Sign up for The Decision, a newsletter featuring our 2024 election coverage.

Charleroi is a small mill town south of Pittsburgh whose dozen blocks, running along the tracks of the Norfolk Southern Railway, are nestled in a valley between the Monongahela River and the worn-down foothills of western Appalachia. Going back more than a century, Charleroi (nicknamed “Magic City”) has made glassware, with a peak population of more than 11,000, a unionized workforce, and a dominant Democratic Party. By the 1970s the factories had begun to disappear, and with them many of the people. By 2020, after half a century of deindustrialization, Charleroi was a town of vacant stores and about 4,200 souls, most of them Republicans. It’s the saga of the Rust Belt, writ small and ongoing.

When I asked Joe Manning, the borough manager, what moved Charleroi from blue to red, he replied: “2016. I know people who were lifelong, dyed-in-the-wool, staunch Democrats who, during that period, went out and changed their registration so that they could vote for Trump.”

Following Donald Trump’s victory that year, academics and journalists embarked on a search for an explanation. Progressives quickly lighted on racism as the sole answer. This conclusion was a costly mistake. Analytically, it ignored important causes that anticipated coming trends; politically, it alienated the unconverted and made discussion more difficult. Kamala Harris appears determined not to repeat the mistake as she downplays identity as a theme in her campaign. Race is only part of the reason for Trump’s persistent base of support, and one that’s grown less significant. The starkest division in American politics is class, as defined by education—the wide gap between voters with and without a college degree—which explains why more working-class Latino and Black citizens have begun to vote Republican. But in a more complex way, political behavior in the Trump era is determined by how class and race interact. The most convincing accounts of the 2016 presidential election found that the leading determinant of support for Trump was residence in a declining white community that had recently seen the arrival of nonwhite immigrants, which brought rapid cultural change and created a sense that the country was becoming unrecognizable.

[Watch: Shifting campaign strategies]

In 2020, Getro Bernabe, an American-trained officer with the Haitian Coast Guard, fled Haiti’s gang violence and arrived in Charleroi looking for work. “It was like a ghost town,” he told me. “It looked like a beautiful place, but now abandoned.” In the past few years Charleroi has gained 2,000 immigrants, mostly Haitians drawn by empty houses and low-wage jobs, raising the town’s population close to its 1970 number. “The newcomers, the new residents in Charleroi, are like a glimmer of light to the economy of this town,” Bernabe said. “I like one of the core values of America—it is on the American coin.” He meant E pluribus unum, which he interpreted as referring to a unified nation of people from different backgrounds and beliefs. “That’s the beauty of America to me.”

Kristin Hopkins-Calcek, the borough-council president, has lived her whole life in Charleroi. “I watched the town deteriorate over time, and it was very hurtful for us that stayed,” she told me when we met in the council chamber. “Coming from owning a house here, watching my son fall into addiction, and seeing the fentanyl and Oxy problem that we had here, and the overdoses, the crime, and even to some extent the prostitution in town, and the ruination and the blight of our property, and the absentee landlords, and, it seems when you’re older, like the instant decline of our town—when the immigrants came in, it was a breath of fresh air. There were people on the streets; there were businesses opening.”

Charleroi is a fragile place: buoyed by the new grocery stores and bakeries of immigrant entrepreneurs, and new renters and taxpayers; strained by insufficient resources, traffic mishaps, and resentment. There’s no prosperous professional class in Charleroi. Its half-deserted streets and sidewalks are shared by two working-class populations: aging white residents whose families have lived here for generations, and younger Black immigrants who arrived in the past few years. This is Trump country—festooned with Trump flags, Trump yard signs, and, on the deck of a trailer in the woods outside town, a Trump banner boasting: IMPEACHED. ARRESTED. CONVICTED. SHOT. STILL STANDING. In a variety shop on Fallowfield Avenue, half the items for sale are Trump paraphernalia.

Last month, two disasters befell Charleroi almost simultaneously. On September 4, the Pyrex factory on the river, which has produced glassware since the 1890s, told its more than 300 union workers that the owners will close the plant by the end of the year and move operations to Ohio. Then Trump heard about Charleroi.

A campaign sign for Republican presidential nominee and former President Donald Trump is seen as an immigrant walks along a street in downtown Charleroi on September 24, 2024. (Carlos Barria / Reuters)

Joe Manning was watching the presidential debate on September 10 when Trump repeated a false story about Haitians eating the cats and dogs of Springfield, Ohio. “Oh my goodness,” Manning thought, “let it just be Springfield.” His wish went unanswered. On September 12, at a rally in Arizona, Trump locked onto Charleroi. “What a beautiful name, but it’s not so beautiful now,” he said. “It has experienced a 2,000 percent increase in the population of Haitian migrants under Kamala Harris. So, Pennsylvania, remember this when you go to vote. This is a small town, and all of a sudden they got thousands of people … The town is virtually bankrupt. This flood of illegal aliens is bringing massive crime to the town and every place near it.” At a rally in Pennsylvania on September 24, he repeated the attack on Charleroi: “Has your beautiful town changed? It’s composed of lawless gangs.”

The “2,000 percent” figure was nonsensical. The Haitians in Charleroi came legally, in search of jobs, and found ones that Americans wouldn’t take, such as food preparation on assembly lines in 40-degree temperatures. The town isn’t bankrupt, there are no gangs, and crime has not gone up, according to Hopkins-Calcek, who sits on the regional police board. “The most heinous crime recently was an infanticide,” Manning told me, “and the parents were both arrested, and they’re both as white as us.”

None of this mattered to Trump. He had found a small, tender wound in a crucial swing state and stuck a finger inside. Then he moved on to other targets, but the effect in Charleroi was overwhelming. Manning and Hopkins-Calcek received threats. A flyer addressed to “White Citizens of Charleroi” and signed by “Trinity White Knights of the Ku Klux Klan” circulated, warning: “Arm yourselves white America, protect your families. White people are the only victims to immigrant brutality.” Passing drivers were emboldened to shout at Haitians, “Trump is coming!” Bernabe, who is the borough’s immigrant-community liaison, heard from people who were afraid to send their children to school and thinking of leaving the state. “All of a sudden, we’ve been seeing a certain fear among the immigrant people, like they feel like they are not welcome, comfortable,” he told me earlier this month. “You see them less and less outside.” Charleroi began to look like the ghost town it had recently been.

For Hopkins-Calcek, Trump’s damage brought back the nightmare of her town’s descent. “It got really quiet, and it got scary again,” she said, beginning to cry. “When they went back in the houses, it felt like it was bad again.” With the imminent departure of Charleroi’s legacy industry, along with its tax revenue, “I feel as if we’re being kicked when we’re down,” she said.

Trump never mentioned the Pyrex factory.

One afternoon earlier this month, I sat with five members of the United Steel Workers Local 53G in a McDonald’s near the Charleroi railroad tracks. They had spent most of the day negotiating the end of their livelihood with lawyers from Anchor Hocking—the glassware company, owned by a New York investment firm called Centre Lane Partners, that plans to close the Pyrex factory. Daniele Byrne, the local’s vice president, and her husband, Rob, an electrician, have worked at the Charleroi plant for a total of 71 years. Before Daniele, her grandfather put in 50 years and set his wall clock by the noon whistle. As severance, the company was offering two months’ health insurance, plus a day’s pay for every year of employment—about $8,000 for two-thirds of Daniele’s life.

She didn’t hide her disgust. “Here you go, be on your way, merry Christmas, happy Kwanzaa,” she said. “What’s the Jewish one?”

Rob asked if I had read Glass House, a book about Lancaster, Ohio, a fading industrial town three hours west, where Anchor Hocking has a glass plant and plans to move the Charleroi factory, along with up to half its workforce. “It’s about the 1 percent economy that started Trumpism,” Rob said. “How they control everything, buying and selling and making all these maneuvers. The billionaires keep getting more and more while everybody else suffers.”

The workers’ hostility toward corporations and billionaires didn’t translate automatically into support for a candidate or party. Their alienation from politics and distrust of elites was too great. The word I kept hearing, in Charleroi and around western Pennsylvania, was care—as in, “They don’t care about us.” It conveyed a deep sense of abandonment.

Half a dozen Haitians work at the Pyrex factory. Daniele, who’s in charge of scheduling, told me they were better workers than the American ones. “I don’t think the problem is the immigrants,” Rob said. But he and the others had complaints about the sudden arrival of so many foreigners in their small town: overcrowded school buses and classrooms, overextended teachers, government benefits the locals didn’t get, and—despite what I’d heard from town officials—higher crime. They claimed that a new immigrant-owned grocery store had put up a sign barring white shoppers. Finding this implausible, I asked Getro Bernabe about it later. He explained that the sign had advertised food from Africa, Asia, and the Caribbean, while omitting American food. When he rushed to the store and told the owner that local people were complaining, she was aghast: “My God, I didn’t think of that.”

[Read: Harris’s best answer to Trump’s resilient appeal]

“Please, put American,” Bernabe urged, but to avoid problems she replaced the sign with one that said simply Queen’s Market. When I visited the store, it was selling live crabs, dried fish, and other products that seemed a little unusual for western Pennsylvania. The owner, an American citizen of Sierra Leonean origin, had put a sign behind the counter that said Trump 2024. This detail, which went against the grounds for local displeasure, hadn’t become a story.

False rumors can be more revealing than true ones, and there are tensions in Charleroi that shouldn’t be either wished away or inflamed. “It’s not hatred so much as—” Daniele began.

“Envy,” Rob said. “Jealousy.”

Longtime residents felt as if they didn’t matter. The Pyrex closing got far less attention than Trump’s commentary on Haitians. Every four years, the political and media class takes an interest in towns like Charleroi for a few autumn weeks. “If Kamala comes here, she’s right now in the battle of the Haitians because she wants the immigrants here and he wants them gone,” Daniele said. “They forget about us and go straight to the immigrants again.” She added, “I don’t pay attention to politics; I’ll be honest. I think they’re all crooks. I’d sooner watch Barney Miller. I can’t wait ’til November’s over so I can watch regular commercials about what razors to buy.” The workers didn’t hate all politicians—just the ones who made promises they didn’t keep and exploited the problems of people like them. Pennsylvania’s Senator Bob Casey is pushing the federal government to examine Anchor Hocking’s acquisition of the factory in a bankruptcy sale earlier this year for a possible violation of antitrust law. This effort won credit even from the scathing Daniele Byrne.

Two nights after we met, Rob and Daniele went to see the Steelers play the Cowboys in Pittsburgh. A friend had gotten me a ticket, and early in the first quarter, people near me suddenly began turning to look behind us and cheer. Thirty feet above, a man in a black blazer and black cap was standing in a luxury box, waving a yellow Steelers towel and grinning. It was Elon Musk—fresh from hopping around onstage at Trump’s return to the scene of his shooting in nearby Butler, now basking in a football crowd’s adoration of wealth and celebrity.

When I told Daniele, she said: “Ah, the fucker.”

A resident chats with an immigrant in downtown Charleroi on September 24, 2024. (Carlos Barria / Reuters)

The convergence of working-class decline, corporate greed, and nativist anger will shape next month’s election in places like Charleroi and throughout the Rust Belt. Northwest of town, Pennsylvania’s Seventeenth Congressional District is represented by Congressman Chris Deluzio. He’s a first-term Democrat, having narrowly won in 2022 in a competitive district of farmland, Pittsburgh suburbs, and mill towns along the Ohio River. Deluzio is a 40-year-old Navy veteran and attorney, neatly groomed, polite, and analytical in a way that doesn’t scream “populist.” But he’s running for reelection on the bet that his pro-labor, anti-corporate positions will prevail over the hostility toward immigrants that Trump and other Republicans are stirring up. (The campaign of Deluzio’s opponent, State Representative Rob Mercuri, didn’t respond to my request for an interview.)

“The Wall Street guys bankrolling Trump and my opponent are the guys who devastated these communities,” Deluzio told me as we drove between campaign events. “They tried to strip us for parts for decades. The mills didn’t just leave; they were taken away by an ideology and a set of policies that said cheaper and weaker labor rules and cheaper and weaker environmental rules is what they’re after. Your family’s hard work and sacrifice didn’t matter to these guys.” After a Norfolk Southern freight train carrying toxic chemicals derailed last year in East Palestine, Ohio, just across the state line from Deluzio’s district, he drafted legislation to tighten the regulation of rail freight, which Ohio’s Senator J. D. Vance co-sponsored. The Railway Safety Act, opposed by the Koch political network, is currently stalled by Republicans in both houses of Congress. Even though few of Deluzio’s constituents were directly affected by the spill, it’s the kind of issue that he hopes will distinguish Democrats like him from pro-corporate, anti-regulation Republicans.

Deluzio argued that Trump villainizes new immigrants to distract local people—themselves the descendants of immigrants and legitimately anxious about rapid change in their towns—from the true causes of their pain: monopolistic corporations and the politicians they fund. He acknowledged that the national Democratic Party failed for years to make this case and pursued trade policies that undermined it. An idea took hold that college-educated voters would soon outnumber the party’s old base of a moribund working class. “For every blue-collar Democrat we lose in western Pennsylvania, we will pick up two moderate Republicans in the suburbs in Philadelphia,” Senator Chuck Schumer predicted in 2016, shortly before Trump won Pennsylvania, and with it the presidency.

The Biden administration has tried to earn the loyalty of working-class voters with pro-union policies and legislation to create jobs in depressed regions. But people I spoke with in western Pennsylvania seemed to have only a vague idea how the Democratic Party is trying to woo them back. The rising cost of living mattered more to them than low unemployment and new manufacturing and Harris’s tax plans. When underinformed and undecided voters say that they want to hear more details about a candidate’s policies, it usually means they don’t believe that policies will make any difference in their lives. To overcome ingrained skepticism after decades of disinvestment, a politician has to show up, look voters in the eye, shake their hand, and then deliver help—or at least be seen to care enough to try.

Curtis and Annie Lloyd live in Darlington, a rural borough on the Ohio border a few miles from the site of last year’s chemical spill. When the Lloyds saw a gray cloud rise into the sky near their house, they found it almost impossible to get solid information about the freight disaster: The county paper is a ghost of its former self, and social media predictably swarmed with conflicting and false stories. But Trump paid a visit to the area, Annie told me, while President Biden didn’t for more than a year—and that made a stronger impression than Deluzio’s effort, thwarted by Republicans, to pass regulatory reform. “People are living their lives, and they don’t delve that easily into policy,” she said. “All they know is Trump was here buying everyone McDonald’s.”

Fifteen miles away, in the town of Rochester, I met a woman named Erin Gabriel at the headquarters of the Beaver County Democratic Party. The office was a hive of activity, with canvassers on their way in or out and Harris/Walz signs stacked against the walls. Gabriel told me that politics was personal to her. While working full-time and chairing the county party, she cares for her three disabled children (her teenage daughter, Abby, who suffers from a devastating neurodegenerative disease, was sitting in the next room with headphones on). “Every single government policy affects my children,” Gabriel said. Without the Affordable Care Act, Abby would have no health insurance for the rest of her life. During Trump’s presidency, Gabriel’s congressman, a Republican, promised her that he would do everything he could to protect Abby’s access to health care. Then he voted for Trump’s bill to overturn Obamacare.

“That’s when I got really active,” Gabriel said. “This is visceral to me.”

For a moment, southwestern Pennsylvania has outsize power and attention. Yard signs appeared everywhere; cashiers in bakeries counted sales of their Trump and Harris cookies. National politics is tribal and hardly open to persuasion. Local politics feels different—less hateful and more flexible, with plenty of ticket splitting. Rico Elmore, a young Republican councilman in Rochester, told me, “We have to find the commonalities and say, ‘We may be different on criminal-justice reform, on taxes, on immigration, but we can come together. My streets need paved; you believe they need paved. Let’s get it done. Let’s find those common goals and work towards that.’”

Elmore, a Black Air Force guardsman, was at the rally in Butler where Trump was shot, and rushed to render first aid to Corey Comperatore, the man who was killed; Comperatore’s family then invited Elmore to speak at Trump’s second Butler rally. He’s a rising star in local Republican politics, and in 2022, in an unsuccessful race for state representative, he knocked on 13,000 doors. He found even Democrats willing to listen, and from both sides he heard something that almost everyone I met, even the strongest partisans, also voiced: an overwhelming desire to move past polarization. Elmore wondered whether America is headed for the fate of the Roman empire. “Are we at that point in history? What are we doing to prevent that from happening? We are becoming a nation that is being divided and will fall. We cannot stand divided.”

On a crystalline October afternoon, Chris Deluzio went door-to-door in a new subdivision of Allegheny County. He was wearing a half-zip pullover that said NAVY—a way, it seemed, to let constituents know that his status as their congressman and a former scholar at the University of Pittsburgh’s Institute for Cyber Law, Policy, and Security didn’t mean he wasn’t one of them. Both Democrats and Republicans lived on the cul-de-sac of single-family homes. At one, a young man in a USC cap named Aaron was working on a truck in his driveway. “You already got my vote,” he told Deluzio. Aaron described himself as a moderate Democrat from California who couldn’t stand what Republicans were doing. “I grew up with Latinos my entire life, I love ’em. I actually miss ’em, being out here, and the way they talk about ’em, it bothers me. If I were on the Republican side, I’d be on the Schwarzenegger middle of the road.”

“Does that exist anymore, those guys?” Deluzio asked.

“From what I see on that side, no. I see it in the blues, but just not on that side. It’s just gone too far.”

[Gilad Edelman: The man who’s sure that Harris will win]

The next house had a Trump yard sign, but Deluzio rang the doorbell anyway. A big-bodied older man with a crew cut answered. He was a police officer in Ambridge, a town on the Ohio River. I had driven through Ambridge, where steel was once fabricated for the Empire State Building: another depressed mill town, with dollar stores, vape shops, and a World War II memorial park with a Four Freedoms monument that belongs to an earlier century.

The policeman, whose name was Mike, said that he had met the congressman in Ambridge. Deluzio reminded him that he had the endorsement of the county’s police union. “I keep an open mind,” Mike said. “I just have a problem with the border and the crime, because I see it down in Ambridge. It’s just a big immigration problem.” Most of the town’s immigrants came from Latin American countries like Venezuela, Mike said, and they brought “DUIs, drunkenness, domestics, a lot of fights.” He would vote on crime and border security.

An elderly woman called out something from the back of the house.

“My mom, she’s on Social Security,” Mike said, “and these people are getting $4,000 a month, and that’s more than she gets. She’s upset they get more—and I’m gonna tell you, my mom voted Democratic her whole life. She switched to Republican.”

I’d heard complaints in Charleroi about government handouts to immigrants. Joe Manning, the borough manager, had explained, “I don’t have a line item in my budget for Haitians. They don’t need my resources. They’re all gainfully employed.”

But Deluzio didn’t question Mike’s story, or argue with him about crime and immigration, or try to persuade him of anything. He had made a connection. Maybe that would be enough.

Seventy Miles in the Darién Gap

For the September 2024 issue, Caitlin Dickerson reported on the impossible path to America.

As a Colombian American, I was deeply moved by “Seventy Miles in the Darién Gap.” Thank you, Caitlin Dickerson, for your courage. I had the deep fortune of migrating to the United States legally with my parents in the 1990s, so I didn’t experience the Darién Gap personally. Recently, I’ve been helping a Colombian refugee who traveled through the Darién Gap. He began to tell me of his experiences there, and I was astounded by his story. He is understandably still processing what he witnessed, and I am letting him go at his own pace. Dickerson’s reporting offered a remarkable window onto a harrowing journey undertaken by the most desperate of people. Thank you for investing in such solid journalistic work. Now I’m going to go hug my dogs and wife.

Carlos Enrique Gomez
Union City, Calif.

As a citizen of the United States and an avid consumer of its news, I’m saddened that most mainstream-media coverage of our immigration woes focuses on controlling our borders and not the underlying reasons people risk, and even lose, their lives in their attempts to immigrate here.

For those who only listen to sound bites, the word immigrant conjures frightening notions—outsiders on a quest to thwart our border security and take some of what we consider to be ours. In Donald Trump’s view, they are murderers, criminals, and rapists.

Caitlin Dickerson’s article reveals that these are mostly just normal people forced to flee their countries due to conditions beyond their control. I can’t imagine how dire circumstances would have to be for me to leave my home! It’s telling and sad to see that U.S. policy to discourage immigration has had the effect of increasing death rates among those who are already so helpless. Not to mention driving new profits for drug cartels.

I hope we can have more coverage centered on the root causes of immigration. After all, U.S. policy created many of the problems that plague countries in Latin America.

Peter Brown
Lyman, Maine

I teach high-school English in Columbus, Ohio. Last year, one of my students wrote an essay about his experience traveling through the Darién Gap. It was the first time I had ever heard of it. This student was hardworking and kind, and I was amazed by his story. When he wrote it, he had been in the United States for just over a year. It’s 288 words, with minimal punctuation and no paragraph breaks.

He left his home country in South America with his mom and sisters. After passing through the Darién Gap, they spent time in Costa Rica, Nicaragua, Honduras, Guatemala, and Mexico, before eventually ending up in Ohio.

He concluded his essay with this: “It affects me in what? I got a lot of depression and stress and I won’t do something like that again.” I feel lucky to have taught this student, and I appreciate that The Atlantic covered this topic.

Chase Montana
Columbus, Ohio

‘Lord, Help Us Make America Great Again’

In the September 2024 issue, McKay Coppins considered the most revealing moment of a Donald Trump rally.

McKay Coppins’s close reading of Trump-rally prayers was unsettling, even frightening. I was concerned less by the apocalyptic fear and strange theology that the prayers mobilized, and more by the unnerving similarity I saw to the rhetoric marshaled against Trump and Republicans by their adversaries.

I confess to seeing in Trump’s opponents—and I count myself among them—the same tendency toward exaggeration (Trump is a modern Hitler, Trump is an existential threat to democracy). Conservatives have been quick to argue that many progressives behave with a quasi-religious zeal: Popular slogans echo liturgy; cancel culture exists as a penalty for heresy.

I’d like to think that there are differences between Trump and his critics that I’m not discerning. Could The Atlantic do a similar sort of analysis of the weirder expressions by Democrats and progressives?

Gary Gaffield
Fort Myers, Fla.

My Mother the Revolutionary

For the September 2024 issue, Xochitl Gonzalez considered what happens when fomenting socialist revolution conflicts with raising a family.

As a mother of young children and a committed socialist organizer, I found that Xochitl Gonzalez’s recent article presented an unrealistic and at times bizarre portrait of the lives of people like me. The bulk of the article is a slippery mix of memory, feeling, and fact—understandable if its purpose was to explore the bitter process of reconciliation between an absent parent and her child, but unsatisfying if it aims to provide an accurate political analysis.

What moved me to comment was the strange choice, 6,000 words into an almost-7,000-word essay, to pivot to a discussion of the presidential campaign of Claudia De la Cruz and Karina Garcia, who are running on the ticket of the Party for Socialism and Liberation. Although the author conducted an interview with the candidates, the only remnant of that interaction was a physical description of them (They are—“not that it matters—beautiful”) and a hasty reduction of their political platform (Burn it all down. Start from scratch). What a shame to silence these women and conflate their candidacy with the aberrant personal experiences of the author.

Polls show that more and more young adults like me have positive attitudes toward socialism. We see the failures of capitalism all around us, and we are eagerly dedicating ourselves to building a socialist future. Although the article depicts socialist activism as a kind of personal obliteration, a subordination of our individual selves to the menacing whims of “the party,” the reality, in my experience, could not be further from the truth.

I proudly support Claudia and Karina, not just because their politics offer the only viable path out of poverty, imperialist wars, and ecological crisis, but also because they are working mothers like me. When they speak about inflation at the grocery store, it is from experience. When they speak about the astronomical cost of child care, it is from experience. When they speak about fighting for a world that truly nurtures our children, it is proof that our identities as mothers are an asset, not a liability, in this struggle.

Moira Casados Cassidy
Denver, Colo.

Behind the Cover

In “Washington’s Nightmare,” Tom Nichols revisits the life of George Washington, whose bravery and self-command established an ideal that all future presidents would, with varying degrees of success, attempt to emulate. All, that is, save Donald Trump, a man who shares none of Washington’s qualities and exhibits the kind of base motives that the first president saw as a threat to the republic. For the cover, we turned to Gilbert Stuart’s The Athenaeum Portrait. The unfinished nature of the work suggests the ongoing American experiment, but also the existential danger that a second Trump term poses.

— Elizabeth Hart, Art Director

Correction

“You Think You’re So Heterodox” (October) misstated where Joe Rogan’s home is located. It is west of Austin, not east of Austin.

This article appears in the November 2024 print edition with the headline “The Commons.”

Illustrations by Anuj Shrestha

It’s called the “longest-swim problem”: If you had to drop someone at the place in the ocean farthest from any speck of land—the remotest spot on Earth—where would that place be? The answer, proposed only a few decades ago, is a location in the South Pacific with the coordinates 4852.5291ʹS 12323.5116ʹW: the “oceanic point of inaccessibility,” to use the formal name. It doesn’t get many visitors. But one morning last year, I met several people who had just come from there.

They had been sailing a 60-foot foiling boat, the Mālama, in the Ocean Race, a round-the-world yachting competition, and had passed near that very spot, halfway between New Zealand and South America. Now, two months later, they had paused briefly in Newport, Rhode Island, before tackling the final stretch across the Atlantic. (And the Mālama would win the race.) I spoke with some members of the five-person crew before going out with them for a sail on Narragansett Bay. When I asked about their experience at the oceanic pole of inaccessibility, they all brought up the weather.

With a test pilot’s understatement, the crew described the conditions as “significant” or “strong” or “noteworthy” (or, once, “incredibly noteworthy”). The Southern Ocean, which girds the planet in the latitudes above Antarctica and below the other continents, has the worst weather in the world because its waters circulate without any landmass to slow them down. The Antarctic Circumpolar Current is the most powerful on Earth, a conveyor belt that never stops and that in recent years has been moving faster. These are the waters that tossed Roald Amundsen and Ernest Shackleton. The winds are cold and brutal. Waves reach 60 or 70 feet. In a second, a racing boat’s speed can drop from 30 knots to five, then jump back to 30. You may have to ride out these conditions, slammed and jammed, for five days, 10 days, trimming sails from inside a tiny sealed cockpit, unable to stand up fully all that time. To sleep, you strap yourself into a harness. You may wake up bruised.

[Read: The last place on Earth any tourist should go]

This is not a forgiving environment for a sailboat. But it’s a natural habitat for the albatross you find yourself watching through a foggy pane as it floats on air blowing across the water’s surface—gliding tightly down one enormous wave and then tightly up the next. The bird has a 10-foot wingspan, but the wings do not pump; locked and motionless, they achieve aerodynamic perfection. The albatross gives no thought to the longest swim. It may not have touched land in years.

The oceanic pole of inaccessibility goes by a more colloquial name: Point Nemo. The reference is not to the Disney fish, but to the captain in Jules Verne’s novel Twenty Thousand Leagues Under the Sea. In Latin, nemo means “no one,” which is appropriate because there is nothing and no one here. Point Nemo lies beyond any national jurisdiction. According to Flightradar24, a tracking site, the occasional commercial flight from Sydney or Auckland to Santiago flies overhead, when the wind is right. But no shipping lanes pass through Point Nemo. No country maintains a naval presence. Owing to eccentricities of the South Pacific Gyre, the sea here lacks nutrients to sustain much in the way of life—it is a marine desert. Because biological activity is minimal, the water is the clearest of any ocean.

What you do find in the broad swath of ocean around Point Nemo—at the bottom of the sea, two and a half miles below the surface—are the remains of spacecraft. They were brought down deliberately by means of a controlled deorbit, the idea being that the oceanic point of inaccessibility makes a better landing zone than someone’s rooftop in Florida or North Carolina. Parts of the old Soviet Mir space station are here somewhere, as are bits and pieces of more than 250 other spacecraft and their payloads. They had been sent beyond the planet’s atmosphere by half a dozen space agencies and a few private companies, and then their lives came to an end. There is a symmetry in the outer-space connection: If you are on a boat at Point Nemo, the closest human beings will likely be the astronauts aboard the International Space Station; it periodically passes directly above, at an altitude of about 250 miles. When their paths crossed at Point Nemo, the ISS astronauts and the sailors aboard the Mālama exchanged messages.

Illustration by Anuj Shrestha

The Mālama’s crew spoke with me about the experience of remoteness. At Point Nemo, they noted, there is no place to escape to. If a mast breaks, the closest help, by ship, from Chile or New Zealand, could be a week or two away. You need to be able to fix anything—sails, engines, electronics, the hull itself. The crew described sensations of rare clarity and acuity brought on by the sheer scale of risk. The austral environment adds a stark visual dimension. At this far-southern latitude, the interplay of light and cloud can be intense: the darks so very dark, the brights so very bright.

Simon Fisher, the Mālama’s navigator, described feeling like a trespasser as the boat approached Point Nemo—intruding where human beings do not belong. Crew members also described feelings of privilege and power. “There’s something very special,” Fisher said, “about knowing you’re someplace where everybody else isn’t.”

We all know the feeling. Rain-swept moors, trackless deserts, unpeopled islands. For me, such places are hard to resist. Metaphorically, of course, remoteness can be found anywhere—cities, books, relationships. But physical remoteness is a category of its own. It is an enhancer: It can make the glorious better and the terrible worse. The oceanic pole of inaccessibility distills physical remoteness on our planet into a pure and absolute form. There are continental poles of inaccessibility too—the place on each landmass that is farthest from the sea. But these locations are not always so remote. You can drive to some of them. People may live nearby. (The North American pole of inaccessibility is on the Pine Ridge Reservation, in South Dakota.) But Point Nemo is nearly impossible to get to and offers nothing when you arrive, not even a place to stand. It is the anti-Everest: It beckons because nothing is there.

I first heard the name Point Nemo in 1997, when hydrophones on the floor of the South Pacific, thousands of miles apart, picked up the loudest underwater sound ever recorded. This got headlines, and the sound was quickly named the “Bloop.” What could be its source? Some speculated about an undiscovered form of marine life lurking in the abyssal depths. There was dark talk about Russian or American military activity. Readers of H. P. Lovecraft remembered that his undersea zombie city of R’lyeh was supposedly not far away. Scientists at the National Oceanic and Atmospheric Administration eventually concluded that the sound had come from the fracturing or calving of ice in Antarctica. In this instance, freakish conditions had directed the sound of an Antarctic event northward, toward a lonely expanse of ocean. Faraway hydrophones then picked up the sound and mistook its place of origin. News reports noted the proximity to Point Nemo.

[Video: The loudest underwater sound ever recorded has no scientific explanation]

You might have thought that a planetary feature as singular as the oceanic pole of inaccessibility would be as familiar as the North Pole or the equator. In a sci-fi story, this spot in the South Pacific might be a portal to some other dimension—or possibly the nexus of the universe, as the intersection of First and First in Manhattan was once said to be. Yet at the time of the Bloop, the location of the oceanic pole of inaccessibility had been known and named for only five years.

I have not been to Point Nemo, though it has maintained a curious hold on me for decades. Not long ago, I set out to find the handful of people on Earth who have some sort of personal connection to the place. I started with the man who put it on the map.

Hrvoje Lukatela, a Croatian-born engineer, left his homeland in the 1970s as political and intellectual life there became turbulent. At the University of Zagreb, he had studied geodesy—the science of measuring Earth’s physical properties, such as its shape and its orientation in space. Degree in hand, he eventually found his way to Calgary, Alberta, where he still lives and where I spent a few days with him last fall. At 81, he is no longer the avid mountaineer he once was, but he remains fit and bluff and gregarious. A trim gray beard and unkempt hair add a slight Ewok cast to his features.

After arriving in Canada, Lukatela was employed as a survey engineer. For several years, he worked on the Alaska Highway natural-gas pipeline. For another company, he determined the qibla—the precise alignment toward the Kaaba, in Mecca—for a new university and its mosque in Jeddah, Saudi Arabia. In time, he created a software company whose product he named after the Greek astronomer Hipparchus. This was in the 1980s, when digital cartography was advancing rapidly and civilian GPS systems were on the horizon. The Hipparchus software library—“a family of algorithms that dealt with differential geometry on the surface of an ellipsoid,” as he described it, intending to be helpful—made it easier to bridge, mathematically, three-dimensional and two-dimensional geographical measurements. Lukatela can go on at length about the capabilities of Hipparchus, which he eventually sold to Microsoft, but two of the most significant were its power and its accuracy.

By his own admission, Lukatela is the kind of man who will not ask for directions. But he has a taste for geographical puzzles. He heard about the longest-swim problem from a friend at the Woods Hole Oceanographic Institution and was immediately engaged. You could twirl a classroom globe and guess, correctly, that the oceanic pole of inaccessibility must lie in the South Pacific, probably concealed by the rectangle where most publishers of maps and globes put their logo. But no one had tried to establish the exact location. As Lukatela saw it, the logic of the search process was simple. It takes three points to define a circle. Lukatela needed to find the largest oceanic circle that met two criteria: The circumference had to be defined by three points of dry land. And inside the circle there could be no land at all. The oceanic point of inaccessibility would be the center of that circle.

I’ll leave the computational churning aside, except to say that Hipparchus was made for a problem like this. Drawing on a digitized cartographic database, it could generate millions of random locations in the ocean and calculate the distance from each on a spherical surface to the nearest point of land. Lukatela eventually found the three “proximity vertices” he needed. One of them is Ducie Island, a tiny atoll notable for a shark-infested lagoon. It is part of the Pitcairn Islands, a British overseas territory, where in 1790 the Bounty mutineers made their unhappy home. A second vertex is the even tinier Motu Nui, a Chilean possession, whose crags rise to the west of Easter Island. The character Moana, in the animated movie, comes from there. The third vertex is desolate Maher Island, off the coast of Antarctica. It is a breeding ground for Adélie penguins. The three islands define a circle of ocean larger than the old Soviet Union. Point Nemo, at the center, lies 1,670.4 miles from each vertex. For perspective, that is roughly the distance from Manhattan to Santa Fe.

Lukatela completed his calculations in 1992, and quietly shared the results with his friend at Woods Hole and a few other colleagues. As the young internet gained users, word about Point Nemo spread among a small subculture of geodesists, techies, and the simply curious. In time, new cartographic databases became available, moving the triangulation points slightly. Lukatela tried out two of the databases, each recalibration giving Point Nemo itself a nudge, but not by much.

Lukatela had named the oceanic pole of inaccessibility after the mysterious captain in the Jules Verne novel he had loved as a boy. Submerged in his steampunk submarine, Captain Nemo sought to keep his distance from terrestrial woes: “Here alone do I find independence! Here I recognize no superiors! Here I’m free!”

But Captain Nemo couldn’t entirely stay aloof from the rest of the planet, and neither can Point Nemo. Many of the boats in the Ocean Race carry a “science package”— equipment for collecting weather data and water samples from regions of the sea that are otherwise nearly impossible to monitor. Data collected by their instruments, later given to labs, reveal the presence of microplastics: Even at the oceanic point of inaccessibility, you are not beyond the reach of humanity.

An article this past spring in the journal Nature reported the results of a scientific expedition that bored deep into the sediment of the ocean floor near Point Nemo. The focus was on the fluctuating character, over millions of years, of the Antarctic Circumpolar Current, whose existence became possible after tectonic forces separated Australia and South America from Antarctica. The current helps regulate temperatures worldwide and keep Antarctica cold. But, as the Nature article explained, its character is changing.

I spent several hours recently with one of the article’s authors, Gisela Winckler, at Columbia University’s Lamont-Doherty Earth Observatory, high on the Palisades overlooking the Hudson River. Winckler is a physicist and an oceanographer, and her interest in oceans and paleoclimate goes back to her graduate-school days at Heidelberg University, in Germany. She confessed that she’d first learned about Point Nemo not from a scientific paper but from the 2010 album Plastic Beach, by Damon Albarn’s project Gorillaz. Winckler is intrepid; early in her career, a quarter century ago, she descended to the Pacific floor in the submersible Alvin, looking for gas hydrates and methane seeps. Yellow foul-weather gear hangs behind her office door. On a table sits a drill bit used for collecting sediment samples. Water from Point Nemo is preserved in a vial.

Illustration by Anuj Shrestha

Winckler’s two-month expedition aboard the drilling vessel JOIDES Resolution, in 2019, was arduous. Scientists and crew members set out from Punta Arenas, Chile, near the start of the dark austral winter; they would not encounter another ship. The seas turned angry as soon as the Resolution left the Strait of Magellan, and stayed that way. The shipboard doctor got to know everyone. Winckler shrugged at the memory. That’s the Southern Ocean for you. The drill sites had been chosen because the South Pacific is understudied and because the area around Point Nemo had sediment of the right character: so thick and dense with datable microfossils that you can go back a million years and sometimes be able to tell what was happening century by century. The team went back further in time than that. The drills punched through the Pleistocene and into the Pliocene, collecting core samples down to a depth corresponding to 5 million years ago and beyond.

The work was frequently interrupted by WOW alerts—the acronym stands for “waiting on weather”—when the heave of the ship made drilling too dangerous. Five weeks into the expedition, a violent weather system the size of Australia came roaring from the west. The alert status hit the highest level—RAW, for “run away from weather”—and the Resolution ran.

But the team had collected enough. It would spend the next five years comparing sediment data with what is known or surmised about global temperatures through the ages. A 5-million-year pattern began to emerge. As Winckler explained, “During colder times, the Antarctic Circumpolar Current itself becomes cooler and slows down, shifting a little bit northward, toward the equator. But during warmer times, it warms and speeds up, shifting its latitude a little bit southward, toward the pole.” The current is warming now and therefore speeding up, and its course is more southerly—all of which erodes the Antarctic ice sheet. Warm water does more damage to ice than warm air can do.

Before I left the Palisades, Winckler walked me over to the Lamont-Doherty Core Repository, a sediment library where more than 20,000 tubes from decades of expeditions are stacked on floor-to-ceiling racks. The library was very cold—it’s kept at 2 degrees centigrade, the temperature of the sea bottom—and very humid. Open a tube, and the sediment may still be moist. I wondered idly if in her Point Nemo investigations Winckler had ever run into a bit of space junk. She laughed. No, the expedition hadn’t deployed underwater video, and the chances would have been infinitesimal anyway. Then again, she said, you never know. Some 30 years ago, during an expedition in the North Atlantic, she had seen a bottle of Beck’s beer from an array of cameras being towed a mile or two below the surface. In 2022, in the South Pacific, the headlights of a submersible at the bottom of the Mariana Trench—about seven miles down, the deepest spot in any ocean—picked up the glassy green of another beer bottle resting in the sediment.

Jonathan McDowell has never been to the ocean floor, but he does have a rough idea where the world’s oceanic space junk can be found. McDowell is an astronomer and astrophysicist at the Harvard-Smithsonian Center for Astrophysics, in Cambridge, Massachusetts. He is also part of the team that manages science operations for the Chandra deep-space X-ray telescope. At more or less monthly intervals, he publishes a newsletter, Jonathan’s Space Report, notable for its wide-ranging expertise and quirky humor. He has written about Point Nemo and its environs, and in an annual report, he provides long lists, in teletype font, with the coordinates of known debris splashdowns.

British by parentage and upbringing, McDowell looks ready to step into the role of Doctor Who: rumpled dark suit, colorful T-shirt, hair like a yogi’s. He is 64, which he mentioned was 34 if you count in Martian years. I met him at his lair, in a gritty district near Cambridge—some 1,900 square feet of loft space crammed with books and computers, maps and globes. One shelf displays a plush-toy Tribble from a famous Star Trek episode. A small container on another shelf holds a washer from the camera of a U.S. spy satellite launched into orbit in 1962.

McDowell has been preoccupied by spaceflight all his life. His father was a physicist who taught at Royal Holloway, University of London. As a teenager, he began keeping track of rocket launches. In maturity, McDowell has realized a grander ambition: documenting the history of every object that has left the planet for outer space. Nothing is beneath his notice. He has studied orbiting bins of garbage discarded decades ago by Russia’s Salyut space stations. If a Beck’s bottle were circling the planet, he’d probably know. McDowell estimates that the thousands of files in binder boxes on his shelves hold physical records of 99 percent of all the objects that have made it into orbit. For what it covers, no database in the world matches the one in McDowell’s loft.

Unless something is in very high orbit, what goes up eventually comes down, by means of a controlled or uncontrolled deorbit. The pieces of rockets and satellites and space stations large enough to survive atmospheric reentry have to hit the planet’s surface somewhere. McDowell pulled several pages from a printer—colored maps with tiny dots showing places around the world where space debris has fallen. The maps reveal a cluster of dots spanning the South Pacific, like a mirror held up to the Milky Way.

Guiding objects carefully back to Earth became a priority after 1979, when the reentry of the American space station Skylab went awry and large chunks of debris rained down on southern Australia. No one was hurt, McDowell said, but NASA became an object of ridicule. The coastal town of Esperance made international news when it tried to fine the space agency for littering. From the 1990s on, more and more satellites were launched into orbit; the rockets that put them there were designed to fall back to Earth. The empty ocean around Point Nemo became a primary target zone: a “spacecraft cemetery,” as it’s sometimes called. That’s where Mir came down, in 2001. It’s where most of the spacecraft that supply the International Space Station come down. There are other cemeteries in other oceans, but the South Pacific is Forest Lawn. The reentry process is not an exact science, so the potential paths, while narrow, may be 1,000 miles long. When reentry is imminent, warnings go out to keep ships away.

Illustration by Anuj Shrestha

When I mentioned the conversation between the Mālama crew and its nearest neighbors, the space-station astronauts, McDowell pointed me toward a bank of flatscreens. He called up a three-dimensional image of Earth and then showed me the orbital path of the ISS over the previous 24 hours. Relative to the universe, he explained, the plane of the ISS orbit never changes—the station goes round and round, 16 times a day, five miles a second. But because the globe is spinning underneath, each orbit covers a different slice of the world—now China, now India, now Arabia. McDowell retrieved a moment from the day before. The red line of the orbit unspooled from between Antarctica and New Zealand and traced a path northeast across the Pacific. He pointed to the time stamp and the location. At least once a day, he said, the space station will be above Point Nemo.

[Read: A close look at the most distant object NASA has ever explored]

McDowell is drawn to the idea of remoteness, which maybe shouldn’t be surprising: To an astrophysicist, remoteness is never far away. But, he said, “there are layers and layers when it comes to how you think about it.” In 2019, a space probe relayed pictures of a 22-mile-long rock known as Arrokoth, the most distant object in our solar system ever to be visited by a spacecraft. That’s one kind of remote. More recently, the James Webb Space Telescope has found galaxies more distant from our own than any known before. That’s another kind. McDowell brought the subject almost back to Earth. On our planet, he said, Point Nemo is definitely remote—as remote as you can get. “But I’m always moved by the thought of Mike Collins, who was the first person to be completely isolated from the rest of humanity when his two friends were on the moon and he was orbiting the far side, and he had the moon between him and every other human being who has ever lived.”

Collins himself wrote of that moment: “I am alone now, truly alone, and absolutely isolated from any known life. I am it.”

I joined Hrvoje Lukatela and his wife, Dunja, for dinner one evening at their home near the University of Calgary. Hrvoje and Dunja had met at university as young mountaineers—outdoors clubs offered a form of insulation from the Communist regime. They emigrated together soon after their marriage. In the basement office of their home, he still keeps his boyhood copy (in Croatian) of Twenty Thousand Leagues Under the Sea. Lukatela spread maps and computer printouts on the table as we ate.

Lukatela might wish to be remembered for the Hipparchus software library, but he accepts that the first line of his obituary will probably be about Point Nemo. He is proud of his discovery, and like a man with a hammer, he has a tendency to see everything as a nail. He and Dunja spend part of the year in Croatia, and in an email this past spring, he sent me some new calculations that solve the longest-swim problem for the Adriatic Sea (“with millimetric numerical precision”). Set him down alongside Loch Ness or the Central Park Reservoir, and I can guess what he’d be thinking.

Lukatela has a dream for Point Nemo, though probably not one that he can pursue alone. His hope is that someone, someday, will venture into the South Pacific and leave GPS receivers on Ducie Island, Motu Nui Island, and Maher Island, establishing the location of the triangulation points more accurately than ever before. While they’re at it, they might also drive brass geodetic markers into the rock. Ducie and Motu Nui would be relatively easy to get to—“I could do it on my own,” he ventured. (Dunja, listening, did not seem overly concerned.) Access to Maher Island, Lukatela went on, with its inhospitable location and brutal weather, might require some sort of government expedition.

[From the February 1906 issue: A history and future of human exploration]

What government would that even be? Lukatela indicated Maher Island on a map. Officially, it is part of Marie Byrd Land, one of the planet’s few remaining tracts of terra nullius—land claimed by no one. But Lukatela recalled hearing that Maher Island had recently come under the jurisdiction of one of those start-up micronations that people invent to advance some cause.

He was right. Maher is one of five Antarctic islands claimed by the Grand Duchy of Flandrensis, a Belgium-based micronation devoted to raising ecological awareness. At international conferences, the grand duke, Nicholas de Mersch d’Oyenberghe, wears military dress blues with handsome decorations and a yellow sash. But he answers his own email. Asked about Lukatela’s ambition, he explained that his country is the only one in the world that seeks to bar all human beings from its territory; the thousand or so people who have registered as citizens are all nonresidents. “No humans, only nature!” is the Grand Duchy’s motto. However, he went on, a mission to install a GPS receiver and a geodetic-survey marker would be deemed scientific, and welcomed. The Grand Duchy would be happy to provide a flag.

The astronaut Steve Bowen has orbited above Maher Island and Point Nemo many times. Before being selected by NASA, Bowen was a submariner; he knows a lot about life in a sealed container far from anywhere. He was one of the crew members aboard the International Space Station who spoke with the Ocean Race sailors as their trajectories crossed at Point Nemo. When I caught up with him this past summer, he compared his circumstances and theirs. The astronauts sleep a lot better, he said—in microgravity, you don’t wake up bruised. But the environment never changes. There is no fresh air, no wind, no rain. Bowen remembered the exhilaration whenever his submarine surfaced in open sea and he would emerge topside into the briny spray, tethered to the boat, taking in a view of nothing but water in every direction.

In the space station, Bowen would often float his way to the seven-window cupola—the observation module—and gaze at the planet below. From that altitude, you have a sight line extending 1,000 miles in every direction, an area about the size of Brazil. In a swath of the planet that big, Bowen said, you can almost always find a reference point—an island, a peninsula, something. The one exception: when the orbit takes you above Point Nemo. For a while, the view through the windows is all ocean.

That same expanse of ocean will one day receive the International Space Station. When it is decommissioned, in 2031, the parts that don’t burn up in the atmosphere will descend toward the South Pacific and its spacecraft cemetery.

Last March, aboard a chartered ship called the Hanse Explorer, a Yorkshire businessman named Chris Brown, 62, exchanged messages with Lukatela to make sure that he had the coordinates he needed—the original computation and the later variations. Brown values precision. As he explained when I reached him at his home in Harrogate after his return from the South Pacific, he and his son Mika had been determined to reach Point Nemo, and even have a swim, and he wanted to be certain he was in the right neighborhood.

This wasn’t just a lark. Brown has been attempting to visit all eight of the planet’s poles of inaccessibility, and he had already knocked off most of the continental ones. Point Nemo, the oceanic pole, was by far the most difficult. Brown is an adventurer, but he is also pragmatic. He once made arrangements to descend to the Titanic aboard the Titan submersible but withdrew in short order because of safety concerns—well founded, as it turned out, given the Titan’s tragic implosion in 2023. The ship he was chartering now could stay at sea for 40 days and was built for ice. Autumn had just begun in the Southern Hemisphere when the Browns left Puerto Montt, Chile, and the weather turned unfriendly at once. “Nausea was never far away,” he recalled.

[Read: The Titanic sub and the draw of extreme tourism]

But approaching Point Nemo, eight days later, the Hanse Explorer found a brief window of calm. Steering a Zodiac inflatable boat and guided by a GPS device, Brown made his way to 4852.5291ʹS 12323.5116ʹW. He and Mika slipped overboard in their wetsuits, becoming the first human beings to enter the ocean here. A video of the event includes photos of the men being ferociously attacked by an albatross. While treading water, they managed to display the maritime flags for the letters N, E, M, and O. Then, mindful of Lukatela’s further calculations, they headed for two other spots, a few miles distant—just to be safe. Admiral Robert Peary’s claim to have been the first person to reach the North Pole, in 1909, has long been disputed; his math was almost certainly off. Brown did not want to become the Peary of Point Nemo.

He isn’t, of course. I think of him, rather, as Point Nemo’s Leif Erikson, the man credited with the first New World toe-touch by a European. I think of Hrvoje Lukatela as some combination of Juan de la Cosa and Martin Waldseemüller, the cartographers who first mapped and named the Western Hemisphere. Jonathan McDowell is perhaps Point Nemo’s Alexander von Humboldt, Gisela Winckler its Charles Lyell and Gertrude Bell. Steve Bowen and the Ocean Race crew, circumnavigating the globe in their different ways, have a wide choice of forebears. The grand duke of Flandrensis may not be Metternich, but he introduces a hint of geopolitics.

Unpopulated and in the middle of nowhere, Point Nemo is starting to have a history.

This article appears in the November 2024 print edition with the headline “The Most Remote Place in the World.” When you buy a book using a link on this page, we receive a commission. Thank you for supporting The Atlantic.

Subscribe here: Apple Podcasts | Spotify | YouTube | Overcast | Pocket Casts

The transformation of several East Asian countries from developing, agrarian economies to highly developed, industrialized economies is one of the great success stories of the 20th century. According to one World Bank report, almost a billion people were pulled out of poverty as a result of fast growth in the region. But the questions of why this happened and how it can be replicated by other countries remain essential to answer for the roughly 700 million people trying to survive on less than $2.15 a day.

In today’s episode of Good on Paper, I talk with Oliver Kim, an economist working at Open Philanthropy, whose recent paper co-authored with Jen-Kuan Wang, a Ph.D. student at Penn State, investigates one country that was part of the “East Asian miracle”: Taiwan. In the 1950s, Taiwan pursued a series of land reforms that were widely credited for transforming its economy. Other countries in the region had pursued similar reforms—including mainland China, Japan, and South Korea—adding to the sense that these specific changes were important for understanding the region’s development. In broad terms, the story went like this: Taiwan redistributed land to the peasantry, which significantly increased the nation’s agricultural productivity and helped finance the country’s industrialization.

But Kim and Wang’s research casts doubt on this story. Diving into the data reveals a far more complicated picture of how land reforms spurred development in Taiwan, with implications for developing nations around the world.

“It is definitely true that Taiwan got richer during this time period,” Kim explains. “But you also have to remember from a critical historical perspective that this is playing an important propaganda role. And so a lot of Taiwanese historians, though they didn’t have the data that we have, have been questioning this narrative.”

He adds, “And actually, if you think about the experience of land reform more broadly in a global sense, I think our results actually bring the Taiwanese experience closer to the global experience, which is generally defined that land reform actually has been fairly disappointing in terms of its productivity impacts.”

The following is a transcript of the episode:

[Music]

Jerusalem Demsas: How does a nation pull its residents out of poverty and into the developed world? I think this is the most important question in economics, and it’s one researchers have struggled to answer.

To development economists, the rise of South Korea, Japan, and Taiwan, among others—what’s often called the East Asian miracle—has been a source of deep fascination. How did these countries so quickly enter the ranks of the global elite?

On today’s episode, we’re going to focus on Taiwan. How did this country go from a Japanese colony to an advanced industrial economy? And what lessons does it hold for other developing nations?

Over the course of the 1950s, Taiwan’s agricultural productivity took off, setting the stage for its transition to an industrial economy. Over essentially one decade, rice yields grew by more than 40 percent, unlocking a period of rapid economic growth. The traditional narrative is that land reforms are the key to development, particularly a set of reforms that redistributed land from wealthy landlords to the disaffected peasantry and thereby increased productivity.

It’s a nice story, one that puts equity and efficiency on the same side. But a new study casts doubt on whether this story is actually true.

[Music]

This is Good on Paper, a policy show that questions what we really know about popular narratives. My name is Jerusalem Demsas. I’m a staff writer here at The Atlantic, and today I’m joined by Oliver Kim. He recently co-authored a paper that is challenging deeply held beliefs about how development works.

His white paper shows that while redistribution of the land might have been a great policy on its own terms, when you dig into the data, it doesn’t explain Taiwan’s sudden burst of productivity. Other explanations make a lot more sense. We’re going to dig into them today.

All right. Oliver, welcome to the show.

Oliver Kim: Thank you for having me.

Demsas: So we’re here to talk about a very interesting new paper you wrote, but because it’s quite specific, I want us to step back a bit. There’s fundamentally one big question that development economics is trying to answer, and it’s: Why do some countries grow into developed nations where their residents are able to access high standards of living, and why do other nations fail to do so?

And one of the most important debates is centered on the divergence of East Asian countries like South Korea, Japan, and places like Taiwan from places in Latin America, Africa, and Southeast Asia. I’m hoping you can just situate us for a second back in the early 1960s. What was the state of these countries that we now think of as highly developed nations, at par or ahead of many Western nations, relative to what we consider developing nations today?

Kim: Sure. At the end of the Second World War—outside of basically Europe, North America, some other European offshoots—the basic condition in most of the world was poverty. And since then, in the present day, outside of basically if you’re lucky enough to find yourself sitting on a giant pool of oil, the only countries to really sustainably grow their economies to high-income status are really the East Asian tigers—Japan, South Korea, Taiwan, and then Singapore and Hong Kong.

So I think one of the central questions in development is, Why did this miracle happen? And one of the primary reasons that’s been put forward by economists and economic historians is land reform, which is basically redistributing land from landlords to peasant farmers. Right around the end of the Second World War and the beginnings of decolonization, all these major East Asian economies had large, sweeping land reform. Japan had one around 1947 under American occupation. South Korea, similarly, had one in 1950. And Taiwan had perhaps the largest and most sweeping land reform, which occurred from 1950 to 1958.

And the logic behind land reform is fairly simple. If you think about what most developing countries do—if you go and visit a developing country or area, what most people do to survive is: they farm, right? So I think in the present day—I think the latest stat’s from 2022—something like 40 percent of the world’s population are farmers. And this is where most of the extreme sort of poverty is.

In Asia, immediately in the postwar period, agricultural productivity went up in the 1950s and early ’60s. The most famous example is that in Taiwan, rice yields went up by 40 percent over the course of the 1950s. And if you think about the rural poor, this is a huge increase to their incomes. And so a very influential sort of view in economics is to try and connect these two things and say that land reform had something to do with the growth in agricultural productivity.

And so this was famously articulated by Joe Studwell in his book, How Asia Works. And the idea is that if you basically redistribute land—you take it away from landlords who happen to own a lot of it, and you give ownership to the peasants who actually work on it—you can improve productivity. And so you can get something that’s actually very rare in economics: You can get something that’s good for both equity and efficiency. And so the idea is that East Asia had these large, sweeping, major land reforms, and other developing areas in the world didn’t. And so this was a major contributor to the East Asian divergence.

Demsas: Okay. This story then is about landlords, and by that we’re talking about landlords of agricultural land. And so essentially, at the base, it’s just redistribution, right? You’re talking about redistributing land from these large landholders to people who are farming small hectares of land.

I’m hoping you could actually walk us through, more specifically, what this is, because I know land reform looks very different in different places, and we’re largely going to be talking about your research in Taiwan. So what was land reform there? What were the three phases? And which is the important one?

Kim: Yeah. Land reform is this big, amorphous term in development, and it means a lot of different things to a lot of different people. But I think most people would agree that the most sort of meaningful form of land reform is what you described, which is redistribution.

I guess the model you can have in your head is that in a lot of developing countries, there are these landlords who own a lot of the land, but they don’t necessarily work it themselves. And in the worst case, you have absentee landlords, who don’t even live on the land that they farm, and they have poor, oppressed peasants who are basically doing all the work and tilling the land. And the idea behind land reform—which I think is pretty compelling, at least on the face of it—is that you take that land, and you give the peasants who actually work it rights to that land. And in the most sweeping case, you give them ownership of that land.

And so in Taiwan, as you mentioned, there were several different phases of this. Just to set the historical background, Taiwan has this very unique political history. The government that runs Taiwan, the Republic of China, is basically an exile regime. And so you had Chiang Kai-Shek’s Nationalist Party, the KMT regime, that basically lose the civil war to Mao Zedong on the mainland of China, and they fled to Taiwan with something like a million of their supporters. And so this was a very sort of foreign regime that is almost a neocolonial setup, where they actually had very little ties to the local Taiwanese population, and they’ve just been kicked out by what was essentially a peasant revolution on the mainland of China.

And so you have this foreign regime coming into Taiwan, and they need to build their own domestic base of support. And they had this very recent experience where they were not very popular amongst the peasants. The KMT’s base on mainland China had been amongst landlords. It had been under urban elites. And so they needed to do something different, basically, to avoid getting thrown out a second time.

The first thing that they did was that they capped rent. So this happened in 1949. This is almost like rent control in a modern sort of rental sense. They capped the amount that these tenant farmers had to pay to the landlords. This is actually a fairly limited form of land reform, and it didn’t really do that much.

So the second thing that they did was: They just kicked out the Japanese, who had previously been the colonial overlords of Taiwan. And they took all the land that had previously been held by the Japanese, and they redistributed it to the peasant farmers. And that redistributed something around 20 percent of all arable land in Taiwan.

This still didn’t go far enough from a political perspective. And so the last stage, which ran from 1953 to the ’60s, is that they took any privately held land that was over a certain cutoff—this varied a little bit in terms of the quality of the land, but the most common cutoff was around three hectares for rice-paddy land—and they basically said, Any holding that is over that is going to get split up. So the landlord can retain up to three hectares, and then anything above that gets redistributed to the tenants who farm it. And these three phases collectively basically redistributed about a quarter of all arable land in Taiwan. And this ended up being, I think in a global sense, one of the most sweeping land redistributions in terms of the share of land that was redistributed.

Demsas: Yeah. You find that in all three stages—the three stages of land reform redistribute over 215,000 hectares of land. And for people like me who had to Google what a hectare was, that’s almost 2.5 acres per hectare. So that’s over 500,000 acres of land redistributed. So this is pretty massive. This is not some small amount of shift that’s happening here.

I appreciate you brought the historical perspective here, because part of the question that I had when I was learning about this is: Why didn’t the landholders become a political problem for the government, as well? There’s a lot of power that’s held by landholders. What did they compensate them with?

Kim: Yeah. So part of this has to do with the elite split that I was describing, where the KMT regime that comes over with a million followers—a lot of them who are soldiers with guns—they don’t feel a lot of connection to the local Taiwanese elites, the landlords there. And the KMT regime pretty quickly turns out to be very unpopular.

And so the Taiwanese actually rise up in 1947. It’s this deeply traumatic event in Taiwanese history called the 228 Incident. Local Taiwanese rise up. There’s a huge uprising throughout the island. They take over major cities. And the KMT, which still controls part of the mainland at this time, has to rush troops over, and they basically end up killing tens of thousands of people. This basically massacres a lot of the traditional Taiwanese elite. So that’s an important feature here, where there was very strong political and military coercion that essentially crushes what would be the nascent elite in Taiwan.

And so the split between the fact that you had this almost carpetbagger regime that came over, that had very little connection to the local Taiwanese—that enabled it. And there was some compensation that the KMT did. They did pay the landlords, in theory, for the land that they redistributed. But these payments actually, I guess, were not that economically significant in the final analysis.

Demsas: And so politically, this helps pacify the population. But that’s not the main mechanism by which most people think of land reform as having helped lead to the East Asian miracle, right? And it’s not just in Taiwan. As you mentioned, in the late 1940s, Japan and South Korea have similar land reforms. And in the 1970s, mainland China also has land reform.

And the mechanism is usually that land reform leads to greater agricultural productivity, right? So then you get a ton more productivity. That kind of output growth spurs the ability for people to move into manufacturing sectors. Is that the story? Can you break that down for us?

Kim: Yeah. The traditional story, which was famously articulated by Joe Studwell, is that land reform boosts agricultural productivity. And there’s a number of different mechanisms by which this can be true. But the most common one is that as you shrink the sizes of farms, paradoxically, yields—which is agricultural productivity over a unit of area, so the amount of stuff that you can grow over a hectare of land—that tends to go up.

And this is a deeply disputed sort of relationship in economics. There are a lot of advocates on one side. There are advocates on the other. But the historical view is—at least, in East Asia, in Taiwan and Japan—breaking up these land holdings that the landlords previously held and giving them to tenants had this effect of raising agricultural productivity.

And what my research basically finds with Jen-Kuan Wang (we’ve recently released this working paper) is that if you actually digitize the data—and this hadn’t been done before—but if you actually go into the Taiwanese archives, and you look in a very straightforward way at what actually happened in townships where there was more land reform, do you see more agricultural productivity growth? Do you see rice yields go up more?

We actually don’t really find a lot of strong evidence that was the case. So what we find, basically, is that in areas where there was more of phase two of the land reform—so remember, that’s where you redistribute the previously Japanese-held public land—in areas where there was more of that public land redistribution, you actually do find some small effect on rice yields. So rice yields go up by around 6 percent over 10 years as a result of this phase two of land reform.

Demsas: Is that because they’re converting things from sugar to rice?

Kim: Yeah. The idea there, basically, is that these are previously colonially held lands. The Japanese were previously mandating that you grow sugar. This was essentially feeding the Japanese empire, in a large sense. And once you lift that constraint, and you make farmers free to grow what they want, they choose to plant rice instead of sugar. And this is a classic economically liberal argument, but: You give people the freedom to choose, and they make more efficient decisions.

But the reform that everyone tends to focus on more—because it’s very rare, actually, that you have a case where you have a colonial overlord that gets kicked out, and you can redistribute their property. That can only usually happen once. The case that people typically fixate on in the case of Taiwan is phase three of land reform—this redistribution. And this obviously has a lot of resonance in present-day debates whenever there’s a lot of inequality in the world.

And the idea was that, basically, by taking land away from these landlords who own these larger estates, and you give them to these peasants, you boost agricultural productivity. But when we look in the data, when we go into the township-level data, and we run regressions on the stuff, we actually really can’t find this effect.

Demsas: So much really good economics is just in the data, right? It is in finding and cleaning data, actually going and testing theories. And one of the things that you make clear in your white paper is that this theory had not been tested by looking at the actual records on the ground in Taiwan. So can you walk us through what you did there and how you found that? Did not have access to this before?

Kim: Actually, it is a little surprising that this hadn’t been done before, because a lot of this was actually just sitting there. My co-author—I mentioned who I wrote this paper with, Jen-Kuan Wang—is Taiwanese, and he did a lot of digitization work, going into the government data archives. But the basic work, the initial regressions that we ran, actually it was just a book I pulled from the UC Berkeley library.

Demsas: What was the book?

Kim: The book was—they actually had, at the township level, the land-distribution statistics. So they had for each township in Taiwan, you know, How many people own plots that are between, like, zero and 0.5 hectares? Between 0.5 and 1 hectare? etc.

And so this very sort of commonly held story that this phase three of redistribution, that land-to-the-tiller reforms boost agricultural productivity—it doesn’t actually seem to be the case. And one reason that we hypothesize is, basically, that the farms that this created were just too small, actually, to be economically viable.

And so if you actually think about the context here—Taiwan is actually a pretty small place—the initial land holdings held by these landlords were just relatively small, and they just got broken down to an even smaller extent, to something like 0.2 hectares, which is, I think, less than an acre.

And these are just too small to support a growing family. And this had the perverse effect, basically, of pushing labor out of agriculture and into industry because these farms are just too small to support the families that were living on them. So this is very different.

Demsas: So people sold their land and moved to the city?

Kim: They often kept it, basically. In a lot of developing societies, I think there’s a lot of attachment to the land as a form of security, but it was just not enough to provide enough income to support a whole family. And so you had people leaving agriculture and starting to pursue part-time or even full-time employment in manufacturing, and that’s maybe related to the broader sort of transformation of Taiwan that people typically think of when they think of the East Asian miracle.

Demsas: And I want to ask you first: Were you surprised by the results of your study? Or did you expect to find productivity growth?

Kim: Yeah, I did. I mean, people on Twitter and stuff have been like, Oh, there’s been a lot of fighting, over, like, Oh, what this means for the Joe Studwell book, but like, Oh, this was an attempt to take down How Asia Works.

But I actually really like that book. And I was motivated to do this because I really did believe the story that land reform is good for productivity, and it had all these sort of positive effects on East Asian development. So yeah, I did go into this expecting to find a very strong result of land reform. So it was surprising to say the least.

Demsas: Okay, cool. I also enjoyed the Studwell book, so I’m glad I don’t have to throw that one out.

So there’s this question about whether small farms would be more productive than big farms, which is actually really, really important. In general, you would expect there to be economies of scale that come with bigger farms: You’re able to diversify your crops more in case of a problem. You’re able to invest in productivity-enhancing technology, whether that’s irrigation or whatever it is.

At some level, do you feel like the original land-reform story felt suspicious on face?

Kim: Yeah. So one thing that’s very interesting, particularly if you dive into the Taiwanese literature, is that local Taiwanese scholars are very skeptical of this story. And I brought up this political context, right? There’s a split—the KMT come in; they’re this foreign regime. This was a very brutal authoritarian regime. In a modern sense, this was an autocracy, right? And they need to build their local base of support, and so they play up this idea of land reform being this benevolent policy that was so good for the Taiwanese people, because it suits their interests, right?

Part of that might be true. It is definitely true that Taiwan got richer during this time period. But you also have to remember, from a critical historical perspective, that this is playing an important propaganda role. And so a lot of Taiwanese historians, though they didn’t have the data that we have, have been questioning this narrative and have been pointing to examples of the fact that, Hey, a lot of these farms were actually really small. A lot of people were actually pushed off the farm. Maybe the sort of miracle doesn’t have actually that much to do with KMT policy.

And so, yeah. I think adopting this critical historical perspective and saying, Hey, something about the story doesn’t seem quite right. And actually, if you think about the experience of land reform more broadly in a global sense, I think our results actually bring the Taiwanese experience closer to the global experience, which is generally defined that land reform actually has been fairly disappointing in terms of its productivity impacts.

Demsas: And I think that brings me to this broader observation that I’m hoping you can expand on, which is that so many debates about development, especially development—these are obviously developing nations—are being shoehorned into debates that are happening in Western countries about what sorts of economic policies are better.

And so in some ways it seems like there’s this larger debate about whether industrial policy, whether the government intervention in the market is an appropriate way to get to development. And then you have folks on the other side— the Washington Consensus, the more neoliberal economists—who are saying, No. We want free and open trade. We don’t want government intervention. And the industrial-policy people really, really like this land-reform story because it shows that there is a role for government to be intervening in the market. Is that how you see it, too?

Kim: Yeah. I think there’s always a bit of a danger from trying to transpose lessons in developing contexts to developed contexts, right? They’re deeply different in very fundamental ways. On the industrial-policy stuff, the example that everyone likes to bring up is: South Korea did a ton of industrial policy and experienced this massive developmental miracle, right? So there was massive intervention. The state heavily subsidized certain sectors—like steelmaking, like auto production, like shipbuilding—and then created these very successful industries and very successful firms that were able to compete on the global marketplace.

And I think there’s a real strong sort of temptation to try and take that logic and apply it to things like CHIPS [and Science Act] in the United States. I’m not saying, necessarily, that’s wrong. But you have to remember that the context in which Korea is implementing these policies is very different than what the United States is trying to do. Like, Korea when it was trying to build its domestic steel industry, or if you want to lump agricultural policy, so like Taiwan when it’s trying to develop its local agriculture—these are not new, frontier industries. These are not at the technological frontier.

You broadly know that you should be growing more food. You broadly know that a developed country, at least in the ’60s or ’70s—the hallmark of being a rich country is that you make your own steel, right? It’s not a mystery. Especially in Korea and Taiwan, which were former Japanese colonies, you can look over the water, and you can see your former colonial overlord, Japan, that’s doing pretty well economically by developing its domestic industries, doing all this industrial-policy stuff. And so it’s fairly simple to mimic that policy recipe and try to develop an industrial mix that catches up to countries that have already reached the frontier.

With things like CHIPS and modern industrial policy or things like electric cars, it’s much more difficult to have a guide in that sense. It’s unclear what the hallmarks of a highly rich society in the 21st or 22nd century is going to look like. So I think there’s a fundamental difference there. When you have the playbook and you have clear sort of historical examples—economists call these demonstration effects—when you have a demonstration effect that you can point to, the problem of what industrial sectors to prioritize becomes a lot easier.

[Music]

Demsas: All right. We’ll have more with Oliver after this break.

[Break]

Demsas: I want to step back broader and talk about why this really even matters. One mechanism that you guys are really undermining in your paper is the idea that land reform led to the East Asian miracle by increasing agriculture productivity.

But you hinted at this already, but there are other ways in which land reform could have helped propel Taiwan, South Korea, all these other countries into becoming a developed nation, into these high-GDP-per-capita countries. I’m hoping we can talk through some of those, because the elimination of that one causal pathway doesn’t actually mean we can say land reform isn’t important for understanding Taiwan’s or other countries’ pathway to development.

So I guess one pathway, which you hinted at, is a sort of political pathway, that you’re able to pacify the population of peasants who have already had this really, as you mentioned, traumatic uprising, have martial law now imposed, but you need to actually make sure that they’re not going to overthrow you. And so it staves off political demands for communism. Is that a pathway that you think is super valid for understanding what happened here?

Kim: Yeah. I think that’s central. I think you can’t really understand the history of East Asian development without understanding the fact that this all is happening in the context of the Cold War, right?

So I mentioned that Japan, South Korea, Taiwan all had these land reforms. These are all non-communist countries. But if you look at the map of East Asia, and you put the dates of land reforms across all countries—not just these three—on a map, you’ll notice there’s this kind of mirroring pattern. Japan has a land reform in ’47, but China also has one from ’45 to the ’50s, as Mao’s Communists take over. South Korea has one in 1950, but that’s because North Korea has one a couple years earlier.

And so the motivations, I think, are fairly obvious, which is that these are poor societies where most of the people are farmers, and across the border, in a lot of cases, you have very viable military threats, which are doing actually fairly popular programs of land redistribution that build up the support among the peasantry.

And so this is a very unique circumstance where the United States supports a lot of these very strong redistributionary measures in order to basically forestall the possibility of communist takeover. And this is not the kind of thing that’s very easy to do in a regression, like a formal statistical sense, but just knowing the historical context and the narrative of this history, it’s hard to see how these regimes could have survived without at least doing some form of land redistribution, at least building up some support amongst the farmers that made up the majority of their population. If these regimes had failed to survive, and they’d been swallowed up—let’s say Taiwan had been consumed by mainland China—you wouldn’t be talking about an East Asian miracle, probably.

And the exception that proves the rule is South Vietnam. So people forget that there was this other kind of American proto-colony in Southeast Asia. Just like Korea, Vietnam gets divided into two. There’s a communist northern half. There’s a capitalist southern half. And for particular reasons, the South Vietnamese regime basically avoids doing systematic land reform. And the American government very famously spends a lot of effort, blood, and treasure in this horrific and tragic war, which is called the Vietnam War, basically, trying to solve through coercive means what ultimately, I think, is an agrarian and a peasant sort of revolt. And part of the problem is that the South Vietnamese regime is beholden to the landlords. It’s a landlord-led government. And they’re not particularly interested until very late in the war in doing large-scale redistribution, because that’s where their wealth is tied up.

And what basically happens, of course, is that the South Vietnamese state gets hollowed out. It ultimately falls to a conventional invasion from the North, but it’s hard to argue, I think, that if they hadn’t done land reform earlier, they might have continued to exist. And failing to have done systematic land reform basically results in the regime failing to survive. And I think the East Asian states that did successfully do land reform avoided this fate.

Demsas: Okay. It sounds like you place a lot of value in that story. But as you mentioned, Joe Studwell, who writes How Asia Works—he has a few other pathways that he also theorizes. One is that landlords are forced to go do more productive things than just rent seeking. Previously before land reforms, you’re a landlord, you have a bunch of money, you have peasants working the farm, you get to raise your rents really high, and you have no incentive to really invest in that land and to increase productivity, because you’re just making a lot of money by just collecting those rents and not having to do anything else. Once your land is forcibly redistributed, in order to—you want to still be rich? You want to make money? You have to go do something more productive. Do you find that to be a valuable way of thinking about this?

Kim: Yeah. We’ve tried to find evidence for this. So the story in Taiwan is that, in compensation for the land reform, the landlords were compensated by the state, either in the form of land bonds, which basically is a claim on agricultural output. So I don’t know—you’re entitled to some number of kilograms of sweet potatoes in the next five years or something like that. My understanding is that those actually turned out to be fairly worthless.

And then the second thing that landlords were compensated with were shares of industrial companies. So Taiwan had a little bit of industry that had been leftover by the Japanese. The KMT state claims it, and they give shares to the landlords. And I think this is a little bit of what Joe Studwell is gesturing at. I’m a little bit more skeptical that this is as developmentally important, because one of the things that happens in Taiwan is that a lot of the growth in manufacturing—there is growth on the state-led sector, but the really most dynamic kind of growth-leading sectors of the economy are small-scale manufacturing.

And so these don’t really have much to do with the former—like, these lumbering, large-scale industrial enterprises that the government had redistributed these shares of. It’s more these new enterprises that sprout up in the countryside. And so that, I think, is maybe evidence against this view that it’s the landlords, in particular, who are the importance of entrepreneurs in development.

Demsas: Okay. To recap land reform: Your belief is that it largely is operating positively for growth by staving off communism and, potentially also, the low yields lead people to work more manufacturing. But you don’t really believe that there is some large productivity increase that comes from redistributing the land to the peasantry.

Kim: Yeah. That’s right. And I think this helps reconcile a little bit Taiwan with the global historical experience, right? It wasn’t unknown that South Korea and Taiwan did this. Throughout the ’70s, there were actually a lot of attempts at land reform in Latin America, elsewhere in Southeast Asia.

Now, there are a host of different reasons that this didn’t work. Often, as I mentioned, as I alluded to, in the case of South Vietnam, for political reasons, a lot of these land reforms ended up getting co-opted. But even in the instances where there was genuine redistribution, there just never were sort of the claimed productivity effects that you saw, or at least you claim to see, in South Korea and Taiwan.

And at least in the Taiwanese case, the study brings Taiwan back into line, I think, with the rest of historical experience. It’s not to say that agricultural productivity didn’t go up. Clearly, Taiwan got richer. Rice yields went up. But I think the balance of explanations and how much explanatory power you want to put on land reform versus more prosaic things—like the fact that there were, you know, improvements in fertilizer use, the fact that there were the development of higher-yield varieties sort of presaging the Green Revolution—these sort of more prosaic, technical developments probably are more important than this sort of very flashy, large-scale institutional change of land reform.

Demsas: But I also want to take a step back from land reform in particular, because in general, if you are downgrading the importance of one explanation for East Asian development, then you should be upgrading other explanations for East Asian development, right?

So I wanted to talk about some of the other reasons people often have for why nations develop or why nations fail, to quote Acemoglu, and see what you think of it now that you’ve done this study. Okay, let’s start with export discipline. Can you explain what that is and whether you feel more confident that it’s a reason for East Asian development?

Kim: Yeah. The idea there, basically, is that when you think of the East Asian countries, you typically actually think of manufacturing development. There’s actually this Calvin and Hobbes comic, I think from 1980s, where Calvin asked Hobbes to check the tag of his shirt because it says where he comes from. And Hobbes checks the shirt, and it says made in taiwan. It’s like he came from Taiwan.

I mean, the idea that Taiwan was making shirts back in the ’80s and now is doing things like TSMC and making microchips—that’s a tremendous transformation. But the common thread throughout this stuff is that it’s about making physical goods, industrial production that gets exported abroad. And the idea with export discipline is that there was a lot of industrial policy throughout the developing world throughout the ’60s and ’70s. And it was actually the policy consensus that the state has to do things to foster development.

And Joe Studwell’s argument in How Asia Works is that the differentiating kind of factor for the East Asian economies was the fact that they actually forced their big sort of industrial champions to export. And this basically disciplines the firm when a firm can just stay in the domestic market. So imagine you’re surrounded by big, tall tariff barriers; there’s really no incentive to innovate or to try and become more efficient.

Demsas: You have a captive audience.

Kim: You have a captive market. You have a captive audience. And in the comparative sense, this is what a lot of scholars view happened, for instance, in South Asia and India, in a lot of Latin American countries, where you had initially very well-intentioned policies, where it’s like, Okay. We want to foster an infant industry. We want to get industrialization going. To support that industry, to prevent it from just being strangled in the cradle, we need to have all these supports. We need to put up tariff barriers. We need to subsidize credit, all this kind of stuff.

Demsas: To make it easier for these industries not to have to compete with—

Kim: Yeah, to make these industries grow and to get them to a sufficient size that economies of scale kick in.

This is a very standard kind of economic argument, and it’s actually fairly sound. The problem is the political stuff, which is that once that firm is born and grows a little bit, it starts to create a constituency that actually really likes to have these barriers. It’s really nice to not have to compete with all these scary, big foreign firms that are potentially more efficient than you are.

And so the view is that what happened in South Asia and Latin America is, like, a lot of these controls and these barriers stayed up, and the firms are never forced to export—unlike, for instance, the Hyundais and Kias of the South Korean experience. In other countries, they never developed the productivity enhancements to become globally competitive, and so you didn’t have the sort of miraculous growth. I think this view is basically still pretty sound.

Demsas: So I think the more-popular ones that people generally tend to hear when they think about development and why it happens are these temperature and education explanations for development. So: Hotter places don’t develop well. And also, people should just invest in education. That’s why the East Asian tigers do so well, is because they have such well-educated populations. How do you think about those explanations?

Kim: Yeah. The temperature one is interesting. I think that goes back to Montesquieu or whatever.

Demsas: Yeah. It’s an old one.

Kim: Yeah, like, tropical. There’s always a little bit of, That’s a little suspect, in there.

Demsas: Oh, yeah.

Kim: I mean, Taiwan is pretty hot. Singapore, which I grew up partly in, is really hot. I don’t think there are very many economists who would seriously defend it. In a more analytical sense, Taiwan, Singapore, Vietnam, which is growing very quickly—they’re tropical climates, essentially. And so I don’t think that climate can necessarily be, like, the best explanatory sort of variable there.

The second one that you mentioned is education, right? Which I think is still very live as a hypothesis, right? And so the idea is that maybe because of cultural factors, like the sort of legacy of Confucianism, this legacy of meritocratic exams in these Confucian regimes, East Asian cultures valued a lot of education. And so that carried over to the modern period, where the educational attainment was relatively high. I think that education is obviously really, really important. I think it’s important from a rights basis. As an individual, it improves your earnings.

It’s a little bit more unclear to me how this works out at a societal level. So when you move from the specific, the micro, to the macro and the aggregate, this is a huge live debate in the development literature—how much, actually, educational attainment has gone up in the developing world.

Clearly, actual growth experience has been kind of disappointing, especially over the past four or five years. At least on paper, there have been a lot of reforms—understanding the East Asian experience, trying to apply it—there have been a lot of reforms trying to promote free primary education, particularly in sub-Saharan Africa. There’s at least a lot of rhetoric from governments that they should be prioritizing education. Whether that has actually translated into increases in learning—like the actual test scores, for instance—that is a huge open question. And I don’t think there’s very good sort of international data, particularly from these very poor, developing countries, to understand how much these test scores have gone up.

There is also a lot of political incentives. I think Justin Sandefur from the Center for Global Development has an interesting paper showing that when Western donors attach a lot of conditions, basically, to—let’s say you want to hit certain targets in vaccination rates or something like that—the official government data goes up a lot, but when you actually do a survey, it doesn’t show that much of an increase.

So there’s a lot of reporting incentives, also, to say that educational attainment goes up a lot. So that’s just a very long tangent to say that I think there’s a strong case to be made that this was an important component of success in East Asia. Whether it’s on its own enough to explain stuff, and whether it can explain growth in development more broadly, in a global sense, the jury is still out on that.

Demsas: And then, the last one is sort of—I mean, the Why Nations Fail is the big developmental econ book that gets popular amongst noneconomists, too. And to oversimplify Acemoglu’s thesis here, there are these inclusive institutions that make development possible. Part of why I think that framework can be helpful here is that when you think about why a country is able to pursue or willing to pursue land reforms or is not able to or willing to pursue it—even if they have a lot of the information, there’s a level at which it’s almost like, Yeah. There are places where people can pursue good economic policy for kind of amorphous institutional reasons. And that’s very, very hard to isolate as a thing that you can tell countries to do.

Kim: Yeah. This is sometimes phrased a little bit pejoratively as the “get a better history” kind of view. Which is that, Oh, there are these long-standing, almost fundamental historical differences that determine your economic destiny. And there’s very little that you can do about that. Obviously, history matters. Countries don’t just drop out of coconut trees. They exist in the context that came before them, right? But history is not everything.

There was scholarly work, for instance, done about China and Confucian societies in the 1950s with the view that’s very different from today, which is that Confucian societies are actually not compatible with modern economic development.

So if you’re taking the vantage point of somebody from the 1950s, you just see in Chiang Kai-Shek’s regime collapse on the mainland, it’s not even like Korea and Taiwan are doing particularly well in a global sense. Like, the miracle stuff is sort of very nascent. And so it’s a very rational kind of response, as a sociologist, to say, Hey. Maybe these Confucian regimes are just not very good at adapting to the modern world.

Fast-forward 30 years, you have Lee Kuan Yew, the prime minister of Singapore, making a very forceful case globally, going around to different countries and saying, Hey. The reason you’re not doing well is because you don’t have Confucian values. Confucian societies—we value education. We’re more respectful of the government. We’re more organized, all this kind of stuff.

And so I think a lot of these more-ingrained cultural and institutional factors are actually not so ingrained when you take a broader view of history. They’re more malleable than people can make them out to be. And so there’s always this game that you can play about, Are institutions inclusive? Or are they exclusive?

So a first approximation looking at Taiwan in the 1950s, this period, which we’ve been describing as the period where the miracle is laying down roots, this is an autocracy. This is an authoritarian, right-wing, military dictatorship. This is not like a democracy where—actually, the other famous example in How Asia Works and in the history of development is South Korea, which we know is the example of a miraculous country.

One of the first things that happened under the dictator Park Chung Hee was that he took a lot of the prominent capitalists, and he threw them in prison. And he basically threatened to confiscate their property unless they started to dedicate their resources to the goal of national development. And throughout this period, there was a lot of state intervention in the economy. And so this just, prima facie, does not sound like inclusive institutions in the “democracy is good” kind of sense.

This is not to condone dictatorship and autocracy as a route to development, but it’s just saying that history is complicated, and, like, our views of this stuff kind of change depending on our vantage point. And I think it’s, maybe, a little bit of a blind spot to think of these things as too fixed, and they can move around, and there are opportunities I think that exists for nations to reshape their economic destinies.

Demsas: And so what’s really important about your paper—and I was reading through some of the other economists that have really focused on land reform in explaining East Asia’s divergence from Latin America or other countries or Southeast Asia—it’s almost like other countries have this template. They need to do land reform, and then they can develop. I wonder if you can talk to us a little bit about: Why wasn’t it possible for other countries to pursue this? And do you think that if they had, they would have gotten the benefits? Or is it kind of like, this is not the one-trick pony for development?

Kim: Yeah. So I think the political factors are really important to foreground here. There’s a scholar, I believe at the University of Chicago, called Michael Albertus who has a great book called Autocracy and Redistribution, which I think of as the canonical theory of, Why does actual, large-sweeping instances of land reform happen? And the answer that he basically gives is that land reform occurs when there’s a split in the elites, right? So as I mentioned, this is a fact you should always bear in mind.

What Albertus basically posits is that what you need to happen is that there’s a split in the elites. And so you have one group that advocates for reform and then separates from the landed class. And so this is what happened in East Asia, right? So I mentioned Japan and South Korea had these land reforms. Well, who actually implemented the land reforms? It was the American Army, in essence, occupying these countries in the aftermath of the collapse of Japan after World War II. In Taiwan, it was less an American military presence, but it was actually, again, this foreign KMT regime that didn’t have very close ties with the local elites that was free, basically, to impose a land reform through this essentially autocratic, authoritarian means.

And so there’s this weird kind of thing where, actually, authoritarian regimes—Albertus points out—are actually, in a lot of cases, more likely to engage in land reform because they have the capacity and the willingness to disempower and remove property from the main political class, who in a democracy might actually have more of a voice.

Demsas: Yeah. That reminds me of my favorite economist and philosopher, Amartya Sen, who has a book called Development as Freedom. There’s this larger debate where people talk about whether or not democracy or authoritarianism is going to lead to economic growth. And what Sen points out is that the ends of economic growth are freedom, whether it’s freedom from hunger or thirst or whatever it is, but those are just basic political freedoms.

People are wanting to have access to these basic goods—whether you’re disrespected inside, or you’re not going to be murdered on the basis of immutable characteristics or treated differently by your government—and so not treating democracy as, Oh. Well, authoritarian government might get you closer to having a higher GDP. But people also care about basic freedoms. And so I always think that that’s an interesting—

Kim: Yeah. That’s right. And also, Sen has this view of—it’s called the capabilities approach. What matters, basically, is, you know—it’s not necessarily just wealth or income; it’s what that wealth or income enables you to do and what sort of agency it gives you.

And I think that’s also a framing with which to view the land-reform stuff, which is that even if it didn’t necessarily have the claimed productivity impacts—so both having this equity boost but also this efficiency boost, so having your cake and eating it too—it’s enough just to have the equity boost.

Demsas: Yes. Exactly.

Kim: You made society more equitable. You gave a lot of very, very poor peasant farmers a little bit more security. You gave them control over their land. This is something that, politically, they’d been advocating for and fighting for for a very long time.

I think a lot of people on Twitter and elsewhere have been trying to read the paper as, Land reform is bad, or, Joe Studwell is wrong. I think that’s maybe too much of a simplification. Giving the poor more land is, I think, a pretty good thing in itself, and you didn’t actually lower productivity as a result. So I think the view is maybe a little bit more ambivalent than the triumphalist sort of narrative. But I think land reform should still be on the table.

Demsas: Well, we have one final question for you, which is: What is an idea that you initially thought was a good one but ended up only being good on paper?

Kim: Like a lot of people—maybe who are listening to this podcast—I got pretty involved in political-betting markets around 2016 or so. And I was peak econ brain at this point in time. One of the first things you learn if you’ve taken a little bit of intermediate micro[economics] theory is that conditional on the probabilities, you want to equalize your marginal utility across different states of the world. And that’s just a very nerdy, complicated way of saying that you want to hedge your bets, right?

And so when it comes to political-betting markets, I imagine like a lot of people listening to this podcast, I have very strong preferences about the outcomes. And so in the 2016 election, I put it as a small bet—like $25 or whatever—and Donald Trump as, like, as an emotional hedge, basically.

Demsas: So that if Hillary Clinton loses, at least you have some money.

Kim: Yeah. Exactly. You feel a little bit better about yourself. And in Homo economicus logic, this makes a whole lot of sense on paper. And I fully expected Hillary Clinton to win, just like a lot of people did. And of course, she loses. And I get my little, dinky payout, and I don’t really feel a lot better about myself. In fact, I feel a little bit dirty in having—

Demsas: How much money did you win?

Kim: I don’t remember. It was on the order of $30 or $40. It was not a whole lot of money. But yeah, it just didn’t feel good relative to how horrible those months were. So I think, yeah, there’s a sense in which the Homo economicus model is missing something. So that’s something that’s maybe better on paper than it actually is in reality.

Demsas: Well, we would need a randomized controlled trial here to see if you would have felt even worse if you hadn’t had the $30. So it seems unclear.

Kim: Yeah. Maybe. Maybe.

Demsas: Thank you so much for joining the show, Oliver.

Kim: Yeah. Thanks. This was a lot of fun.

[Music]

Demsas: Good on Paper is produced by Jinae West. It was edited by Dave Shaw, fact-checked by Ena Alvarado, and engineered by Erica Huang. Our theme music is composed by Rob Smierciak. Claudine Ebeid is the executive producer of Atlantic audio, and Andrea Valdez is our managing editor.

And hey, if you like what you’re hearing, please leave us a rating and review on Apple Podcasts.

I’m Jerusalem Demsas, and we’ll see you next week.

Last week, Mark Zuckerberg stood on a stage in California holding what appeared to be a pair of thick black eyeglasses. His baggy T-shirt displayed Latin text that seemed to compare him to Julius Caesar—aut Zuck aut nihil—and he offered a bold declaration: These are Orion, “the most advanced glasses the world has ever seen.”

Those glasses, just a prototype for now, allow users to take video calls, watch movies, and play games in so-called augmented reality, where digital imagery is overlaid on the real world. Demo videos at Meta Connect, the company’s annual conference, showed people playing Pong on the glasses, their hands functioning as paddles, as well as using the glasses to project a TV screen onto an otherwise blank wall. “A lot of people have said that this is the craziest technology they’ve ever seen,” Zuckerberg said. And although you will not be able to buy the glasses anytime soon, Meta is hawking much simpler products in the meantime: a new Quest headset and a new round of software updates to the company’s smart Ray-Bans, which have cameras and an AI audio assistant on board, but no screen in the lenses.

Orion seems like an attempt to fuse those two devices, bringing a fully immersive computerized experience into a technology that people might actually be comfortable putting on their face. And it is not, you may have noticed, the only smart-glasses product to have emerged in recent months. Amazon, Google, Apple, and Snap are all either officially working on some version of the technology or rumored to be doing so. Their implementations are each slightly different, but they point to a single idea: that the future is about integrating computing more seamlessly into everyday life.

Smartphones are no longer exciting, and the market for them has been declining for the past few years. The primary new idea there is foldable screens, which effectively allow your phone to turn into a tablet—though tablet sales have slowed too. The virtual-reality headsets that companies have spent billions developing aren’t being widely adopted.

These companies are betting big that people want to be able to check the weather without pulling out a smartphone—and that they are more willing to wear a pair of Ray-Bans with cameras than spend hours in the metaverse. And after years of false starts on the glasses front, they’re betting that AI—despite some high-profile flops—will be what finally helps them achieve this vision.

Tech companies have been working on smart frames for decades. The first real consumer smart glasses started appearing in the late 1980s and ’90s, but none broke through. At last, in 2013, Google released its infamous Glass eyewear. A thin metal frame with a camera and tiny screen above one eye, Glass could be used to check emails, take photos, and get directions. They were advanced for their time, but the general public was spooked by the idea of face-cameras constantly surveilling them. In 2015, Google abandoned the idea that Glass might ever be a consumer product, though the frames lived on as an enterprise device until last year.

Glass’s failure didn’t deter other companies from taking a swing. In 2016, Snapchat launched its first generation of Spectacles, glasses that allowed users to capture pictures and videos from cameras mounted above each eye, then post them on their account. In 2019, Amazon jumped in, teasing its Echo Frames—camera-less smart glasses with Alexa built in—which went on sale to the public the following year. Meta, then called Facebook, launched the first iteration of its collaboration with Ray-Ban in 2021, though the frames didn’t catch on.

Then there are the virtual-reality headsets, such as Meta’s Quest line. Last summer, after Apple announced the Vision Pro, my colleague Ian Bogost deemed this the “age of goggles,” pointing out that companies have been spending billions developing immersive technology, even though the exact purpose of these expensive headsets is unclear.

Consumers also seem to be wondering what that purpose is. One analyst reports that sales of the Vision Pro were so dismal that Apple scaled back production. According to The Information, the company paused work on the next model, while Meta canceled its competitor device entirely.

[Read: The age of goggles has arrived]

In some ways, this glasses moment is something of a retreat: an acknowledgment that people may be less likely to go all in on virtual reality than they are to throw on a pair of sunglasses that happens to be able to record video. These devices are supposed to look and feel more natural, while allowing for ambient-computing features, such as the ability to play music anywhere just by speaking or start a phone call without having to put in headphones.

AI is a big part of this pitch. New advances in large language models are making modern chatbots seem smarter and more conversational, and this technology is already finding its way into the glasses. Both the Meta and Amazon frames have audio assistants built in that can answer questions (How do whales breathe?) and cue up music (play “Teenage Dirtbag”). Meta’s Ray-Bans can “look” using their cameras, offering an audio description of whatever is in their field of vision. (In my experience, accuracy can be hit or miss: When I asked the audio assistant to find a book of poetry on my bookshelf, it said there wasn’t one, overlooking an anthology with the word poetry in the title, though it did identify my copy of Joseph Rodota’s The Watergate when I asked it to find a book about the Washington landmark.). At Connect, Zuckerberg said that the company plans to keep improving the AI, with a couple of big releases coming in the next few months. These updates will give the glasses the ability to do translation in real time, as well as scan QR codes and phone numbers on flyers in front of you. The AI will also, he said, be able to “remember” such things as where you parked your car. One demo showed a woman ruffling through a closet and asking the AI assistant to help her choose an outfit for a theme party.

[Read: The end of foreign-language education]

But whether AI assistants will actually be smart enough to realize all of this is still somewhat of an open question. In general, generative AI struggles to cite its sources and frequently gets things wrong, which may limit smart glasses’ overall usefulness. And, though the companies say the technology will only get better and better, that’s not entirely certain: The Wall Street Journal recently reported that, when Amazon attempted to infuse Alexa with new large language models, the assistant actually became less reliable for certain tasks.

Products such as Orion, which promise not just AI features but a full, seamless integration of the digital world into physical reality, face even steeper challenges. It’s really, really difficult to squish so many capabilities into eyewear that looks semi-normal. You need to be able to fit a battery, a camera, speakers, and processing chips all into a single device. Right now, even some of the most state-of-the-art glasses require you to be tethered to additional hardware to use them. According to The Verge’s Alex Heath, the Orion glasses require a wireless “compute puck” that can be no more than about 12 feet away from them—something Zuckerberg certainly did not mention onstage. Snap’s newest Spectacles, announced earlier this month, don’t require any extra hardware—but they have a battery life of only 45 minutes, and definitely still look big and clunky. The hardware problem has bedeviled generations of smart glasses, and there still isn’t a neat fix.

But perhaps the biggest challenge facing this generation of smart glasses is neither hardware nor software. It’s philosophical. People are stressed right now about how thoroughly technology has seeped into our everyday interactions. They feel addicted to their phones. These companies are pitching smart glasses as a salve—proposing that they could, for example, allow you to handle a text message without interrupting quality time with your toddler. “Instead of having to pull out your phone, there will just be a little hologram,” Zuckerberg said of Orion during his presentation. “And with a few subtle gestures, you can reply without getting pulled away from the moment.”

Yet committing to a world in which devices are worn on our face means committing to a world in which we might always be at least a little distracted. We could use them to quietly read our emails or scroll Instagram at a restaurant without our partner knowing. We could check our messages during a meeting while looking like we’re still paying attention. We may not need to check our phones so much, because our phones will effectively be connected to our eyeballs. Smart glasses walk a thin line between helping us be less obsessively on the internet and tethering us even more closely to it.

I spent some time this spring talking with a number of people who worked on early smart glasses. One of them was Babak Parviz, a partner at Madrona, a venture-capital firm, who previously led Google’s Glass project. We discussed the history of computers: They used to be bulky things that lived in research settings—then we got laptops, then smartphones. With Glass, the team aimed to shorten the time needed to retrieve information to seconds. “The question is, how much further do you need to take that? Do you really need to be immersed in information all the time, and have access to much faster information?” Parvis told me he’d changed his mind about what he called “information snacking,” or getting fed small bits of information throughout the day. “I think constant interruption of our regular flow by reaching out to information sources doesn’t feel very healthy to me.”

In my conversations, I asked experts whether they thought smart glasses were inevitable—and what it would take to unseat the smartphone. Some saw glasses not as a smartphone replacement at all, but as a potential addition. In general, they thought that new hardware would have to give us the ability to do something we can’t do today. Right now, companies are hoping that AI will be the thing to unlock this potential. But as with so much of the broader conversation around that technology, it’s unclear how much of this hype will actually pan out.

These devices still feel more like sketches of what could be, rather than fully realized products. The Ray-Bans and other such products can be fun and occasionally useful, but they still stumble. And although we might be closer than ever to mainstream AR glasses, they still seem a long way off.

Maybe Zuckerberg is right that Orion is the world’s most advanced pair of glasses. The question is really whether his big vision for the future is what the rest of us actually want. Glasses could be awesome. They could also be just another distraction.