When we talk about “secular stagnation,” we refer to a really big question – how long does it take the nation’s standard of living to double – one generation, two, or more? By “stagnation” we don’t mean a situation of no growth at all. Rather, we mean an epochal slowdown in growth that drastically lengthens the amount of time it takes for the standard of living to double. For the total U.S. economy, the benchmark was set during 1920-70, when the growth rate of labor productivity in the total U.S. economy was 2.9 percent per year. That means the standard of living doubled every 24 years.
What has happened since 1970? With the exception of a nine-year period between 1995 and 2004, the growth rate of labor productivity has slowed to 1.3 percent per year during the 49 years between 1970 and 2019. That means the standard of living now doubles every 53 years, more than twice as long as required before 1970. That is a profound slowdown, meaning that the doubling of the standard of living takes two generations rather than one. Instead of our children enjoying twice our standard of living, that privilege now awaits our grandchildren.
Two qualifications are important. First, these growth rate numbers refer to the total U.S. economy, not the three-quarters of the economy in the private business sector for which productivity measures are normally published. This is the right measure, for the standard of living depends on everything produced in the economy, not just in the private sector.
Second, I omitted 1995-2004, when productivity growth temporarily revived to 2.6 percent, a revival usually credited to the invention of the internet and a wave of computer-related “ICT” investment. And that contrast is meaningful, between the 50 years of 2.9 percent during 1920-70 and nine years of 2.6 percent during 1995-2004. In my analysis, the pre-1970 half century of rapid growth reflected the importance of the “Great Inventions,” particularly electricity (and all its spin-offs) and the internal combustion engine (which made possible not only motor vehicles but also air transport). In contrast computer-related ICT investment was a pale shadow, boosting productivity growth back temporarily to pre-1970 rates only for nine years.
Why has productivity growth slowed down by so much? A growing consensus has formed that the process of innovation has encountered diminishing returns. A group of Stanford economists showed that “New Ideas Are Getting Harder to Find,” documenting the multifold increase in the number of research workers needed to achieve continued miniaturization of computer chips, or to discover new drugs. Other research shows that an explosion in the number of new patents has been more than offset by the diminishing importance of these patents. The number of robots in manufacturing doubled over the past decade, but productivity growth in manufacturing was actually zero between 2010 and 2019. Much-heralded innovations like autonomous cars are like “waiting for Godot” as the tallies of deaths from Tesla auto-driving software keep mounting up. These examples support the interpretation that the bigger the stock of existing inventions, the tougher it is to find new ones.