A group of distinguished trade economists have quietly released a paper estimating that if Chinese imports had grown half as quickly over the past 15 years, Hillary Clinton rather than Donald Trump would be preparing to move into the White House right now.
There’s reason to be skeptical about this specific result, since it focuses on regions harmed by trade with China and doesn’t factor in benefits enjoyed by people elsewhere in the country. Nevertheless, it underscores just how big an impact trade with China has had on the American economy — and on our politics.
For decades, experts have argued that freer trade is good for the US economy and downplayed the economic harms that trade can cause. On the campaign trail, Donald Trump did the opposite, railing against trade with Mexico and China and promising to stop the decline of the manufacturing sector.
To the surprise of many experts, Trump won. And new research suggests that Trump knew exactly what he was doing when he made trade a central theme of his campaign.
In recent years, a group of economists led by MIT’s David Autor have been studying the economic effects of Chinese imports on American communities. Their research demonstrated that Chinese imports have had significant and long-lasting economic impact in towns where local manufacturers have faced competition from Chinese imports.
“Manufacturing employment has been really important to non-college-educated workers,” Autor says. “And I do think its decline has had real effects on their earnings.”
More recently, they’ve found that Chinese imports have had significant political effects. In 2010, communities that were hard hit by Chinese competition tended to elect more extreme candidates to represent them in Congress.
And those same communities saw the strongest swing toward Donald Trump in the 2016 presidential race. Indeed, the economists estimate that if Chinese imports had grown half as quickly over the past 15 years, Hillary Clinton would have won Wisconsin, Michigan, Pennsylvania, and North Carolina — and hence the presidency.
“The foundation for this sharp division on the lines of trade was put in motion in the 2000s, long before the presidential election,” Autor says. “You really see this fracturing between areas that were trade-impacted and not impacted.”
This does not, of course, prove that expanding trade with China was a bad idea. Lots of Americans in other parts of the country benefited from lower prices as well as from jobs created by exports to China. But it does pose a challenge to the complacency of many economists on the issue. In theory, the government can compensate the losers so that everyone winds up better off. But Autor’s work — and the populist backlash that helped put Trump in the White House — suggests the US hasn’t done enough to help trade’s losers.
Chinese imports have devastated many US communities
Autor worked with economists David Dorn and Gordon Hanson on a 2013 study that looked at how rising Chinese imports during the 2000s affected specific regions of the United States.
The US economy is complex, so nailing down cause and effect isn’t easy. Autor and his co-authors developed a clever strategy to try to isolate the effect of rising Chinese imports. They assembled data on the growth of Chinese imports in a bunch of different industries — narrow categories like “luggage,” “rubber and plastic footwear,” and “die-cut paperboard.” They also collected data on where competing US manufacturers in each category were located.
By cross-referencing these two data sets, the researchers were able to compare towns where the local industry faced strong competition from Chinese imports with other manufacturing towns where Chinese imports were not a major threat. The results were striking. Not only did Chinese trade create big job losses in towns most affected by Chinese competition, but the effects lasted a lot longer than many economists expected.
In other words, economists had overestimated the flexibility of the US labor market. Workers who lose their jobs don’t necessarily bounce back and find new work in other industries as conventional economic theories predicted. A lot of them become persistently unemployed, eventually retiring early or going on disability.
And because these manufacturing job losses are heavily concentrated in specific towns, they have larger social effects. “When a city loses manufacturing jobs, it doesn't just lose manufacturing jobs,” says economist Enrico Moretti. “There are additional job losses outside of manufacturing because the factory jobs provided the wealth for local services. If you look numerically, the biggest losses are outside manufacturing.”
People who run stores, restaurants, and other local businesses suffer as their customers have less money to spend. The town needs fewer teachers, doctors, and construction workers. A declining tax base strains the budget of the local government.
And these changes can also have profound personal consequences. A recent study by economists Justin Pierce and Peter Schott found that counties whose local industries were more exposed to Chinese competition saw modestly higher mortality rates. The authors speculate that losing jobs and health care coverage led to deteriorating health for some affected workers.
How seriously should you take these findings? Economics isn’t an exact science, but a lot of economists find the methodology used in these studies to be compelling. Lots of factors have contributed to the general decline in manufacturing employment across the US economy in general. But the industry-by-industry and town-by-town correlation between imports and job losses suggests that trade with China really has played a causal role.
Chinese imports helped make Congress more polarized
In a follow-up study published this year, Autor, Dorn, Hanson, and Kaveh Majlesi found that trade with China has also had a big impact on US politics. They compared voting patterns in congressional races between 2002 and 2010. And as in their earlier research, they compared areas where local industry was hard hit by Chinese competitors against areas facing less Chinese competition.
The result: Congressional districts that were more exposed to Chinese imports were more likely to elect ideological extremists to office than districts that were less affected by Chinese imports. Voter anger about the economic effects of trade helped drive the Tea Party wave that put Republicans in control of the House of Representatives after the 2010 election.
Congressional districts that were represented by a moderate Republican in 2002 were more likely to be elected by a conservative Republican by 2010 if the district was heavily exposed to Chinese imports. (Legislators were classified as liberal, moderate, or conservative using a standard metric based on members’ roll call votes.)
Congressional districts that were initially represented by a moderate Democrat were more likely to be represented by either a liberal Democrat or a conservative Republican eight years later if they were heavily impacted by Chinese imports. Among these trade-affected, initially Democratic districts, majority-white districts were more likely to elect conservative Republicans, while majority-minority districts tended to elect Democrats who were further to the left.
This wasn’t primarily because formerly moderate members of Congress shifted to the right over the course of the 2000s. Rather, a lot of moderate legislators from 2002 were out of Congress by 2010 — replaced by more conservative (or more liberal) members.
And often this was the result of political instability within these districts: a moderate Republican would lose his seat to a Democrat in 2006 or 2008, and the Democrat, in turn, would lose her seat to a different Republican in 2010. And because the Republican Party is becoming more conservative overall, the new members of Congress tended to be more conservative than the old ones.
“Growing import competition from China has contributed to the disappearance of moderate legislators in Congress, a shift in congressional voting toward ideological extremes, and net gains in the number of conservative Republican representatives, including those affiliated with the Tea Party movement,” Autor and his co-authors write.
The researchers say that between 2002 and 2010, the fraction of moderates in Congress — Democrat as well as Republican — declined from 57 percent to 37 percent.
It would be a mistake to conclude from this that trade with China is the cause of growing ideological polarization in Congress. This trend has been underway since the 1970s. But trade with China seems to have accelerated the process.
Chinese trade boosted Trump in Wisconsin, Michigan, and Pennsylvania
Earlier this month, Autor and his collaborators crunched the numbers and found a similar pattern in the 2016 presidential race. They compared voting results between 2000 and 2016, examining how much better (or worse) Donald Trump did in various parts of the country than George W. Bush had done 16 years earlier. They found that Trump enjoyed the biggest gains — relative to Bush’s 2000 results — in areas that were most strongly affected by Chinese imports. And the effect appears to be large enough to have tipped the election to Trump.
They estimate that in an alternative universe where Chinese import gains between 2002 and 2014 were 25 percent smaller — and Rust Belt job losses had been correspondingly less severe — Donald Trump would have lost Wisconsin by 1.7 percent instead of winning it by 0.8 percent. He would have lost Michigan by 1.7 percent instead of winning it by 0.3 percent.
They further estimate that if Chinese imports gains had been 50 percent smaller, Trump would have lost North Carolina and Pennsylvania as well as Wisconsin and Michigan. And flipping those four states would have been enough to throw the Electoral College to Hillary Clinton.
In short, China’s rising economic might provides a powerful explanation for the surprising shift of Michigan, Wisconsin, and Pennsylvania — states that had previously gone Democratic in every presidential election since 1992 — into Donald Trump’s column. Trump’s anti-trade rhetoric and his general attitude of economic pessimism seem to have resonated particularly strongly with voters in these states. And it especially resonated in parts of those states where local manufacturers had faced strong competition from Chinese imports.
As the authors themselves acknowledge, it’s a mistake to put too much stock in the specific figures here. Many things would be different in a hypothetical world where Chinese imports grew half as fast. Most obviously, people in other parts of the country would not be enjoying lower prices or jobs producing goods for export to China, and those benefits are not factored into the analysis. So if Chinese trade had been smaller, Clinton might have done better in the Rust Belt — but worse in states that have thrived in the era of globalization.
What the research does make clear, however, is that Donald Trump’s anti-trade rhetoric was tapping into widespread voter sentiments that had been largely ignored by conventional politicians. Chinese imports had negatively affected the economic circumstances of a lot of voters in Michigan, Pennsylvania, and Wisconsin, creating a fertile environment for Trump’s brand of economic populism.
Trade’s harmful effects won’t be easy to reverse
If Chinese imports devastated certain communities in the United States, an obvious response would be to reverse the policies that allowed Chinese imports into the country in the first place. Donald Trump has proposed as much, calling for a 45 percent tariff on Chinese imports.
But economists say this isn’t a good solution — for two big reasons.
One is that trade really has been good for the United States as a whole, even if it’s been bad for certain communities. Consumers across the US have benefited from lower prices and greater product variety. US companies have benefited from exporting their goods to the Chinese market.
And the US and Chinese economies have become increasingly interconnected, so disrupting trade between the two countries could cause a lot of economic damage. Some American manufacturers rely on components or raw materials from China, and China is a big export market for many US companies. A trade war with China would harm a lot of US businesses — and the employees that work for them — as well as Chinese firms.
The larger issue, though, is that even if we were willing to accept that economic pain, reintroducing trade barriers wouldn’t necessarily reverse the damage to communities that have lost jobs over the past 15 years. Higher trade barriers might generate some new manufacturing jobs, but it probably wouldn’t create those jobs in the same Rust Belt towns that have lost jobs over the past 15 years. They’d be more likely to locate in Sunbelt towns with nice weather, growing populations, and business-friendly labor laws.
And many of those jobs wouldn’t come back at all.
“A lot of the jobs that we lost to China were somewhat vestigial,” Autor says. He argues that without Chinese imports, “we wouldn't have lost them so quickly, but we would have lost them over the next 15 to 20 years. I don't think we're going to get them back.”
In short, if companies were forced to shift production back to the US, they’d likely build factories with more robots and fewer human workers.
The conventional solution to the disruptive effects of trade is to offer trade assistance — government programs that offer laid-off workers job training, relocation benefits, and other services to help them get back on their feet.
However, says economist Enrico Moretti, “it's not always completely feasible to help the affected workers. If you are 50, and you've been working at a factory all your life, it's really hard to retrain you.”
One thing that probably would have helped — and might still help — would be for the Federal Reserve to do more to support demand across the economy in general. Over the past 15 years, Chinese have imports sharply reduced demand for goods manufactured in certain US towns, but since 2008 that has been combined with a persistent demand shortfall across the US economy as a whole.
Weak domestic demand has made it hard for towns hit by Chinese competition to nurture new businesses in other industries that are less exposed to trade. And it has also made it harder for workers who lose their jobs in Rust Belt towns to find work in nearby cities. Having your town’s biggest factory shut down is going to be a grim situation regardless of the broader macroeconomic context, but it’s much worse for it to happen when the broader economy is suffering from a sluggish recovery.