The geography of voting — and not voting

People around the country can pass judgment on their government Nov. 6 in the first national election in two years — if they’re registered to vote, and cast a ballot.

Many will not, if recent history holds true. Only about 60 percent of U.S. adult citizens voted in the last national election, a turnout that ranked in the bottom third of the world’s developed countries.

Turnout in 2016 was especially low in almost a thousand U.S. counties. Most of them form a geographic belt that curves from upstate New York, down along the Appalachians, across Tennessee and into the Southwest. But 2018 has brought statewide races predicted to be competitive to the heart of the low-voting belt, in West Virginia, Tennessee, Oklahoma, Texas and New Mexico. The November results will be studied for signs of closer races and greater voter involvement to come in 2020.

These maps show a turnout percentage based on two numbers for each county: ballots cast in the 2016 presidential race and the estimated population of voting-age citizens. Among the low-voting counties, those with turnout of 55 percent or less, there are some striking differences.

For example, low-voting areas span the entire urban-rural spectrum. On the map, rural areas and small towns may dominate visually. But there are far more potential votes in low-turnout metropolitan counties.

Fifteen large central metro counties with low 2016 turnout

There are low-voting counties even in the cores of some of the nation’s largest urban areas, including four of New York City’s five boroughs, as well as part or all of Houston, Dallas, Memphis, St. Louis, Baltimore and Oklahoma City.

They also vary in their racial makeup. Low-turnout counties include more than 400 that are predominantly (more than 80 percent) white. But in another 200, nonwhites are a majority, often including a significant Hispanic population.

There are demographic traits that consistently set apart places where people don't vote. The most well-known is education. In more than 700 low-turnout counties, a majority of people had attained only a high school diploma or less schooling. In comparison, education beyond high school is far more common among high-voting counties.

The map also highlights stark differences between states, with high-voting areas just across the border of low-voting states such as West Virginia and Tennessee. In some cases, states differ in how convenient it is to register and vote. Six of the top seven highest-voting states allow voters to register on Election Day. In Colorado, Oregon and Washington, residents get a ballot by mail, and can vote by mailing it back.

Stark differences in neighboring states also reflect in part the differing culture and traditions in areas such as civic involvement and in minority relations, said Jay Barth, a politics professor at Hendrix College in Arkansas.

For midterm elections like 2018, turnout is even worse than in presidential years, lower by about 20 percentage points, and trending down. Since 1800, there have been only three midterms with turnout lower than in the last midterm in 2014, according to figures compiled by Michael McDonald, a political science professor at the University of Florida.

But McDonald and others are expecting higher turnout in November. And some are looking for signs that 2018 marks a shift in voter participation that changes the nation’s political map. There are more competitive races, and voters are paying more attention.

A recent Washington Post-ABC News poll found that voters are significantly more interested in voting than they were four years ago. It was especially true for young people under 25, whose turnout in 2016 lagged the national average by 18 percentage points, and for nonwhites.

As of Tuesday, Cook Political Report rated 72 House races as competitive, dozens more than in the same week of 2014. Those competitive districts include part or all of 121 counties that were low-voting.

More competitive races could fuel greater voter participation because competition drives voter interest. More close races could even push up turnout in swing states where tiny winning margins of a fraction of one percentage point showed in 2016 how relatively few votes could be decisive. President Trump won Wisconsin by 0.8 of a percentage point, Pennsylvania by 0.7 and in Michigan by only 0.2.

“There is something that happens from living in a state where there is actually campaign activity, and voters sense that their vote matters, because the parties have invested time in running advertisements and reaching out to them,” said Barry Burden, director of the Elections Research Center at the University of Wisconsin-Madison.

Those sharp state lines that appear on the voting map, often following the boundaries between states with competitive races and those without, could begin to blur.

“It will be interesting to see if turnout is up in some of those places this year because they have become competitive after long periods of not being competitive,” Burden said. “That may be kind of a signal as to what we might see in the next electoral college vote where some of these states come into play.”