The most common restaurant cuisine in every state, and a chain-restaurant mystery

Signage outside a McDonald's in Shepherdsville, Ky. (Luke Sharrett/Bloomberg News)
7 min

When Clio Andris and Xiaofan Liang gave us early access to the latest update of their delightful data on chain restaurants, they already had identified its most compelling mystery: Places that support Donald Trump also tend to have the most franchise foods. But why?

It turns out “the foodscape is very political,” said Liang, a PhD candidate at Georgia Tech’s School of City & Regional Planning. “Places with a high percentage of Trump voters have a higher percentage of chains. We didn’t expect it.”

Chain restaurants — those ubiquitous monuments to corporate consistency, from Applebee’s to Arby’s, Olive Garden to Pizza Hut — are most common in Kentucky, West Virginia and Alabama. They’re rarest in Vermont, Alaska and Hawaii. Maine, New York and D.C. also tend to have fewer chains.

The chain restaurant capital of the country is the metro area around Anniston, Ala., home to the Talladega Superspeedway. Nearly 3 in 5 restaurants there are chains.

Department of Data
We here at the Department of Data are dedicated to exploring the weird and wondrous power of the data that defines our world. Read more.

Nestled in the southern reaches of Appalachia, off the interstate between Birmingham, Ala., and Atlanta, Anniston is accustomed to life as a national punching bag. It has been named among the “most dangerous” and “fastest shrinking” cities and appears on lists of the worst places to live and the places where workers are most likely to be replaced by robots. In 2019, local reporter and author Tim Lockette wrote a helpful guide for residents titled, “FIVE THINGS to know when Anniston lands on a ‘10 worst’ list again.

Anniston lies in Calhoun County, which Trump won in 2020 with 73 percent of the two-party vote, which excludes votes cast for third-party candidates. That makes it an exemplar of the Trump-chain restaurant nexus. In the Trumpiest fifth of the United States, counties where Trump received at least 63.3 percent of the two-party vote in the past presidential election, 37 percent of the restaurants are chains. In the least Trumpy fifth, where Trump received less than 32.1 percent of the vote, it’s 23 percent.

There are exceptions. One of the chainiest cities in the country is Waldorf, Md., a fast-growing, majority Black D.C. suburb in Charles County, which President Biden won with 71 percent of the two-party vote. But it’s often safe to assume that the higher the Trump vote, the more chain restaurants you’ll find.

Ocean City, N.J., tops the list of least-chainy metros in America, which is dominated by tourist hubs in places like the Hawaiian island of Maui and Cape Cod in Massachusetts. Their ultra-chainy counterparts at both the city and metro levels typically look more like Anniston: working-class cities, often in Appalachia.

We defined chains as restaurants with at least 50 outlets with the same name nationwide, based on Andris and Liang’s analysis of a database of more than 700,000 restaurants from marketing-data firm Leads Deposit. That includes almost 400 businesses, from heavyweights such as Subway and Dunkin’ to specialists such as HuHot Mongolian Grill and Morton’s steakhouse.

From there, we could weed out names such as China Garden or Joe’s Pizza that were common but not all owned by the same folks.

Of the major cuisines, fast food has the highest chain concentration, while ethnic foods (apart from pizza) tend to have the lowest. This may help explain why we often see fewer chain restaurants in areas with higher immigrant populations.

While the Trump vote correlates with the presence of chain restaurants, it clearly doesn’t explain it. Our gut told us to look at population density. The density divide is the Ur-cleavage from which so many other modern American divisions flow. As places get more rural, education and income levels fall and Trump support rises.

But chains don’t fit perfectly into this worldview. Chain restaurant concentration peaks in midsize cities and suburbs and tends to be lower in both the most urban and most rural areas. And at every density level, the political divide remains: Rural areas won by Biden have fewer chains than rural areas won by Trump. Same goes for suburbs and major cities.

We tested other variables. Education seemed promising. Lower education levels tend to bring both more chain restaurants and more Republican votes. But even within areas with similar levels of education, differences persisted. In the most educated Biden counties, about 26 percent of restaurants were chains. In the most educated Trump counties, 37 percent were — the same as in the least educated Trump counties.

Something else was at work. One by one, we ruled out the possibilities: It wasn’t age, either of people or of the community (as measured by the year a typical dwelling was erected). It wasn’t concentration of White population. It wasn’t income.

In the end, we identified one factor that transcended politics and explained the presence of chain restaurants throughout the nation: driving. Specifically, the share of the workforce that drives to work each day.

The places that drive the most tend to have the same high share of chain restaurants regardless of whether they voted for Trump or Biden. As car commuting decreases, chain restaurants decrease at roughly the same rate, no matter which candidate most residents supported.

If the link between cars and chains transcends partisanship, why does it look like Trump counties have more chain restaurants? It’s at least in part because he won more of the places with the most car commuters!

About 83 percent of workers commute by car nationally, but only 80 percent of folks in Biden counties do so, compared with 90 percent of workers in Trump counties. The share of car commuters ranges from 55 percent in the deep-blue New York City metro area to 96 percent around bright red Decatur, Ala.

Alabama as a whole ranks second in car commuting (behind Mississippi) and third in chain restaurants (behind Kentucky and West Virginia) as people in the South tend to be more likely to drive to work than folks in other regions.

We still aren’t sure why Trump won areas with more car commuters — it could be linked to deeper cultural divides and hidden factors that influence both driving and voting. The link between car commuting and chain restaurants seems clearer.

Andris, a Georgia Tech professor and director of its Friendly Cities Lab, says it’s all about highways. Highways and chains have been linked from the beginning, when a burgeoning car culture and a fast-metastasizing web of interstates gave rise to the roadside Howard Johnsons, one of America’s first great chains.

Ever since, the chains have followed the interstate lanes, creating a quick, transactional filling-station-for-the-human-body experience that’s essentially the opposite of the unique atmosphere and gastronomy available in the independent restaurants that dot the nation’s waterfronts, ski towns and walkable, historic Main Streets.

Whether they’re commuting, running errands or road-tripping, people in car-dominated spaces tend to crave the speed and predictability of chains.

“The legacy of the highway seems to perpetuate chain restaurants,” Andris said.

How do you do, fellow nerds? The Department of Data needs your data-driven questions. Let us know what you’re curious about: the relationship between driving and voting for Trump? The least traveled highways? The cities with the most parking lots? The math behind Zillow and Redfin’s home-value estimates? Just ask!

To get every question, answer and factoid in your inbox as soon as we publish, sign up here. If your question inspires a column, we’ll send a button and an official Department of Data ID card. This week, the button goes to Washington’s JR Pelkola, a childhood friend who flagged the Georgia Tech analysis long before the Department blinked into existence.