Democratic presidential nominee Hillary Clinton and Republican nominee Donald Trump during the second presidential debate Oct. 9 at Washington University in St. Louis, Mo. (Jim Lo Scalzo/EPA)

China’s emergence as one of America’s biggest trading partners over the past 15 years may have cost Hillary Clinton the election.

At least those are the findings of a hypothetical exercise by a noted group of economists who have studied how increasing competition from China has changed not just economic realities for Americans, but political life as well.

In a recent note, economists David Autor, David Dorn, Gordon Hanson and Kaveh Majlesi calculated that if Chinese import penetration had been 50 percent lower since 2000, Michigan, Wisconsin, Pennsylvania and North Carolina would have elected the Democratic instead of the Republican candidate. That would have been enough to put Clinton in the White House instead of President-elect Donald Trump.

The research showed how large the effect of trade and globalization — topics addressed time and time again by Trump during the presidential campaign — may have been on the election.

The economists have done work in the past on what the shock of China’s entrance into the global economy has meant for American workers. China, a massive country with one-fifth of the world’s population, had been largely closed off from the global economy after Communists took over in 1949, but started to reintegrate beginning in the 1980s and 1990s.

In 2001, China joined the World Trade Organization, dismantling many of the country's remaining barriers to trade, and creating a huge new source of competition for American manufacturers. Much of that effect has now been absorbed, but the political consequences have not.

In a previous paper, Autor, Dorn, Hanson and Majlesi showed that increasing trade with China had contributed to political polarization in the U.S. through the 2000s. Areas that had been more exposed to trade competition with China ended up less likely to elect moderate politicians to Congress, on both the right and the left.

“No matter how you sliced it, these trade impacts led to the removal of moderates. And so we speculated that we might see something similar in the [2016] general election,” Autor said.

In the current paper, Autor and his colleagues compared something called the “two-party vote share” — how many people voted for the Republican candidate, divided by the total number of voters — in the 2016 election and in the election of George W. Bush in 2000, before China’s integration into global trade. They examined election results for 2,971 counties and compared those with how Chinese imports affected industries and jobs in those areas between 2002 and 2014.

They found that Republicans gained votes in counties that were more exposed to Chinese competition, with a one percentage point increase in average Chinese import penetration in a county leading to a 2.09 percentage point gain in the Republican two-party vote share compared with 2000.

“Areas that were more exposed had a significantly greater increase in the Republican two-party vote share, and because the election was pretty close it seems to have mattered a fair amount,” Autor said.

Autor pointed out that the political changes were probably not “just about trade policy, per se. We think it feeds into a ... sense of nationalism, the sense that the American way of life is potentially threatened, and that a certain type of employment and a structure that went with that is going extinct,” Autor said. These political trends are also related to the fact that economic prospects for non-college-educated Americans, especially non-college-educated white men, look dimmer now than they did 30 years ago, Autor said.

The economists also carried out a fascinating exercise looking at how voting might have differed if China hadn’t become such a manufacturing powerhouse. The table below shows that analysis.

With 10 or 25 percent less Chinese import growth, Michigan and Wisconsin would have voted for the Democratic candidate in the 2016 election, but Trump still would have won the election with more electoral votes than Clinton, the economists found. But with 50 percent less import growth, they find the effect is strong enough to flip Pennsylvania and North Carolina as well, turning the election in Clinton’s favor.

The economists emphasized that this hypothetical scenario is “extremely restrictive.” If there had been no trade shock from China, the world would have been quite different, as would the 2016 election, the researchers say.

Yet, Autor said, trade with China did have a big effect on how people voted. “The places that were trade impacted did swing heavily toward Trump, relative to their votes in previous elections.”

You might also like:

Trump just announced he'd abandon the TPP on day one. Here's what happens next

A single chart everyone needs to look at before Trump's big fight over bringing back American jobs

Donald Trump's threats work better on some countries than others