US productivity growth is shrinking, and nobody knows why

The United States is making less with more. Needless to say, that’s an ominous sign for economic growth.

According to the US Bureau of Labor Statistics, for the first time since the global financial crisis, multifactor productivity growth was negative in 2016. Multifactor productivity is the most basic measure of how well the economy is turning inputs, like wood and steel, into outputs, like tables and cars. After growing each year from 2010 through 2015, the latest data show that, in 2016, productivity declined by 0.2%.

The productivity decline of 2016 is part of a long term trend. Economist Robert Gordon has shown that US productivity growth slowed from a pace of about 3% per year between 1930 to 1970, to an average of about 1% since—excluding a brief spurt in the late 1990s and early 2000s.

Economists don’t agree on why this is happening. A new report by the McKinsey Global Institute gives three possible explanations for the “productivity puzzle.” The report focuses on the US, but McKinsey believes that America’s problems are a good proxy for the developed world—Western Europe and Japan have also experienced productivity slowdowns.

One explanation is that we may not be measuring productivity correctly. Some scholars argue that productivity growth is much harder to assess in a modern service-based economy than it was when manufacturing dominated. For example, McKinsey estimates that about 27% of the benefits from free products like Google search, Facebook, and Skype, are not accounted for in productivity measurements. Still, even the largest estimates for mis-measured productivity would only explain a small part of the decline.

Another possible explanation is that businesses haven’t been investing in becoming more productive. McKinsey points out that, even with historically low interest rates, in the past decade the share of US domestic investment to GDP fell to its lowest point in more than 60 years. Spending on new machinery and technology has a direct relationship to productivity growth. Researchers at the Brookings Institution believe this could explain about half of the productivity slowdown.

Finally, there are those who believe that the productivity slowdown is really an innovation slowdown. Robert Gordon believes that the swift pace of productivity growth of the early 20th century was mostly due to the momentous innovations of the industrial revolution and the advent of electricity. He doesn’t see any of today’s inventions producing a similar effect.

The analysts at McKinsey disagree. They see huge opportunities for growth from new technologies like smartphones and the internet of things, but believe that the private sector that has not taken full advantage. What is unclear is whether companies have not harnessed the power new technologies because of structural issues, like bad management or perverse incentives, or if there is just a lag, and a productivity boom is coming soon.

Whatever the reasons, the productivity slowdown is unquestionably bad for economic growth. As aging populations shrink the working-age population in developed countries, the majority of GDP growth comes from increases in productivity. The big GDP growth numbers US president Donald Trump has promised are unlikely to appear unless this trend reverses itself.