Here’s a curious ritual of American politics: Whenever a large energy project is proposed, the ensuing debate revolves chiefly around how many jobs it will create.
For example, backers of the proposed Atlantic Coast Pipeline, a contested 600-mile natural gas conduit that would run from West Virginia to North Carolina, have maintained that it would create nearly 20,000 construction and manufacturing jobs.
Similarly, President Trump’s announcement last year that he was reviving the Keystone XL pipeline suggested that it would generate 28,000 construction jobs. On the other hand, The Washington Post concluded that the real number was 3,900 construction jobs, based on a State Department estimate.
Renewable energy advocates play the same game. An Environmental Defense Fund report published last year asserted that the United States’ clean energy sector employs an estimated four million to 4.5 million people. The solar and wind industries “are each creating jobs at a rate 12 times faster than that of the rest of the U.S. economy,” the report said.
The problem with statistics like these isn’t just their purveyor’s bias: It’s that all the numbers are inherently misleading. Regardless of political outlook, economists generally agree that job numbers don’t convey the value of a given project or sector. Severin Borenstein, a University of California, Berkeley, economist, could have been speaking for all of them in his 2015 blog post called “The Job Creation Shuffle.”
“Most of the people who take a newly ‘created’ job are leaving an existing job. Or would have found another job,” Mr. Borenstein wrote. “Then there are people who are displaced by the new jobs created — the coal miners who worked for the mine that is shut down” or “the fracking oil drillers in North Dakota laid off when cheaper crude from the tar sands is carried to market by the Keystone XL.”
The “fundamental fallacy” of counting jobs, Mr. Borenstein continued, is that all sorts of things, from government policies to variations in energy prices and taxes, affect employment. As a result, “reports of ‘green job creation’ or the ‘jobs that will be created by Keystone’ are just data cherry picking, not real analysis,” he wrote.
A deep recession justifies the sole exception to this rule: In 2008, joblessness was rampant and stimulus projects clearly led to increased employment. But at other times, Mr. Borenstein told me, the idea that projects “take people who have no job and now they have one is completely misleading.”
The right way to assess the worthiness of energy projects has nothing to do with counting jobs, Mr. Borenstein added. It involves doing cost-benefit analyses that include both the projects’ direct costs — the sort of thing that investors base decisions on — and their “externalities,” the costs and benefits to society such as air or water pollution or useful knowledge uncovered by research and development.
When fossil fuels are involved, a sophisticated cost-benefit analysis requires an additional calculation using what has become known as “the social cost of carbon”— an estimated number in dollars per ton of carbon dioxide that reflects the climate change consequences of a project’s emissions. This is an imprecise number, intended to take into account a deluge of woes, an array of constantly changing and sometimes unpredictable climate phenomena including rising temperatures and sea levels, intense droughts and floods, and hampered agricultural production. But the number, pegged by the Obama administration at $45 per ton in 2020, is a better approximation of climate change’s impact than zero, the default number if the calculation isn’t attempted.
Multiply $45 by the number of tons of a project’s expected carbon dioxide emissions and the result is its climate change cost, which should be included in its cost-benefit analysis. If that calculation were performed for all energy projects now, most existing coal-fired power would be considered uneconomical, while gas-fired power plants wouldn’t be significantly affected. Renewables would benefit.
On an individual level, the impact of the social cost of carbon might be modest: For instance, the average American car emits a ton of carbon dioxide every two to two and a half months, hiking the annual expense of driving a car by roughly $250.
But on a national scale, the quantities are enormous. Consider that in 2016 the United States emitted five billion tons of carbon dioxide, nearly a seventh of the global total of 36 billion tons. Adding the social cost of carbon to domestic fossil fuels’ other expenses would have driven up their cost by about $200 billion, enough to help hasten the economy’s conversion to renewable energy and reduce the depredations of climate change.
The idea of the social cost of carbon was discussed in scholarly literature for decades, but it didn’t become consequential until 2007, when the San Francisco-based Ninth Circuit Court of Appeals ruled that the National Highway Traffic Safety Administration had to take climate change impacts into account in devising automobile fuel-efficiency standards. From that point on, federal agencies have devised their own social-cost-of-carbon numbers to apply to regulatory decisions.
In 2009, the Obama administration drew representatives from a dozen government agencies and offices to create a working group that calculated a set of estimates for all agencies to use — eliminating the duplication and confusion arising from allowing the agencies to perform their own, inevitably conflicting calculations. The group’s estimates have been applied to scores of federal regulations since then, as well as the Obama administration’s now-stalled Clean Power Plan. They are also being used by Canada and several state public utilities commissions, including New York’s. Many corporations — oil companies like Exxon Mobil, the onetime climate change denier, among them — use some version of the social cost of carbon to plan investment.
The Trump administration would probably have liked to disregard the social cost of carbon entirely, but it is already too firmly entrenched in law and regulation for that. Instead, two months after taking office, President Trump issued an executive order that disbanded the working group and returned the job of calculating the social cost of carbon to individual agencies. Then, last October, the Environmental Protection Agency unveiled a drastically lowered social cost of carbon, ranging from $1 to $6 in 2020. Other agencies are likely to use similar numbers.
The administration argued, among other things, that the previous calculation erred by including the international impacts of climate change instead of just the national ones. But as Richard Newell, president of an environmental economics think tank called Resources for the Future, pointed out to me, carbon dioxide is a global pollutant that has impacts everywhere, regardless of where it’s emitted. And even if many of the impacts — for example, increased migration or economic or political upheaval — occur in other countries, they can affect the United States in unpleasant ways. Perhaps more important, the United States’ willingness to curb its emissions would encourage other nations to limit theirs, a necessary step toward reducing climate impacts here.
If reinforced by rollbacks planned by the E.P.A. in car fuel economy standards and other regulations, the administration’s distortion will have far-reaching consequences. Vehicles will burn more fuel, and appliances will be less energy-efficient. Proposed renewable energy plants will be disadvantaged compared with fossil-fuel plants (which will generate fewer jobs than predicted). Most ominously, climate change will worsen.
The Trump administration’s daily-soap-opera gyrations attract more attention, but its attempt to ignore global warming is more sinister. One way that business leaders, environmentalists and elected officials at all levels can fight back is to insist on considering the social cost of carbon in every energy decision. Honest accounting, it turns out, is a pillar of a clearsighted society.