District of Columbia from standard government sources. We focus on energy-related carbon dioxide emissions, which account for the great majority of greenhouse gas emissions and are one of the sectors most likely to be regulated under climate legislation. Emissions from electricity generation are attributed to the sectors where electricity is used – residential, commercial, industrial, and transportation. We adjust reported emissions data for interstate electricity sales, and then calculate the fractions of each state’s emissions that come from household activities:  transportation, household fuel use (primarily heating and cooking), and household electricity use. The range from highest to lowest states, excluding Alaska, varies about 3 to 1 in transportation and residential emissions.

The 3-to-1 interstate variation in carbon emissions from household activities is the result of many factors, some more controllable than others. People who live in rural, low-density states drive more than those who live in urban, high-density areas, resulting in more transportation emissions. Some parts of the country are colder than others, and face greater heating requirements; some are hotter, and need more energy for cooling. These factors are difficult or impossible to change. Public policy can provide incentives for improved insulation and more energy-efficient heating and cooling units for residences and businesses; but even the best efficiency programs cannot eliminate the need for heating in colder states, and for air conditioning in hotter, humid states. Note that the District of Columbia, New York, Oregon, Rhode Island, Vermont, and Washington, all colder than the national average, are among the states with the lowest total transportation and residential emissions per capita.

Regression analysis shows that climate (heating and cooling degree days and average annual humidity) and geography (inverse population density) are important but incomplete explanatory variables for transportation and residential emissions. In the regressions that included variables that could be more readily influenced by policy, per capita income was only significant in the case of heating fuels, the smallest of the three categories of emissions we examined. The lack of correlation between income per capita and transportation and electricity emission per capita demonstrates that, at least among states of the U.S., there is no rigid relationship between affluence and emissions. Similar incomes can be associated with very different levels of emissions. It is possible – as evidenced by the contrast between California and Texas – to enjoy the typical American lifestyle with per capita emissions that are widely divergent from the U.S. mean.

Energy prices (average gasoline and retail electricity price) and public infrastructure (share of working population commuting via public transportation) were important in explaining the interstate variation in the use of transportation fuels and residential electricity and are readily addressed by climate and energy policies. The level of gasoline taxes differs widely from state to state; the extent of public transportation differs as well. Both of these policies have a direct, measurable effect on automobile usage and thus on transportation emissions. The reliance on coal power for electricity generation, where fuel choice is a function of existing infrastructure, has large impacts on residential emissions. So too does reliance on oil for home heating. Both factors are amenable to change via public policy. Electricity tax rates, more direct, government-sponsored conservation initiatives, and development of alternative energies can have far-reaching impacts on home electricity consumption and its emissions.

What does our analysis say about the difference between per capita emissions in California and Texas? Transportation emissions are almost one and a half times as great in Texas as in California. Texas is more sparsely populated, it has lower gasoline prices, and 2 percent of its labor force commutes via public transportation, as compared to 5 percent in California. Per capita emissions from electricity are also four times as large in Texas as in California. With lower electricity prices and triple the heating degree days, Texans use twice the electricity per capita of Californians; 37 percent of Texans’ electricity is generated from coal compared to 15 percent in California. Public policy can’t make Texas more densely populated (in the short or even medium run) or cooler, but it could promote public transportation, increase gasoline and electricity prices, and shift electricity production towards less carbon intensive alternatives to coal.

The data analyses provided in this report offer only a partial explanation. There is much more to be learned from a detailed examination of the policies of the lowest- and highest-emitting states. States with the lowest per capita transportation and residential emissions (New York, District of Columbia, Oregon, California, Rhode Island, Washington, Vermont, and New Hampshire) are by no means the poorest in the nation; indeed, these states are all above the national average in per capita income. These high-income, low-emissions states demonstrate that it is possible to produce a comfortable American lifestyle with carbon emissions well below average. Following their example more widely is an important first step on the road to reducing our greenhouse gas emissions to a sustainable level.

Elizabeth A. Stanton is a Senior Economist at the Stockholm Environment Institute.