The fall of coal, and its pollution-linked deaths, is boosting the economy

Many economic activities create what are called “externalities”: costs that aren’t accounted for in their products but are paid for by society at large. Pollution is a major source of externalities, as it can lower the value of property, force people to spend money on medical costs, and even lead to early deaths.

Air pollution is estimated to have caused more than 100,000 early deaths in 2016. Most of these have come due to what are called fine particulates, which are small particles that can be readily inhaled and cause issues like stroke, heart disease, and lung ailments. So a group of researchers at Carnegie Mellon University decided to do an economic analysis of the issue for the United States. The researchers compared the costs of premature deaths from particulate pollution to the value added by the economic activity that produced the pollution to find out which polluting industries might provide a net benefit to the economy.

Their analysis showed that the electricity sector had recently caused more in particulate mortality costs than its direct contributions to the economy. But that has now changed thanks to the drop in coal use, which has left farming as the only major activity that generates more costs than its direct benefits.

Some dismal science

Obviously, it’s not possible to ascribe every single death in the US as being either caused by particulate pollution or not. But various organizations have made models that can take a given amount of pollution and project how many premature deaths it will cause, based on epidemiological studies of smaller populations. And the EPA has good estimates of a variety of sources of particulate pollution, ascribing annual numbers to a variety of industries and even specific activities within industries.

The research team—Peter Tschofena, Inês Azevedo, and Nicholas Muller—has obtained pollution data from the EPA, census data, mortality rates from the CDC, and financial data from the Bureau of Economic Analysis, all broken down at the county level. The researchers fed that data into a model that calculated two things: the contribution of individual industries to the GDP, and the economic damages caused by that industry through the deaths it caused by production of particulate pollution.

We should note that this is a severe underestimation of the true costs of pollution. Particulates themselves cause lots of economic damages due to increased medical costs and the loss of productivity that’s associated with health issues. And many forms of economic activity produce non-particulate pollution, like the mercury emitted through coal burning or the groundwater contamination that can be caused by industrial activity. Still, by focusing on a single source of economic costs, the three researchers lower the overall uncertainty in their analysis.

The data they used runs from 2008 to 2014 and, so, allows them to track trends across this time period. The biggest and most obvious is that particulate pollution has dropped dramatically. In 2008, they estimate that particulates subtracted just over a trillion dollars in economic damages from the economy. By the end of their study period in 2015, that figure had dropped to $790 billion. As a fraction of the total economic activity, that represents a drop from about 6% of the GDP in 2008 to about 4% in 2014.

Death of coal brings life

What has caused this huge drop? Put simply, coal. Particulate production is dominated by four sectors of the economy: agriculture, power generation, manufacturing, and transportation. Everything else accounts for only a quarter of the known sources. Agriculture and manufacturing were relatively stable over this period, while transportation saw a significant drop, primarily through cleaning up diesel emissions from trucks, rail, and boat transport.

But the biggest change was clearly in the power sector, where economic damages dropped from about $300 billion to $150 billion in that period. That’s almost entirely attributable to lower levels of sulfur dioxide particulates, which can be attributed to the shuttering of coal plants.

This has had a dramatic effect on the overall contribution of electrical generation to the economy. In 2008, the authors estimate that the economic damages from deaths due to particulates were actually higher than the value added to the economy by electrical generation. (While, as noted above, this doesn’t include costs due to other forms of pollution and non-fatal costs, it also doesn’t include the economic activity that having a supply of electricity enables.) By 2014, the situation was radically different, with electrical production adding substantially more value than it costs through particulate mortality.

Transportation also saw a considerable drop, with the economic costs per dollar of value added to the economy falling by 70% over this period. This has the effect of making agriculture the only major sector of the economy that causes enough of this form of economic damage to more than offset the value it adds to the economy. It’s now also the single largest source of fine particulate emissions in the US economy.

Agricultural emissions are complicated, as they include particulates from livestock and fertilizers, ammonia emissions, particulates released by the equipment used, and more. But the particulate emissions of crop production have actually gone down substantially, but this drop has largely been offset by a rise in particulates emitted by livestock. Despite this, emissions have dropped slightly, and the agricultural sector has continued to expand. Combined, these have allowed the costs to approach the break-even point with the value it adds to the economy.

Some uncertainties

The authors suggest that this sort of analysis will become more challenging going forward. With most major sources of particulate pollution cleaning up their act, the large and poorly constrained “all others” sector of the economy will become a more significant player. This will include sources like residential housing, office complexes, small businesses, and more. Since each of these sources is, on its own, difficult to measure, most of the information we have about them comes from aggregate estimates and have much larger uncertainties than something like a power plant, where pollution is regularly monitored.

As these costs are definitely a low estimate of the total externalized cost of these industries, it’s also not clear how best to approach the overall issue: that many US industries probably contribute more in costs than they contribute to the economy.

“We certainly do not advocate for closing industries in such cases,” the authors caution. “Rather, this is an indication, within the context of macroeconomic aggregate statistics, that damages are likely inefficiently high.” How to identify and correct those inefficiencies, however, was left as an exercise for the reader.

PNAS, 2019. DOI: 10.1073/pnas.1905030116  (About DOIs).

Correction: the summary of the results at the end of the intro has been modified to more accurately represent the study’s conclusions.