On the campaign trail and in office, President Trump and his advisers have railed against countries that exploit the United States by selling us a lot more stuff than we sell them -- that is, countries with whom we have big trade deficits. The criticism has led them to pursue a confrontational trade strategy, espousing taxes and potentially tariffs that would, at least conceptually, make U.S. exports cheaper and foreign imports more expensive.
Yet David Autor, a widely respected economist at the Massachusetts Institute of Technology who actually has done much of the most groundbreaking research on how trade can kill jobs, says that if the president wants trade to do more to benefit American workers, he should take a different approach.
Autor is perhaps best known for his calculations about how China’s entry into the global trading market had a devastating effect on U.S. manufacturing communities. I spoke with him about Trump's trade philosophy and what he would suggest the new administration do to help workers.
This interview has been edited for length and clarity.
What is the biggest thing you think readers should know about the administration’s trade policy?
Their views are quite mercantilist, meaning they tend to view trade as a zero-sum game, where if you export more than you import, you win, and if you import more than you export, you lose. And that’s an antiquated and unsophisticated understanding of trade.
Why is that unsophisticated?
Economists have understood since the Victorian era that the main benefits of trade come from comparative advantage, the idea that people can specialize in what they’re good at, and then benefit from exchange. The principle is no more mysterious than specialization in the labor market. There’s a reason I write articles and go out for good dinners, because I’m better at research than cooking. And there are people who are much better at cooking than research, so it’s mutually beneficial for us to specialize. Why should every country in the world have a crappy airplane manufacturing facility when they can buy jets from Boeing and Airbus?
The ultimate example of the opposite of trade is a subsistence economy, where every household grows its own food, raises its own animals, chops down its own trees and so on. But of course, as societies get more sophisticated, they subdivide work. We do that within countries as well as across countries, because some countries have expertise or cost advantages. The U.S. tends to export high-tech goods because we have strong comparative advantage there, and we tend to import labor-intensive and less skill-intensive goods that other countries can do more cheaply.
So how does the administration’s trade discussion contrast to that?
Their measure of our failure in trade is the trade deficit with China and Mexico. That’s just not an appropriate metric of success. Mexico is one of our biggest export markets. The fact that in some years we’re importing a few percentage points more from them than they are from us, it’s really second order relative to the benefits of trade.
People tend to think about trade as if it’s competition between companies — if Apple wins, Google loses. But that’s false. Trade makes nations better off in general. Now, I want to be clear. I’m not saying that everything about trade is good and beneficial. Trade also has costs. But starting from the point of view where trade is a game where one country wins and the other loses, and the measure of your victory is how much you export relative to what you import, that’s just crazy. You have to understand the fundamental reason for trade is not because you win by exporting and lose by importing -- it’s that there are things that other countries make that you want to buy, and things you make that they want to buy.
By the way, there is almost no evidence that NAFTA [the North American Free Trade Agreement] was substantially harmful for U.S. workers. That myth has been promulgated by people from Ross Perot to Pat Buchanan to Donald Trump, but there is not any academic support for it.
So what are the costs of trade?
Trade almost necessarily grows the size of the economic pie, but it also changes the size of different slices. It’s quite possible for trade to increase the size of the pie by a few percent, and yet shrink some slices by 20 to 30 percent. Because we’re a high-skill nation, when we trade with the rest of the world, we increase our production of skill-intensive products. So trade tends to increase the earnings of highly educated and skilled workers, and decrease the earnings and employment opportunities for less educated and less skilled workers.
The gains to consumers and high-skilled workers are sufficiently large that they could compensate those that are worse off for their losses. But that process doesn’t happen on its own. And in fact, it almost never happens in reality.
So that’s the real cost, that trade can dramatically reduce the livelihoods of a subset of people, and that is also extremely visible and wrenching. One reason is that manufacturing is geographically concentrated, so when a sector starts to go into decline, everyone in a region loses their job simultaneously, just like what happened with coal mining. It’s like a small bomb going off in your downtown.
Meanwhile, the benefits are quite diffuse. They often mean lower prices for consumers, a greater variety of goods, more demand for highly educated workers, and huge profits for certain companies. When we all get cheaper furniture and textiles at Walmart, that’s a couple percentage points in your cost of living over the next few years.
A second factor is that manufacturing employment in particular is relatively high-paid work for relatively less-educated workers. They earn more per hour than similar people, and they have more stable hours. They generally are not going to find equally good jobs.
We know in general as the labor market has become more skill intensive, women have educated themselves and adapted by moving quickly into other jobs. No one spends a lot of time shedding tears about the loss of all those great clerical jobs, but it is the case that clerical jobs have dramatically contracted. Women have moved on and up.
Whereas for men in manufacturing, there has not been nearly as strong of an educational response. When men are displaced from manufacturing, they tend to move into lower paid jobs, or just move out of the labor market. So they really are losing something they’re not going to replace in any short-term way.
Thinking about trade now, there's some confusion over how much people should question the orthodoxy of free trade. It seems like you’ve done a little bit of that yourself, yet your description of how trade works is still very much economics 101.
Here’s the irony: Orthodox economics has always acknowledged that trade could be very disruptive. But that wasn’t the advice given to policymakers. They were told that trade makes every nation better off — which is true — but not the caveat that it could create real challenges for certain individuals.
Why is that? Because historically we had very little evidence of those disruptive effects. What we learned over the last 20 years is that what theory always said, that trade can be very redistributive, was true in spades. The shock of China’s entry into global manufacturing was unprecedented in the disruption it created for U.S. communities. It created much more hardship than anyone anticipated.
In the work I’ve done with my co-authors, we’re not making the hunky dory case for trade. We emphasize you need to start by understanding that there are many potential benefits, but then you also need to complete the sentence and say that the process can be very redistributive, and if you’re not cognizant of that, people can get really hurt, as they did in the 2000s in particular.
So I don’t not want to be on the record saying it’s all good, and everyone who complains it just a troglodyte. That’s not correct.
You’ve mentioned that if we knew then what we knew now, we might have lowered trade barriers more slowly. But is that a case for raising trade barriers now?
If we knew how disruptive this would be, we would have wanted to do it more slowly. The reason the rate of change matters is there is a natural ebb and flow. People retire, and young people don’t enter contracting sectors. If we know that 20 years from now, no one will be driving motor vehicles for a living, we can adjust to that. If it’s announced that after Monday, there will be no one driving cars for a living, that would be very disruptive. So even knowing the destination, the rate of change really matters.
But secondly, no, I don’t think we can put the genie back in the bottle and expect to get back a lot of labor-intensive manufacturing. Some of that would have eventually succumbed to trade or automation, because it wasn’t an area where the U.S. had comparative advantage anyway, as a high-wage country.
That doesn’t mean you can’t aggressively enforce trade agreements. There are certainly cases where countries engage in dumping, flooding the market with their products and bankrupting their competitors. That probably happened in the case of steel, tires and solar cells. The WTO [World Trade Organization] gives us a lot of leverage. There have been several hundred cases brought in the last 15 years, and many of them resolved in the U.S.’s favor.
The TPP [Trans-Pacific Partnership] was a step in this direction, of creating enforceable contracts and strengthening intellectual property protections. China wasn’t in it. There’s no country that was happier that the TPP was shot down than China. And by the way, there was no country that was more harmed by that than Japan, with the U.S. being second. Trump's decision to withdraw from the TPP was an incredibly short-sighted and theatrical thing to do.
What about strategic sectors? If trade barriers aren’t the answer, is there anything we can do to preserve industries we see as important?
There are reasons to care about certain sectors. We ought to be really concerned that we don’t lose the robotics sector to China, for example. People are afraid of robots, but whether we produce them or not, they will be here. It’s just a question of whether they will be produced by the U.S., Germany or China. Since a lot of the innovation has originated here, we have an intellectual head start, but our investment is woefully behind.
Manufacturing is also extremely important as a hub of innovation. It may not be a mass employer in the future, but it is still incredibly important because so many of our great ideas are produced while making new products, and a lot of our wealth comes from those ideas.
One thing the U.S. ought to do is change our tax code in a way that favors domestic investment rather than simply favoring companies moving profits offshore. So, for example, the idea of replacing our payroll tax with a value-added tax, sometimes called a border adjustment, is a good one. We have an antiquated tax system that has very high, inefficient corporate taxes. Companies spend a lot of energy getting around them, and that creates huge distortions.
Another big step would be for the U.S. to stop cutting the legs out of its investment in research and development by gutting our science and statistical agencies. The U.S. was by far the leader in R&D expenditure and now we’ve fallen behind Germany. China is catching up rapidly.
I wanted to ask you about automation. Some research claims as much as 80 percent of the jobs lost in manufacturing are due to automation. Do you agree?
If we’re talking from 1980 to the present, automation has certainly been a bigger factor in the decline of manufacturing employment than trade. But from 2001 to the present, I think trade is at least equally important. The other thing is that trade is much faster moving, while automation tends to be more gradual.
Automation will occur whether we lead or we lag on it. Automation could benefit us in two ways. One is by increasing productivity. The other is that if we develop those technologies, the industries that grow up around them will create profit and wealth in the U.S. So we ought to be thinking of automation as a strategic asset. As much as we recognize that, just like trade, automation doesn’t make everyone better off, it does raise national wealth, and that gives us the scope to make everyone better off.
What advice would you give the new administration in terms of labor force policies?
I think steps the Obama administration was taking in terms of worker protections, safety enforcement, wage and overtime laws, and so on were important. I think it’s incredibly important to retain access to healthcare and health insurance.
But there’s no substitute for economic growth, so I do want to see pro-growth policies like tax reform and investment. We should be investing in infrastructure and education. The way that public education institutions are being decimated is a huge national loss. And we would benefit from government investment in R&D, for example the National Science Foundation or the National Institutes of Health.
There is a big role for the government in facilitating growth, and it’s not just about deregulating and letting dirty industries pollute more, and letting companies give bad financial advice to pensioners. Growth isn’t a matter of letting one group of people take advantage of another. It’s choosing policies that make the system work better.
You might also like: