Residents walk near a replacement metal wall installed by U.S. workers at the border between Ciudad Juárez, Mexico, and Sunland Park, N.M., on Sept. 12. (Herika Martinez/AFP/Getty Images)

When Donald Trump launched his presidential campaign, he promised to build “a great, great wall on our Southern border.” Since then, pundits and scholars have pointed out the wall would be extremely expensive, ineffective, and harmful to the local ecosystem and economy. Even so, Trump has doubled down on his promise to build an “impenetrable” border wall.

But if Trump is elected and tries to build a wall along the Southern border, support is likely to plummet once people see that to do so would require taking hundreds, if not thousands, of properties from ordinary homeowners and small businesses.

How much private land would the federal government need to take to build a wall with Mexico?

The border between the U.S. and Mexico is 1,954 miles long. Large chunks of land along that border — especially in Texas — are privately owned. Building the wall would require the federal government to take property from hundreds of American citizens living along the border.

Consider what happened when the U.S. built the existing segments of the national border fence. In 2006, Congress passed the Secure Fence Act. The federal government then built roughly 650 miles of fencing along the Southern border. Roughly 100 miles of this were in Texas, where the government had to take property from more than 400 border residents.

But government “takings” are quite unpopular. Here’s how I measured this:

In a recent article, I find that government use of eminent domain — taking property for “public use” — is unpopular in the best of circumstances. It’s extremely unpopular when the public benefit isn’t clear. People are more likely to support takings for projects that will be used by — or at least will be open to — the public.

That includes the traditional “narrow” uses of eminent domain for such purposes as building roads, schools, and hospitals. There’s less support for “broad” use takings, when the public won’t actually use the land, even if there’s supposed to be some indirect benefit like economic growth or more jobs.

So how unpopular is it for government to take private property?

Having government take private property was unpopular no matter what — but respondents were much more opposed to, and angrier at, property being taken for a “broad” use.

Graph by Logan Strother. Source: Original survey experiment conducted at Amazon’s Mechanical Turk workplace.
Public opposition to government takings by eminent domain Graph by Logan Strother. Source: Original survey experiment conducted at Amazon’s Mechanical Turk workplace.

Surprisingly, here’s what didn’t matter: whether government was taking a home or a vacant lot. What people cared about was how the government would use the property. Opposition didn’t grow from sympathy for individuals forced from their homes. People care about why property is taken, not what kind of property is being taken.

 What does that mean for Trump’s border wall?

Taking private property to build a wall is a clear example of a “broad” use taking. The public won’t use it, and any benefits would be indirect.

Those who distrust the government are especially opposed to takings of all kinds, but especially to broad-use takings. And so some of those who support Trump because of his “outsider” status may be particularly angry about takings for a border wall.

Trump also has a history of using eminent domain for private gain, so there seems to be little reason to suspect he wouldn’t use it to build a border wall. In the early 1990s Trump asked Atlantic City’s Casino Reinvestment Development Authority (CRDA) to take properties from several homeowners so he could build a limousine parking lot for his Trump Taj Mahal; when that was challenged, Trump and the CRDA lost in state court. In 1994, Trump lobbied the city of Bridgeport, Conn., to take waterfront properties from five small business to sell it to Trump Enterprises Connecticut so he could develop the land into offices and an entertainment destination. That plan also failed.

Trump has even tried to use this power abroad, attempting (and failing) to force individuals living near his golf course in Scotland to give up their homes so he could expand his development. And in 2005, Trump was asked about the Supreme Court’s widely reviled decision in Kelo v. New London, which declared it constitutional for governments to take property for economic development. Trump answered:

I happen to agree with it 100 percent. If you have a person living in an area that’s not even necessarily a good area, and … government wants to build a tremendous economic development, where a lot of people are going to be put to work and … create thousands upon thousands of jobs and beautification and lots of other things, I think it happens to be good.

But if he tried to build a border wall, many Americans would likely disagree.

Logan Strother is a PhD candidate in the department of political science at the Maxwell School of Syracuse University, and a visiting scholar at the Truman School of Public Affairs at the University of Missouri.