Looking for the roots of total factor productivity growth

To paraphrase the legendary football coach Vince Lombardi, productivity isn’t everything when it comes to economic growth; it’s the only thing. That might be a bit of hyperbole, but economists today agree that the efficiency of the production of goods and services given levels of capital and labor in an economy—its total factor productivity— is the key determinant of the pace of economic growth. Unfortunately, they have yet to figure out exactly what drives growth in total factor productivity, a critical task if we want to boost growth in the long run.

Let’s first take a quick look at the roots of total factor productivity. In two papers published in 1956 and 1957, Robert Solow, now an emeritus professor of economics at the Massachusetts Institute of Technology, showed that the source of long-run economic growth per capita wasn’t increased savings or faster population growth but rather a residual that he labeled “technology”. Calling something a “residual” usually doesn’t make your audience think that it’s the most important thing in the world, but Solow eventually won the Nobel Prize for this work—and the vast majority of economists today agree that this technological residual, now commonly known as total factor productivity, is the key source of long-run growth.

Economists who study economic growth have long been trying to understand the source for differences in total factor productivity. In a recently published working paper, Stanford University economist Charles Jones details a number of facts about economic growth including the role of total factor productivity growth. Jones cites one calculation that 80 percent of the growth in economic output per person since 1948 is due to such growth. And the difference in output per person across countries is very strongly related to differences in total factor productivity.

Jones breaks down total factor productivity into two factors. The first is the stock of knowledge in an economy, which is probably what most people think of when they hear the word technology. A larger stock of knowledge helps economic growth as individuals know how to best use labor and capital to boost economic growth. The second factor is what Jones calls “M.” One possible interpretation of “M,” he says, is that it stands for “misallocation.” Higher total factor productivity growth could be the result of less misallocation in the economy. Jones cites a study that look at how less discrimination against women and black workers in higher-level jobs increased economic growth in the United States as just one example of his M factor,

But “M” might also stand for the “measure of our ignorance,” as Jones points out. Indeed, the possible sources of total factor productivity growth are myriad: misallocation, better government and labor market institutions, “culture,” and a variety of others. In short, M could well stand for Mysterious.

If the concept of M were not daunting enough, economists are also looking at how they measure and conceive of productivity itself. Dietz Vollrath, an economics professor at the University of Houston, points out that the current categorization system for economic firms is tilted toward better understanding of manufacturing firms. The system was created when manufacturing was a larger portion of high-income countries. The result of this better data has been that most studies looking at productivity look at manufacturing firms. In today’s economy that misses out on the majority of economic activity in the services sector

Ryan Decker, a graduate student at the University of Maryland, agrees and adds that data aren’t the only problem. The concept of total factor productivity, at least at the firm level, was conceived for understanding manufacturing firms. It might be difficult to graft that concept onto service-sector firms. As Decker puts it, “The further you get from producing widgets with machines, the harder it is to map the TFP concept to the real world.”

The possibility that the most studied and discussed productivity concept might be flawed is discouraging at best, but perhaps the flaws Vollrath and Decker identify aren’t that large, which would be encouraging. But then economists would still have the task of understanding what actually drives the growth of total factor productivity. Let’s hope that task doesn’t take the economics profession as many centuries as it did before Solow detailed his “residual.”