INTERVIEW: Manufacturing Myths

By Reuvain Borchardt

Economist Don Boudreaux, a professor of economics at George Mason University and senior fellow at George Mason’s Mercatus Center, discusses the state of American manufacturing.

An adjunct scholar at the Cato Institute, Boudreaux previously served as president of the Foundation for Economic Education, and taught law and economics at Clemson University and was a visiting scholar at Cornell Law School.

Boudreaux has written two books: Globalization (2008), on the effect of international trade; and Hypocrites & Half-Wits (2012), a collection of letters to the editor he has penned to various media outlets against politicians, pundits, and professionals who contravene his libertarian views on free trade, low taxes and minimal government regulation. He blogs at Cafe Hayek.

He earned a PhD in economics from Auburn University (1986) and a J.D. from the University of Virginia (1992). He has lectured around the world and been published on topics including antitrust law, monetary policy, and international trade. 

—

A popular talking point among members of both major political parties is hand-wringing over a retreat in American manufacturing. What is the state of manufacturing right now in the United States? 

As it happens, I was just researching that today for a chapter of a book that I’m writing with Phil Gramm [a fiscally conservative former U.S. senator known for sponsoring legislation that rid the banking industry of some regulations]. And the state of manufacturing is pretty good. 

It is simply false to say that manufacturing has been “hollowed out” in the United States. The standard claim is that we manufactured a lot of stuff right after World War II, and then — starting in the ’70s according to some people, and when China joined the World Trade Organization in 2001 according to others — we lost all or at least most of our manufacturing capacity. Trump often says we don’t make things anymore. It’s simply untrue. 

Manufacturing output in America as of February 2024 is only 6% below its all-time high, which it hit in December of 2007, just before the Great Recession. And, of course, as is always true, when a recession hits, output goes down, especially in the manufacturing sector. And so output has grown since the end of the Great Recession, and today, it’s just about 6% below that all-time high. So in 2007 manufacturing was 21% higher than when China joined the WTO. And today, it is 164% higher than it was in the mid ’70s, the last time America ran a trade surplus. 

Since the end of World War II, inflation-adjusted manufacturing output — except in recessions — pretty much steadily rose. And it didn’t decline when we started running trade deficits or when the GATT became the WTO in 1995 or when China entered the WTO in 2001. It didn’t decline when NAFTA kicked into effect, either. 

Its first real decline came with the Great Recession. And as everyone remembers, that was a slow recovery. Around 2015, it finally returned close to its pre-recession level. Since then it’s largely been flat. There was a huge dip with the pandemic, but it went right back up. So it’s largely been flat for about nine years. But this has not been a period where America was pursuing freer trade. It’s an era when protectionism rears its head.

Nevertheless, manufacturing today is only 6% beneath it’s all time high, measured in inflation-adjusted dollars of manufacturing output. If you look at another measure, called “manufacturing value added” — the amount of economic value that American manufacturers add to what they produce — that today is just .05% off its all-time high that it hit in 2021. 

And if you look at industrial capacity — the capacity to produce things — that, too, is near an all-time high. So this claim that we don’t make things anymore and that our manufacturing capacity has been hollowed out is simply false. There’s no data to support it. 

The reason people think it’s true is because what we don’t make anymore is final consumer goods. That’s where you get the “Made in” labels that we see on so many of our products: “Made in China,” “Made in Turkey,” “Made in Canada,” etc. The “Made in” labels mean where the final assembly of products occur. Americans now manufacture pretty high-end things, like industrial equipment, machine tools, construction equipment, and components of things that are assembled abroad. So we do a lot of manufacturing. But when people go shopping at Target or Walmart, or when they open their Amazon packages, they just see the “Made in” label and think they are seeing just things made abroad.  

American manufacturing is quite healthy: America is the second-largest manufacturing economy in the world. It was the largest until 2010, when China took it over. Now America is in second place — but China has four times the number of people America has, so on a per-capita-basis we’re still the largest manufacturing country in the world. On a value-added basis, we are by far the most successful manufacturing country in the world. 

And investment continues to pour into the U.S.

You’ve also argued in your writings that another reason for what you say is a false narrative about a manufacturing decline is that the number of manufacturing jobs is in fact shrinking — but that’s because we’re producing more with fewer workers, which means our productivity is increasing.

The absolute number of Americans employed in manufacturing peaked in June of 1979. The number in terms of a proportion of the workforce hit its peak back in 1944. 

So, as more and more workers move out of manufacturing and into the service sector, the proportion of the workforce employed in manufacturing is much, much lower than it was in the past. And the main reason is that productivity has increased. 

The same thing that happened to agriculture in the 19th century and in the first half of the 20th century is now happening to manufacturing in America and in most of the rest of the world. Output per worker has skyrocketed over the years. It’s leveled off in the past five or six years, a slight stagnation. But the long-run trend since the end of World War II — and even before that, but the good data go back only until then — is a dramatic increase in manufacturing worker productivity. 

So the jobs are being lost because not as many workers are needed to produce the goods?

Yes. We can produce more output with far fewer workers now.

A very well known and competent study was done in 2015 and revised in 2017 by Michael Hicks and Srikant Devaraj of Ball State University.

They found that 88% of manufacturing job loss was due to technology improvements. Less than 12% was due to trade. Some of it was due to changes in consumer tastes; as we get wealthier, we don’t just buy more stuff, but spend more of our money going on vacation or buying more medical services. And, of course, that’s not counted as manufacturing.

So the vast majority is productivity improvements. And no serious person wants to argue that productivity improvements are bad.

The average American seems to think we don’t make things any more. Do you attribute that notion to labor unions — who care about jobs more than productivity — trying to get government to enact union-friendly policies?

I suspect some of it is pushed by unions.

But a lot of it is pushed by manufacturing firms themselves. After all, if U.S. Steel can convince people that America doesn’t make things anymore, their fellow Americans will be more open to having the government protect firms like steel producers and other manufacturers from foreign competition, or to even to subsidize them. 

And a lot of this is also fueled by economic ignorance. Politicians like Trump on the right and Bernie Sanders on the left can play on the economic ignorance of people.

It’s not a crime that the average person doesn’t look at economic statistics — folks like me do that. People are going about their daily lives, and it seems plausible to them, given what they experience in their daily lives, that we don’t make things anymore. So when you have a politician come along and say, “We don’t make things anymore; put me in power and I’m going to make sure we start making things again,” they become more likely to vote for that person. 

So some of it is unions, some of it is manufacturing firms, and a lot of it is just opportunistic politicians playing upon the fact that most people aren’t adequately informed about the true state of the economy. 

By the way, these statistics that I mentioned to you are very easy to find. These are all government-gathered statistics, all published in prominent places like the Bureau of Economic Analysis. The St. Louis Fed has this great data site called FRED Data. These are all publicly available data. These statistics are not mysterious or controversial. 

But you’ve got to look for them, and that takes a little bit of knowing how to work your way around the internet.

You mentioned that there are protectionists on the right like Trump and on the left like Sanders. Prior to Trump, Democrats were more associated with unions and protectionism while Republicans were associated with free trade, but since 2016 it seems Republicans have become more protectionist. Do you think protectionism will remain in the Republican Party going forward regardless of who wins the presidential election this year?

No question about it. 

But keep in mind that Republicans historically have been the party of protection while Democrats were the party of free trade, going back to the early days of the Republican Party in the mid-19th century. And the Smoot-Hawley Tariff Act was signed in 1930 by Herbert Hoover, a Republican president.

That only started to change in the postwar period. By the way, Ronald Reagan talked a better game on trade than he played, due to political compromises. But certainly, from the 1970s through George W. Bush, the Republicans had a legitimate claim to be regarded as the party most favorable to free trade, and it was the Democrats, largely because of their union connections, who were the party of protection.

Now, unfortunately, one of those parties — the GOP — is returning to its protectionist roots, but the other is not returning to its free-trade roots. Historically, one party was the party for protectionism and the other was the party for free trade. Now, for the first time since at least the Civil War, both major parties are explicitly and enthusiastically for protectionism.

One of your biggest pet peeves is the notion that a trade deficit is a bad thing. Arguing against that notion is kind of your life’s work; can you distill that argument here briefly?

Another name for what we today call a trade deficit is a capital-account surplus, or simply “capital surplus.”

A trade deficit means that we import, during some periods, more goods and services, measured in dollars, than we export. If foreigners want to invest in America, they need dollars. If they spend all their dollars buying from us the same amount of dollars they earn selling to us, they have no dollars left to invest. So if month after month, year after year, we were to export as much as we import, foreigners wouldn’t have any dollars left to invest because they would have spent all their dollars buying our exports. 

Therefore, when we run a trade deficit today, practically speaking — it doesn’t have to mean this, but it does mean this in practice — foreigners are investing those dollars in the United States. And why we should be upset at that is a mystery to me. We don’t get upset when our neighbors across the street save and invest their dollars. We applaud that; we think such saving is prudent and we know that it helps the economy. And it helps the economy no less if our neighbors across the ocean save and invest in the American economy. 

So the very fact that foreigners are choosing to invest in the American economy is at least a signal that foreigners believe that, compared to other investment options around the globe, America still looks pretty good. It’s not a sign that we’re losing, that someone’s cheating us, or  that we’re no longer “competitive.”

Furthermore, when the investments come here they help to strengthen our economy — and that’s even when those investments are loaned to the government, as lots of them are. Foreigners don’t make Uncle Sam borrow money. I wish Uncle Sam would borrow a lot less money — but to the extent that non-Americans lend some of their dollars to Uncle Sam, that reduces the burden on Americans of having to fully support Uncle Sam’s fiscal incontinence. And it thus releases resources here to be invested in businesses, research, and development, and worker training. 

That’s the practical effect of a trade deficit: more investment in America. We should applaud that, not bemoan it.

People like Trump argue that America is harmed when other countries sell us cheap goods. What exactly is their argument?

Their main argument is that it takes away our jobs. This is an argument that protectionists have used for two and a half centuries. 

But it’s not true. When foreigners sell us stuff, some particular jobs, in industries that compete with imports, are surely destroyed — but other jobs are created. Because foreigners do one of two things with their dollars: they send them back to us by buying our exports, which creates jobs in industries that produce exports, or they invest those dollars in America. And when they invest those dollars in America that, too, creates jobs.

When Ikea builds a store in New Jersey, that creates construction jobs. And when it operates that store, that creates jobs. If Nippon Steel [a Japanese company currently seeking a takeover of U.S. Steel]  is allowed to purchase U.S. Steel, it’ll pour dollars into upgrading the facilities at U.S. Steel, making the company more productive, and so jobs there are protected. 

You can scour the historical record, and you will find no evidence that increased trade — or decreased trade, by the way — has any effect on the long-run rate of unemployment. Trade doesn’t affect the level of employment. Trade affects where employment occurs. When you have protectionism, you have more jobs in inefficient industries and fewer in efficient industries. When you have free trade, that moves jobs out of inefficient industries, and into industries that are more efficient. And when you have resources and workers moving out of inefficient industries and into efficient industries, you increase the productivity of the economy, including the productivity of the workers — thereby increasing wages and living standards.

It seems that people often feel government should protect American jobs from being “shipped overseas,” as opposed to a job being lost because certain goods or services are no longer needed. Far fewer people seem to be bothered by, for example, telephone operators losing their jobs because we no longer need them, than by an American job moving to another country.

Statistics from 2000 — and it wouldn’t have changed much since then — show that Americans spend roughly 11 cents out of every dollar on goods or services that are directly connected with trade. We’re a gigantic country, and most of the economic activity that takes place is internal. The vast number of jobs that are lost in America are lost to purely internal economic changes. You used the example of telephone operators. The example I always like to use is, back about 25 years ago when Americans became fascinated with the Atkins diet, Krispy Kreme shut some donut shops. They blamed the Atkins diet — and correctly so. 

The vast majority of jobs that are lost have nothing to do with trade, but rather to purely internal factors, like changes in consumer taste, technology, and changes in demographics. Like, Americans have fewer babies today than it did in the past, so we need fewer Americans making diapers. 

I once wrote a piece looking at American job loss and creation for roughly eight years following the Great Recession — these stats are available from the Bureau of Labor Statistics — and this was during a period in which there were no recessions. The average number of jobs lost every month in America during that time period was 1.7 million.

We have an enormous amount of job churn in the U.S.; jobs are being destroyed and created all the time. And because we’re such a large and dynamic country, trade happens to play a relatively small role in that. The share of job losses during that time period that I studied that could plausibly be attributed to America’s trade with China was about 1%. So you add in other countries, maybe 5% or 6% or 7% of job losses can be attributed to trade.

But that gets all the headlines. 

Yeah. Because, first of all, we measure trade more carefully than we measure internal things. We’re not counting every month the number of restaurants that shut down or have layoffs, but we have these trade statistics: “Oh look, we imported more than last month, oh look, the trade deficit went up last month. This must be bad news.”

We have — and I think it’s a good thing — an enormously dynamic labor market, in which jobs are being constantly destroyed and created every month. 

As I’m sure you know, on the first Friday of every month at 8:30 a.m. Eastern Time, the Bureau of Labor Statistics releases the monthly jobs report. One of my pet peeves is how misleading that number is.

In a normal, non-recession month, the news reporter might announce, “There were 200,000 jobs created last month.” But that’s a net figure. That probably means something like 2.1 million jobs were created and 1.9 million were lost. But they just give you the net figure — so people don’t understand the incredibly large amount of job churn we have in the U.S.

At the beginning of our discussion, you mentioned that manufacturing improved after the Great Recession but it’s remained largely flat the past nine years. And you said, “But this has not been a period where America was pursuing freer trade. It’s an era when protectionism rears its head.” I just want to clarify: Are you saying that protectionism is bad not only for the overall economy, but specifically for manufacturing as well? 

It depends how it’s carried out, and it depends on how other countries react. I haven’t studied in enough detail to tell you exactly why manufacturing output has been largely flat for the last nine or so years.

Protectionism, if it’s carried out correctly, can certainly raise the outputs of particular firms or industries. In the case of Trump’s tariffs, I think they were pretty slipshod. Here’s one thing we do know: Trump was protecting steel and aluminum. Now, we are so efficient at producing steel and aluminum that the number of steel and aluminum workers in America is minuscule. It’s only in the hundreds of thousands. But we have many, many more workers — something like 34 times as many — working in industries that use steel and aluminum as inputs. And so if you make steel and aluminum more costly by putting tariffs on them, you are actually harming manufacturing throughout the economy.

And this is not even to mention retaliation by other countries. 

Regarding manufacturing remaining flat the last nine years — how much of that is a COVID effect? 

COVID affected everything. And I think we’re still so near the COVID era that when I look at data now, with very rarely exceptions, I stop at the beginning of 2020. So manufacturing was actually only flat for five years before then. 

Regarding the pandemic, during lockdown it’s easy for college professors and journalists to work at home. But it’s almost impossible for manufacturing workers. So you got this dramatic decline in manufacturing output. And it recovered much faster than it did after the Great Recession, which was a very slow recovery.The five-year period, pre-COVID, of manufacturing being flat, I don’t know that that’s a long enough period for which any conclusions can be drawn. And because COVID happened right after it, the fact that manufacturing output today is roughly where it was 10 years ago, that could be due to the lingering consequences of the COVID lockdowns. 

—

rborchardt@hamodia.com

—

This interview first appeared in Hamodia Prime magazine.