June 6, 2023
Paul Krugman- NYTIMES Opinion
Back when Donald Trump began his political rise, it was common for mainstream pundits to attribute his support to “economic anxiety,” to suggest that MAGA was an understandable, maybe even reasonable response to deindustrialization and the loss of jobs in the American heartland. You don’t hear that very much anymore.
But it’s true that Trump himself was obsessed with trade deficits and that if he indeed had any unorthodox policy ideas — in practice, he was mostly a standard, tax-and-benefit-cutting Republican — they were focused on attempts to revive manufacturing. That, at least, was the main rationale for the trade war he started with China in 2018.
As it turned out, Trump had no visible success in promoting manufacturing. But a funny thing has happened under his successor: Suddenly, investment in manufacturing has surged. What Trump’s trade policies didn’t achieve, President Biden’s industrial policies have.
The numbers are stunning. Here’s an annotated chart of manufacturing construction spending:
The Trump tax cut of 2017, which was sold as a way of promoting U.S. investment, didn’t have any visible effect. Neither did the trade war, which kicked off in earnest in mid-2018. But under Biden, manufacturing construction, as some people put it, has gone parabolic, more than doubling just over the past year.
Raw dollar figures can be misleading, especially when you’re looking at the long run; spending needs to be compared with the size of the economy as a whole. But the Census, which provides these numbers, has annual data going back to 1993; here’s manufacturing construction as a percentage of G.D.P. over the past three decades (the number for 2023 is April spending as a percentage of first-quarter G.D.P.):
It’s still really impressive. And there’s no real question about the causes of the surge. It’s being driven by two major pieces of legislation: the misleadingly named Inflation Reduction Act, whose actual core is subsidies for green energy, and the CHIPS Act (“creating helpful incentives to produce semiconductors” — call in the acronym police!), which is supposed to protect national security by promoting domestic production of, um, chips.
The ultimate impact of these policies will almost surely be much bigger than these numbers suggest. For one thing, planning and beginning work on new manufacturing plants takes time, so there’s probably even more spending in the pipeline. For another, these numbers count only construction — basically, factory buildings. Filling those buildings with machinery and investing in R&D to make the most of the new capacity will probably add hundreds of billions to the total business spending.
Why are Biden policies producing a manufacturing revival but Trump policies didn’t? Well, Trump’s trade policy was simply incompetent: Because it raised tariffs on industrial inputs as well as consumer goods, it raised costs and may well have reduced manufacturing employment. And the Trump tax cut was based on the belief that if you let corporations keep more of their profits, they’ll invest the money rather than use it to, say, buy back shares; this belief was proved wrong.
Biden’s industrial policies, by contrast, are largely focused on creating demand for U.S.-manufactured products, for example by subsidizing the purchase of electric vehicles. And business investment, while far less sensitive to tax rates than legend has it, is very responsive to demand.
And so we’re having a huge manufacturing revival.
Now, there’s a risk that what I’m saying may come across as too uncritically upbeat. So let me offer two major caveats about the Biden manufacturing boom.
First, even if we do have a major manufacturing revival, we’re not going back to 1970, when more than a quarter of U.S. workers were in manufacturing. We’re still going to be overwhelmingly a service economy despite these new policies. The new manufacturing boom may help lagging regions in the U.S. heartland and is specifically designed to help workers without college degrees. But nobody should expect it to turn back the clock on our transition to a postindustrial society.
Second, rapid growth in a sector isn’t necessarily a good thing for the economy. Until recently, for example, there was an explosion of resources devoted to Bitcoin mining. As far as I can tell, these resources produced nothing of value — sorry, crypto enthusiasts, Bitcoin has yet to show that it’s useful for anything besides money laundering. And the Bitcoin boom has both inflicted environmental damage and consumed resources that could have been used to produce things that are actually useful.
So why should we consider Biden’s industrial-policy-driven manufacturing boom a good thing? Mainly because it’s part of an urgently needed transition to renewable energy that may be our last chance to avoid climate catastrophe. And the surge in U.S. manufacturing investment in particular partly reflects protectionist aspects of the legislation that are a bad thing in terms of economic efficiency — but were essential to the political deal-making that made it possible to tackle climate change at all.
The point, then, is that the success of Biden’s manufacturing policy can’t be judged purely by the extremely impressive investment numbers. Still, the policy would clearly have been considered a failure if it hadn’t produced a manufacturing boom. So it’s good news that the boom is happening, indeed exceeding even the most optimistic expectations.