Nonprofit, nonpartisan journalism. Supported by readers.

UCare generously supports MinnPost’s Second Opinion coverage; learn why.

Two data analyses identify health-related variables that may explain why certain counties voted for Trump

A study by The Economist found that health was a strong predictor of how a county voted in the presidential election.

The data suggest that the ill may have been particularly susceptible to Donald Trump’s message.
REUTERS/Aaron Josefczyk

Two separate analyses of county-level U.S. health data — one by a pair of American economists and the other by the editors at the London-based news magazine The Economist — offer interesting new theories into why the death rate for white, middle-aged Americans has increased in recent years and how that increase is associated with President-elect Donald Trump’s popularity among white voters. 

First, some background: A year ago, two Princeton researchers made headlines when they reported that, starting in 1998, the death rate for white middle-aged Americans — those aged 45 to 54 — reversed. Instead of continuing a decades-long downward trajectory, the rate began to steadily increase by about one-half percent per year.

That disturbing death-rate turnaround did not occur among black or Hispanic middle-aged Americans (their death rates continued to decline), nor did it happen among middle-aged whites in other wealthy countries.

The authors of the study said they weren’t sure exactly sure why the life expectancy of middle-aged white Americans (particularly men) had shortened, but their leading theory had three suspects: substance abuse (alcohol, prescription opioids and heroin), suicide and chronic liver disease.

An ‘economic shock’

But what triggered the substance abuse, suicide and liver disease?

Article continues after advertisement

In a study published in draft form last week, Justin Pierce of the Federal Reserve Board and Peter Schott of the Yale School of Management point to a specific “economic shock”: liberalization of trade between the U.S. and China.

The two economists argue that after President Bill Clinton and Republican lawmakers granted permanent normal trade relations to China in 2000, millions of workers in “factory” communities across the U.S. lost their jobs. Many of those workers never recovered economically and fell into depression and addiction, which led, in turn, to greater rates of suicide and drug overdoses.

Here’s a summary of the study by Slate reporter Jordan Weissman:

[It] shows that, compared to the rest of the country, counties where local manufacturing businesses were more vulnerable to Chinese competition saw their job markets deteriorate more severely, and experienced greater increases in deaths by suicide and “accidental poisoning” — a category that includes drug overdoses. They found “mixed” evidence of increased deaths from alcohol-related liver disease.

The rising mortality rates were concentrated among whites. Pierce and Schott generally found no relationship between the change in China’s trade status and suicide or poisonings among blacks or Asians, for instance. The impact was also stronger in communities where fewer people had gone to college. In other words, trade with China seems to have brought more despair and tragedy to working-class white communities.

The study also found, however, a health-related upside to the opening up of trade with China. As Washington Post reporter Max Ehrenfreund notes, “that exposure to Chinese competition significantly reduced the number of heart attacks, possibly because fewer people were doing strenuous labor.” 

Schott told Ehrenfruend that he believes that free trade offers U.S. workers, as a whole, important economic benefits, but that government officials need to do more to help individuals and communities displaced by such policies. 

“I’m in favor of free trade, but I’m also someone who believes that we should be honest about the consequences,” he said. “It doesn’t benefit everyone equally.”

Illness as indicator

The Economist editors also looked at county-level health data, but with a different purpose in mind. They wanted to see if the data could help explain Donald Trump’s election win.

Their results seem remarkably related to those of the Pierce-Schott study. For they found that health was a strong predictor of how a county voted in the presidential election. The sicker a county’s residents (overall), the better Trump did.

Article continues after advertisement

Here is the editors’ explanation of what they looked for and what they found:

Since Donald Trump’s surprising victory in America’s presidential election on November 8th, polling enthusiasts have been poring over the data to try to understand precisely how he won. The single factor that best predicted the amount of votes that Republican’s gained compared to 2012 is the share of voting-age citizens who are both white and do not have a college degree. This variable alone can explain 41% of the county-level swing to Mr. Trump. For many that statistic might provide closure on what was a bitter and forgettable election. But Patrick Ruffini, a pollster, called on “data nerds” to find another variable that matches the explanatory power of this pale pedagogic predictor.

Fighting fit for such a challenge, The Economist has crunched the numbers and discovered a coherent set of variables that beat it: an index of health metrics. Together these variables can explain 43% of the Republican party’s gains over the Democrats. Even when controlling for a battery of other indicators — race, education, age, gender, income, marital status, immigration and employment — these health metrics remain significant and predictive.

The data suggest that the ill may have been particularly susceptible to Mr. Trump’s message. According to our model, if diabetes were just 7% less prevalent in Michigan, Mr. Trump would have gained 0.3 fewer percentage points there, enough to swing the state back to the Democrats. Similarly, if an additional 8% of people in Pennsylvania had engaged in regular physical activity, and the rate of heavy drinking in Wisconsin were 5% lower, Hillary Clinton would have won the electoral college vote and be set to enter the White House.

“Polling data suggests that on the whole, Mr. Trump’s supporters are not particularly down on their luck: within any given level of educational attainment, higher-income respondents are more likely to vote Republican,” the editors add. “But what the geographic numbers do show is that the specific subset of Mr. Trump’s voters that won him the election — those in counties where he outperformed Mr. Romney by large margins — live in communities that are literally dying. Even if Mr. Trump’s policies are unlikely to alleviate their plight, it is not hard to understand why they voted for change.”

FMI: You can read The Economist’s article on the journal’s website. Pierce and Schott’s study was published as a working paper by the nonpartisan National Bureau of Economic Research, where it is, unfortunately, behind a paywall.