Science isn’t done by guesswork or gut instinct. It requires expertise not only to conduct but to evaluate; in-depth research in a field outside of my own is often beyond my ability to critique. I don’t have the knowledge to review a paper on molecular biology, although I might notice a really blindingly obvious flaw.
I have more knowledge of economics than I do of molecular biology. Even so, it’s not my primary field of expertise, so when I saw a recent post by American Enterprise Institute scholar Mark J. Perry, I was a little confused. His “Bottom Line” includes this: “The goal of America’s energy sector isn’t to create as many jobs as possible… we want the fewest number of energy workers.”
Is there something I’m missing? Is AEI actually saying that jobs are bad?
Don’t we want jobs?
Now, the basic economic argument behind Dr. Perry’s argument is that labor carries a cost, and producers of any sort of good seek to keep their costs down. This is a story as old as civilization. As agriculture improved and we needed fewer hands to work on the farms, people moved to the towns and cities and produced new goods and services. In the Industrial Revolution, machines reduced the number of workers needed to produce a given quantity of goods. Luddite rioters smashed a few machines in protest, but more were built, and society adapted.
In the coal sector, automation cost jobs well before the shale gas boom made its impact felt. It’s important to recognize that even if the economy as a whole benefits by replacing people with machines in repetitive tasks, in the near term the individuals who were doing those tasks—coal miners, but also people in many other fields, such as factory workers, cashiers, truck drivers, even accountants and paralegals—will be adversely affected by losing their jobs. The displaced workers and their communities will face economic, social, and health impacts. We need to plan for these shocks better than we have in the past.
Human labor is not simply a fungible, tradeable commodity like a ton of steel or a bushel of wheat; it is far more complex. Human beings have feelings, needs, and rights. The field of “labor economics” examines this topic in more detail.
Coal and solar jobs
Dr. Perry argues that solar’s greater reliance on human labor is a flaw, not an asset. And in doing so, he makes a very basic mathematical error.
The solar industry employed, depending on the source, 260,000 to 374,000 workers in 2016. Solar power produced 56.22 million megawatt-hours according to the U.S. Energy Information Administration (Dr. Perry mistakenly includes a much lower figure). Meanwhile, coal employed about 160,000 workers and produced 1240 million megawatt-hours. Therefore, solar needed about 35-50 times as many workers per unit of electricity produced.
But here’s the major error: the solar workers in 2016 installed systems that will generate power for decades. Or were researching new solar panel chemistries that might pay off decades in the future. The labor of coal workers in 2016 was largely fueling and operating existing power plants for only that year. On an amortized basis of labor-per-kilowatt-hour, the two technologies would be a great deal closer. By a rough estimate, it would be a ratio of about two to one.
There are also differences in costs per worker. Coal has higher labor costs per employee, when you factor in executive bonuses, hazard pay, and the pension and health benefits that unions secured for workers in decades past. These benefits are necessary because of the health risks faced by coal workers. Solar doesn’t pay badly, but coal mining was historically a very good-paying if risky job, as my colleague Jeremy Richardson points out. Factor that in, and the gap in labor costs per unit of electricity shrinks even further.
So is that the end of the story? No. Labor isn’t the only input to production. Industrial output can be modeled with something called a Cobb-Douglas Production Function. This equation states that production requires multiple inputs, like capital, labor, and materials, and that these can be substituted for one another to a greater or lesser degree.
Since the unsubsidized costs of new solar power have now fallen below those of new coal power, if solar power uses more labor, then it must be making more efficient use of capital and materials (in dollar terms) than coal power. Solar gives more people jobs, uses less other stuff in the process, creates less pollution, and comes out ahead.
The value of solar jobs
Is the labor dependence of solar power a bad thing? Not for the men and women who are actually working in the field. Not for you and me and the other consumers, and our utilities who are buying solar power at record-low price levels. Not for a country seeking to create opportunities for its citizens.
In basic economics, prices tell the story. From that point of view, solar’s labor requirements are clearly not a disadvantage. When we consider that labor and jobs are actual people’s livelihoods, and not just numbers on a page, it becomes clear that the labor requirements are actually beneficial.