Three years ago, in my book An Extraordinary Time, I advanced the view that governments have little power to make their economies grow faster over the long run. Economic growth, beyond the next year or two, depends mainly on productivity growth. Drawing on the experience of the 1970s, I asserted that governments cannot manage this in any predictable way; there have been periods when productivity improved quickly, usually when no one expected it, but slow productivity gains are the normal state of affairs.

Predictably, this viewpoint was not very popular. On the right, the usual suspects insisted that fewer regulations and lower tax rates, especially on capital, would usher in a new age of higher productivity and faster economic growth. On the left, I heard objections that I was condemning workers to stagnant living standards; if governments would spend more on education or infrastructure or research or something, productivity might skyrocket as it did in the 1930s and the 1950s.

I’ve been reexamining this issue, looking at the evidence that has come in since An Extraordinary Time appeared. Unfortunately, it seems to support my case.

In the United States, the much-touted 2017 tax reform, which mainly reduced taxes on business, seems to be doing nothing for productivity. Last month, in its latest economic outlook, the Congressional Budget Office projected that total factor productivity in the business sector, which takes account of capital investments and improvements in workers’ skills as well as output per worker hour, will grow about 1.1 percent per year over the coming decade, about the same miserable rate as in the 1970s and far more slowly than in the quarter-century between 1982 and 2007. The combination of this performance and slow population growth will leave the U.S. economy will be hard-pressed to grow at 2 percent annually, CBO found. That is far more slowly than U.S. politicians of either party claim is possible.

This is not merely a domestic trend. The OECD, the research organization of rich-country governments, projects that labor productivity in the wealthy economies, on average, will be a minuscule 4% higher in 2020 than it was in 2010. Some countries, mainly in Eastern Europe, are projected to do much better. Others, though, are seeing no productivity growth at all, which will make it difficult to improve workers’ wages. The Conference Board, a business research organization, finds that productivity growth is slowing in the major emerging economies as well, holding down their growth potential.

Is there any be cause for optimism about a productivity revival? Perhaps. An interesting article published last year suggests that slow growth in productivity in U.S. manufacturing may be related to outsourcing; the authors, economists at the Bureau of Labor Statistics, suggested that “industries that shift their production process toward greater use of intermediate purchases may be doing so at the expense of innovation.” We’ve seen some evidence over the past several years — predating President Trump’s attacks on trade — that manufacturers have been shortening their supply chains and bringing a greater share of their production in-house. If the BLS economists are right, this could stimulate innovation and hence faster productivity growth in the manufacturing sector.

Manufacturing, though, now accounts for only a small part of the economy in most wealthy countries. It’s no sure bet that faster productivity growth in manufacturing would provide enough of an economic boost to give workers the higher living standards they expect government to deliver.