Climbing the economic ladder can happen in two ways: you can move up the ranks relative to your parents (relative mobility) or you can benefit from economic growth that boosts everyone to a higher rung without changing their relative position. In a society in which relative mobility is not especially high, growth, or absolute mobility, takes on added importance. Thus, the recent slowdown in U.S. growth is more than a little distressing.
What’s going on? First, labor force growth has been anemic because the population is aging and because even among working-age adults, men are dropping out of the work force while women’s participation has leveled off after many decades of strong growth.
Even more discouraging are labor productivity trends. Output per hour, which grew at 3.3% for most of the postwar period (1948 to 1973), has now slowed to a crawl (1.3% over the 2004-2015 period). A new paper by our colleagues Martin Baily and Nicholas Montalbano attempts to explain the slowdown, analyzing aggregate and firm-level data and drawing from recent research by experts in the field. They discuss three possible explanations for the lethargic growth in US productivity: (1) mismeasurement of productivity; (2) declining rates of innovation; and (3) a widening gap between the most and least productive firms.
All three appear to have played a role. That said, we are in uncharted waters. In their fascinating new book, The Second Machine Age, Erik Brynjolfsson and Andrew McAfee argue that we are experiencing an inflection point. The current slowdown, they think, is only the calm before the storm. New applications of exponentially growing computing power combined with artificial intelligence or machine learning will revolutionize everything from medicine to driving. Robert Gordon, in contrast, believes that our best days are behind us. In his equally impressive book, The Rise and Fall of American Growth, he argues that recent innovations pale in comparison to the steam engine or electricity. He likes to ask his audiences “which would you rather have: a cell phone or indoor plumbing?”
One can remain agnostic about the future, but one finding in particular is worth more serious attention. A portion of the slowdown can be attributed to notably slow growth in the service sectors, especially in health care and education, which together account for nearly 25 percent of GDP (OECD 2015). They share many inefficient qualities in common: fee for service instead of pay for performance, third-party payment mechanisms, and entrenched institutional and professional interests that tend to be resistant to change. Add to these problems the difficulty of measuring quality improvements in education and health services and it’s easier to understand the lackluster performance of these two sectors.
We need to do a better job of measuring productivity growth in the health care and education sectors. Which medical treatments actually lead to better health? When is an educational credential just a credential and when does it actually signify a higher level of knowledge and competence? And what is each of these worth? But the problems go well beyond the need for better measurement. Until we learn to use technology more effectively in these two sectors (e.g. electronic medical records, computer-based diagnostic tests, computer-assisted classroom instruction, and massively open online courses) and reorganize the two sectors around these promising technologies (e.g., more paraprofessionals in health care, fewer teachers lecturing a classroom), low productivity in this large and growing share of the economy will continue to drag down our growth rate. Reorganizing these sectors would not only improve near-term growth, but by producing more human capital, it could also generate stronger long-term growth.