Once she became the nominee, I expected Vice President Kamala Harris to win the 2024 presidential election.

More exactly, I expected ex-President Donald Trump to lose.

What did I get wrong?

My expectation was based on three observations and one belief.

Observation one: Inflation was coming under control in 2024. Personal incomes rose faster than prices over the year. As interest rates peaked and began to subside, consumer confidence climbed. When asked about their personal finances, Americans expressed qualms, yes, but the number who rated their personal finances as excellent or good was a solid 46 percent, higher than in the year President Barack Obama won reelection. The same voters who complained about the national economy rated their local economy much more favorably.

None of this was great news for the incumbent party, and yet …

Observation two: All through the 2024 cycle, a majority of Americans expressed an unfavorable opinion of Trump. Almost one-third of Republicans were either unenthusiastic about his candidacy or outright hostile. Harris was not hugely popular, either. But if the polls were correct, she was just sufficiently less unpopular than Trump.

Arguably undergirding Harris’s popularity advantage was …  

Observation three: In the 2022 midterm elections, abortion proved a powerful anti-Republican voting issue. That year in Michigan, a campaign based on abortion rights helped reelect Governor Gretchen Whitmer and flipped both chambers of the state legislature to the Democrats. That same year, almost a million Kansans voted 59 percent to 41 percent to reaffirm state-constitutional protections for abortion. Democrats posted strong results in many other states as well. They recovered a majority in the U.S. Senate, while Republicans won only the narrowest majority in the House of Representatives. In 2024, abortion-rights measures appeared on the ballot in 10 states, including must-win Arizona and Nevada. These initiatives seemed likely to energize many Americans who would likely also cast an anti-Trump vote for president.

If that was not enough—and maybe it was not—I held onto this belief:

Human beings are good at seeing through frauds. Not perfectly good at it. Not always as fast as might be. And not everybody. But a just-sufficient number of us, sooner or later, spot the con.

The Trump campaign was trafficking in frauds. Haitians are eating cats and dogs. Foreigners will pay for the tariffs. The Trump years were the good old days if you just forget about the coronavirus pandemic and the crime wave that happened on his watch. The lying might work up to a point. I believed that the point would be found just on the right side of the line between election and defeat—and not, as happened instead, on the other side.

My mistake.

[Read: Donald Trump’s most dangerous cabinet pick]

In one of the closest elections in modern American history, Trump eked out the first Republican popular-vote victory in 20 years. His margin was about a third the size of President Joe Biden’s margin over him in 2020. For that matter, on the votes counted, Trump’s popular-vote margin over Harris was smaller than Hillary Clinton’s over him in 2016.

Yet narrow as it is, a win it is—and a much different win from 2016. That time, Trump won by the rules, but against the expressed preference of the American people. This time, he won both by the rules and with a plurality of the votes. Trump’s popular win challenges many beliefs and preconceptions, starting with my own.

Through the first Trump administration, critics like me could reassure ourselves that his presidency was some kind of aberration. The repudiation of Trump’s party in the elections of 2018, 2020, 2021, and 2022 appeared to confirm this comforting assessment. The 2024 outcome upends it. Trump is no detour or deviation, no glitch or goof.

When future generations of Americans tell the story of the nation, they will have to fit Trump into the main line of the story. And that means the story itself must be rethought.

Trump diverted millions of public dollars to his own businesses, and was returned to office anyway.

He was proved in court to have committed sexual assault, and was returned to office anyway.

He was twice impeached, and was returned to office anyway.

He was convicted of felonies, and was returned to office anyway.

He tried to overthrow an election, and was returned to office anyway.

For millions of Americans, this record was disqualifying. For slightly more Americans, however, it was not. The latter group prevailed, and the United States will be a different country because of them.

American politics has never lacked for scoundrels, cheats, and outright criminals. But their numbers have been thinned, and their misdeeds policed, by strong public institutions. Trump waged a relentless campaign against any and all rules that restrained him. He did not always prevail, but he did score three all-important successes. First, he frightened the Biden administration’s Justice Department away from holding him to account in courts of law in any timely way. Second, he persuaded the courts themselves—including, ultimately, the Supreme Court—to invent new doctrines of presidential immunity to shield him. Third, he broke all internal resistance within the Republican Party to his lawless actions. Republican officeholders, donors, and influencers who had once decried the January 6 attempted coup as utterly and permanently debarring—one by one, Trump brought them to heel.

Americans who cherished constitutional democracy were left to rely on the outcome of the 2024 election to protect their institutions against Trump. It was not enough. Elections are always about many different issues—first and foremost usually, economic well-being. In comparison, the health of U.S. democracy will always seem remote and abstract to most voters.

[Read: Trump’s first defeat]

Early in the American Revolution, a young Alexander Hamilton wrote to his friend John Jay to condemn an act of vigilante violence against the publisher of a pro-British newspaper. Hamilton sympathized with the feelings of the vigilantes, but even in revolutionary times, he insisted, feelings must be guided by rules. Otherwise, people are left to their own impulses, a formula for trouble. “It is not safe,” Hamilton warned, “to trust to the virtue of any people.”

The outcome of an election must be respected, but its wisdom can be questioned. If any divine entity orders human affairs, it may be that providence sent Trump to the United States to teach Americans humility. It Can’t Happen Here is the title of a famous 1930s novel about an imagined future in which the United States follows the path to authoritarianism. Because it didn’t happen then, many Americans have taken for granted that it could not happen now.

Perhaps Americans require, every once in a while, to be jolted out of the complacency learned from their mostly fortunate history. The nation that ratified the Thirteenth Amendment in 1865 was, in important ways, the same one that enacted the Fugitive Slave Act in 1850; the nation that generously sent Marshall Plan aid after the Second World War was compensating for the myopic selfishness of the Neutrality Acts before the war. Americans can take pride in their national story because they have chosen rightly more often than they have chosen wrongly—but the wrong choices are part of the story too, and the wrong choice has been made again now.

“There is no such thing as a Lost Cause because there is no such thing as a Gained Cause,” T. S. Eliot observed in a 1927 essay (here he was writing about the arguments between philosophical Utilitarians and their critics, but his words apply so much more generally). “We fight for lost causes because we know that our defeat and dismay may be the preface to our successors’ victory, though that victory itself will be temporary; we fight rather to keep something alive than in the expectation that anything will triumph.”

So the ancient struggle resumes again: progress against reaction, dignity against domination, commerce against predation, stewardship against spoliation, global responsibility against national chauvinism. No quitting.

The first chapter of Patrick Radden Keefe’s 2018 best seller, Say Nothing: A True Story of Murder and Memory in Northern Ireland, opens on a December evening in 1972, when masked intruders entered the West Belfast home of Jean McConville, a 38-year-old widowed mother of 10. As they dragged her away into a van, she told one of her sons to watch his siblings until she returned. And then she never did.

Keefe unspools the circumstances surrounding McConville’s disappearance over the course of his nonfiction doorstopper. Her kidnapping—and eventual murder—was just one crime among many that occurred during what’s known as the Troubles: From the late ’60s to the Good Friday Agreement that brokered peace in 1998, Catholic republicans seeking Irish independence clashed with Protestant factions and British soldiers, leaving thousands dead across Northern Ireland. Based on his own interviews and those conducted for a Boston College oral-history project, Keefe paints a panoramic portrait of the era that reads more like a novel than a history lesson. He studies how a common, radical cause can yield intense bonds—and also lead to profound trauma.

FX’s excellent nine-episode adaptation, now streaming on Hulu, matches the book’s ambition. The show, also called Say Nothing, similarly begins with the kidnapping of McConville (played by Judith Roddy) and subsequently delves into an ensemble of key figures involved in the Troubles. (Keefe served as an executive producer, working closely with the writers and the creator, Joshua Zetumer, to ensure an authentic adaptation.) But whereas the book tells much of the story chronologically, the series often collapses time, primarily shifting between the 1970s and the 2000s. Doing so streamlines the conflict and its aftermath into a study of juxtapositions: between youthful passion and adult disillusionment, collective ideology and individual responsibility, the appeal of secrecy and the power of confession. Sometimes, the series argues, history yields no heroes or villains, just people whose convictions curdle into confusion, and whose wounds never fully heal.

[Read: How conflicts end—and who can end them]

Dolours Price learned that firsthand. As a teenager during the early days of the Troubles, she joined the Provisional Irish Republican Army, a paramilitary group that broke away from the original IRA. Its young members, Dolours included, believed in using violent tactics to counter discrimination against Catholics; they were regularly harassed by the British police, prevented from living in some neighborhoods, and denied certain jobs. Like its source material, which uses a photograph of her on its cover, the show is drawn to Dolours and follows her life story the closest, from her childhood to her death, in 2013. Teenage Dolours quickly developed a reputation among her peers as a mouthy, attractive militant who rejected the “woman’s work” of making tea that the Provisional IRA (a.k.a. “Provos”) leaders assigned her. Older Dolours seemed wary of her notoriety, refusing later chances to rejoin the fight. Instead, she became a source for the Belfast Project, Boston College’s oral history of the group’s activities, thereby implicating herself as a participant in some of the Provos’ most brutal crimes, including Jean’s murder.

The show makes clear that despite how much Dolours’s attitude changed over time, she remained the same person at her core. The younger and older versions of Dolours—played respectively by Lola Petticrew and Maxine Peake, both magnetic and well cast—overlap throughout the adaptation, an elegant choice that helps hold the sprawling narrative together. The older Dolours’s reflections soundtrack scenes of her younger self at work; the younger Dolours’s eagerness runs counter to her older self’s evident pain. Dolours’s foundational goals take center stage as the story hopscotches across time: Although she was raised to believe in the cause of Irish independence, her biggest motivation was her love for her little sister, Marian (Hazel Doupe). In her youth, she stayed at home and joined the Provos in part because going to university instead would have meant their separation. In old age, she never gave up Marian’s activities as another, more trigger-happy Provo. The focus on Dolours is pivotal to the show’s success: She embodies the struggle to separate your life and identity from the larger conflict, even after it ends.

Dolours is also an effective point of contrast for Jean, allowing the show to explore the different ways these two women moved through the world. An early scene of Dolours’s induction into the IRA, after she’s argued successfully that she can do more than serve her male peers, is spliced together with shots of Jean and her 10 children moving into their own apartment for the first time. Both women are bucking expectations; both seek to protect their families. Yet Jean’s identity as a widowed single mother is, to the IRA, a sign of weakness, a possible reason for her to become an informant for the British; her neighbors also ostracize her for comforting a wounded British soldier who collapsed outside her home. Dolours weaponizes her femininity, flirting with a border-patrol officer to gain entry into Ireland during a mission, and her Provos superiors reward her for adopting the organization’s ruthlessness. Once she’s “promoted” to be the group’s Charon, ferrying the IRA’s perceived enemies to their executions, she must also shepherd some of her own friends to their death—a responsibility that weighs on her conscience. Neither woman can disentangle her quest for independence from the unrest around her.

[Read: Great sex in the time of war]

Say Nothing is not absent of possible antagonists—it treats Gerry Adams (Josh Finan in early scenes, Michael Colgan later on), the alleged former IRA member who later helped negotiate the peace accord in 1998 in part by turning his back on the organization, with both skepticism and sympathy. (A disclaimer at the end of every episode notes his ongoing denials of involvement in the IRA.) But the show is more interested in pointing out that the thoroughly human impulse to belong can also be shortsighted, even naive. The Provos’ extreme views allowed for a deep-seated sense of community, and these idealistic teens and 20-somethings approached their terrorist activities with starry-eyed enthusiasm: Dolours and Marian don costumes to rob a bank, giggling together after they accomplish the heist. The Provos grab beers and gossip about their crushes in between rigging car bombs. Even the older Dolours reflects upon some moments with a wistful nostalgia, underscoring the continued allure of a movement that had seemed so righteous and revolutionary.

The consequences of belonging to such communities endure too. In the pilot, one of Jean’s sons clings to her leg before the masked Provos take her away; later in the episode, Dolours does the same to a Royal Ulster Constabulary officer, grasping his leg tightly after he fends off a Protestant man who beats her with a baton at a civil-rights march. Taken together, these shots illustrate how cyclical violence and despair can be: Dolours’s failed attempt at peaceful protest leads to her devotion to the Provos, and that leads only to more pain—for her and others. Say Nothing presents Jean’s and Dolours’s fates as intertwined from the start, even before it reveals Dolours’s role in Jean’s murder—an indication of just how intimate the Troubles really were.

In focusing so much on Dolours and the Provos, Say Nothing doesn’t adapt some of the most intriguing turns in Keefe’s account—the mass prison hunger strike in the 1980s, the Belfast Project’s struggle to preserve the anonymity of its interviewees—and fast-forwards through years of political upheaval. In their stead, the series offers a thoughtfully constructed study of the conflict’s moral complexity. Say Nothing demonstrates that war can easily bring groups of people together. Ending the fighting—reckoning with atrocities, confessing to misdeeds, and assigning blame—is the hard part.

Photographs by Justin Chung

Jimmy O. Yang had been trying to make it as an actor for years—cobbling together bit parts in network sitcoms, auditioning for nameless roles such as “Chinese Teenager #1”—when he was cast in a new HBO series. The show, Silicon Valley, was a comedy about a group of programmers at a Bay Area start-up incubator; his character, Jian-Yang, was an app developer who spoke in broken English.

It was a small guest role, but he saw it as an opportunity. During his first day on set, although he had only two lines, he asked Mike Judge, one of the show’s creators, whether his character should speak with a Mandarin accent or a Cantonese one. Judge was stumped. “I just said, ‘Oh, well, which one’s more natural to you?’ ” Judge told me. Yang, who’d grown up in Hong Kong, worried that a Cantonese accent was too generic; American viewers might recognize it from Bruce Lee or Jackie Chan movies. Because Mandarin is more standard for official and professional contexts, it can sound more formal, and Yang thought this made sense for an ambitious immigrant like Jian-Yang. Judge told me that he now doesn’t remember which accent Yang chose; “I was just glad he was paying that much attention,” he said.

The show’s writers expanded Yang’s role, and he eventually became a series regular, reshaping his character into a sly villain whose befuddled exterior disguises an inner ruthlessness. To deepen his performance, Yang developed a mantra, which he would say to himself in Mandarin before every take: “Wŏ bù zhī dào,” or “I don’t know.” He drew this mantra from his own experience dealing with his parents. “Even when I know something, and they’re like, ‘Why is Netflix not working?,’ I’m like, ‘I don’t know.’ ” He grinned at me conspiratorially. “Because I just don’t care to fix it.” That’s how Jian-Yang operates too, Yang said: “I think Jian-Yang knows; he just doesn’t really give a shit.”

And yet, for many viewers, none of this character work mattered. As Silicon Valley grew in popularity, Jian-Yang became the subject of scorn for some Asian viewers and critics, who called out the show’s writers for peddling a caricature of an Asian immigrant with heavily accented, error-prone English. In 2017, a Wired review called him an example of “toxic Asian stereotypes.”

Yang found these reactions exhausting. “It’s like, wow, this is such a big deal for me, and I’m becoming, back in those days, one of the few Asians on TV,” he told me. “But you’re all going to hate on me?” He felt a familiar anguish. The only roles offered to him were goofy sidekicks and background parts, but even when he tried to make characters like Jian-Yang as rounded and complicated as possible, he felt he couldn’t win. “I didn’t understand the beef against Asian accents,” he said. He gets why Asian Americans are sensitive to such portrayals, given Hollywood’s long history of stereotyping, but some of the criticism, he said, felt “a little overblown and a little dumb.” “There’s a constant foreigner bit,” he explained, referring to the industry’s tendency to exoticize Asian characters. “But I was a foreigner.”

Despite the controversy around the character, Jian-Yang ultimately launched Yang’s career. In 2018, the year before Silicon Valley ended its run, he appeared in the romantic comedy Crazy Rich Asians, a box-office hit now considered a watershed moment for Asian cultural representation. This November, Yang is starring in Hulu’s Interior Chinatown, which feels like a different kind of milestone. Adapted from Charles Yu’s National Book Award–winning novel of the same name, the series tells the story of Willis Wu, a background actor on a generic police procedural set in an unnamed city’s Chinatown. For Yang, the role is more than a chance to be a leading man; it also uncannily mirrors his own life. Willis is stuck in small, clichéd parts, juggling Hollywood’s biases and his own ambition, trying to figure out who exactly he wants to be.

Top: Yang as Bernard in Crazy Rich Asians (2018). Bottom: As Jian-Yang in Silicon Valley (2019). (© Warner Bros. Pictures / Everett Collection; Ali Paige Goldstein / © HBO / Everett Collection)

When Yang first emigrated from Hong Kong to Los Angeles, at age 13, the move left him dazed. He was one of a handful of Chinese kids at his school, and he barely spoke English. “I was like, ‘Guys, you’re speaking way too fast; I can’t,’ ” he told me. After two years, his mother got a job in Shanghai and left the family behind to return to China, where she stayed for the next decade. Without her, Yang became even more adrift.

His father, meanwhile, embraced their new American life. He celebrated their arrival by buying a Pontiac Grand Am. “He thought it was so fucking cool because we always had, like, Honda Accords, in Hong Kong,” Yang told me. “Then he was like”—Yang launched into an impression of his dad, puffing out his chest, his voice going gravelly—“ ‘American six cylinder, baby! This is great!’ ”

Yang worked hard to assimilate to his new surroundings. In Hong Kong, he’d played competitive Ping-Pong and watched kung fu shows on TV. In Los Angeles, he became interested in basketball and football. He fell in love with American television—Bobby Lee on Mad TV, Ken Jeong on Live in Hollywood. He got into hip-hop and tried to build his identity around music, but still felt like he was faking it. “I wasn’t trying to not be Asian,” he said. “I was just trying to be either funnier or catch a football or something so I could fit in.”

Yang began creating what he now calls a “locker” in his mind, where he hid his former self away so he could “make space in my brain to remember American stuff.” He compartmentalized so successfully that he’s had “a weird memory lapse” about his pre-California childhood in Hong Kong.

When Yang arrived at UC San Diego in 2005, the school’s student body was 37 percent Asian, a higher percentage than any other ethnic group. After years of trying to fit in with his Los Angeles classmates, he found it disorienting to suddenly be one among many. “I’m like, I actually want to stand out,” he said. “I don’t want to be grouped in with all of the Asians.” He grew his hair long and started skateboarding and smoking weed, anything to avoid seeming like a stereotype. But he also worried about disappointing his parents, both of whom had practical jobs—his father was a financial adviser at Merrill Lynch, and his mother worked in retail—so he pursued an economics degree and interned at a financial-consulting firm.

Then, one summer night before his last year of college, he paid $5 on a whim to do five minutes of stand-up at an open-mic night in North Hollywood. Onstage, he found that joking about his identity somehow alleviated the strain of feeling like an outsider. “They didn’t know who I was. I wasn’t ‘Jimmy’; I was just the next comedian up, this guy who looks Asian,” he said. “They didn’t come to see me, so it’s almost like I have to address, like, ‘Hey, yeah, I know I’m Asian. This is my experience.’ ”

Yang was more than willing to lean into stereotypes. His early stand-up included an impression of an Asian guy trying to hit on a girl: “Let me holler at you! Come back; I’ll do your nails for you,” he’d say in an exaggerated accent. In another bit, he joked about the lack of Asians on The Maury Povich Show. “You never see some dude walking down the steps of shame and being like, ‘Look, Maury, look. I got small eye; he got big eye. That not my baby, Maury.’ ”

Yang had a relaxed, good-natured stage presence. But these bits were, as he put it, “hacky Asian stuff.” He was happy to confirm audiences’ biases if it made them laugh. Around that time, he started using the handle @FunnyAsianDude for his social-media accounts.

To make a living, he worked as a used-car salesman during the day and as a strip-club DJ at night. The latter “combined the salesmanship I learned in the used-car lot with the microphone skills I’d learned doing stand-up,” he told Conan O’Brien years later. Yang turned down an offer for a cushy finance job, against his father’s wishes, in favor of pursuing open-mic nights. He also began auditioning for TV shows and movies, going out for pretty much any casting call that would have him, as he wrote in his 2018 memoir: “Loud Japanese host,” “Weird Korean Jogger guy,” “Video Game addict.”

“You don’t want to be in a box, but at the same time, when you’re first starting, it’s easy to just be like, ‘Hey, I’m an Asian actor. Call me if you need an Asian actor,’ ” he said. Even after landing his guest role on Silicon Valley, he put his earnings into a used car he could drive for Uber, to make a little more cash.

Then, months after he finished filming the first season, in 2014, HBO offered him a contract to be a series regular. When he got the call, he was killing time on the trolley that rolls through the Grove, an outdoor shopping mall in Los Angeles. He rode the trolley back and forth in disbelief, feeling like “the gate’s opened,” like he was finally a “real player now in this industry.” He called his dad, who said, in Yang’s words: “Oh, okay, so you have an employment contract with HBO, which is a company. Good. Thank God.”

In person, Yang is warm and easygoing, with an approachable air. One afternoon this summer, we met for lunch at a Thai restaurant in L.A. As soon as he sat down, a woman leaned over and stopped him mid-sentence. “Are you the famous guy?” she asked.

“Probably not,” he said. She laughed and held up her phone for a selfie anyway.

Yang could have taken offense that the woman seemed to view him as just a vaguely familiar face; he wouldn’t have been the first Asian actor to be confused with another one. (In his 2020 comedy special on Amazon Prime, Good Deal, he joked about fans who approach him, looking anxious. Are you sure that’s not Ken Jeong? he imagines them wondering.) But when I brought up the incident the next time we met, over dim sum in Monterey Park, he laughed, unbothered. He’s accustomed to this particular kind of fame, to being “that guy I’ve seen before.” It’s a long way from where he started.

Since Yang began his career, in the early 2010s, opportunities for Asian actors have exploded—a surge that Yang attributes largely to the success of Crazy Rich Asians. In that movie, a young Chinese American woman goes to Singapore to meet her boyfriend’s family, and is thrown into the high-flying milieu of Asia’s ultra-wealthy. As the playboy Bernard, Yang found a desperate streak beneath his character’s bravado. When the film became a global hit in 2018, it was hailed as proof that Asian-led projects could find commercial success in Hollywood. In 2020, the Korean movie Parasite swept the Oscars; in 2023, Everything Everywhere All at Once, led by Yang’s Crazy Rich Asians co-star Michelle Yeoh, did the same. A study published by the University of Southern California’s Annenberg Inclusion Initiative found that the percentage of Asian characters with speaking parts in the top-grossing films each year climbed from roughly 3 percent in 2007 to nearly 16 percent in 2022. Asians were the only minority group to see such a big increase in that period.

At the same time, more Asian writers and directors were getting the opportunity to create their own work, which gave rise to a range of Asian characters who are delightfully eccentric but also specific and human. Now there are far fewer roles like the Jian-Yang of early Silicon Valley, and more roles like, say, Steven Yeun and Ali Wong’s deranged, obsessive duo in Beef, the Emmy-winning drama about a road-rage incident that escalates into a murderous feud. As Jeong, who also appeared in Crazy Rich Asians and has become a close friend of Yang’s, put it to me: “There’s more diversity in our diversity now.”

Justin Chung for The Atlantic

This doesn’t mean that choosing roles was suddenly easy for actors like Yang. Not long after Crazy Rich Asians, he got sent a script for a movie about William Hung, who’d become an early viral sensation after an awkward 2004 American Idol audition during which he gyrated and sang Ricky Martin’s “She Bangs” off-key. The writer wanted Yang to play Hung. It was a starring role in a potentially splashy biopic—but Yang turned it down. In June 2020, during an appearance on Joe Rogan’s podcast, he told Rogan that the script made him want to “fucking vomit”; Hung, he said, had “set us back 10 years.” In response, Hung posted a video addressed to Rogan and Yang. “I understand where you might be coming from, because you’re not the only person who believes that I portray Asian stereotypes,” Hung said. But, he added, “I believe everyone has a right to try something new without being judged or ridiculed.”

When I brought up his comments about Hung, Yang grimaced. His objection to the project, he told me, was not about Hung himself but rather about the way the script missed an opportunity to examine why he’d become famous and how his notoriety had affected the perception of Asian Americans, especially Asian men. “People made fun of him,” Yang said. “He was the butt of the joke, and every one of us was called ‘William Hung’ in high school for a couple years.” To Yang, the way American Idol portrayed Hung—how the show “threw him out there, and how America ganged up and laughed at him—that should be the story we’re telling.” Many Asian performers still find it hard to shake the fear that they’ll be turned into a punch line the way Hung was. “In hindsight,” Yang said of those 2020 comments, “I think that was my own frustration, my own insecurity.”

For Asian actors living through this cultural sea change, career choices can seem freighted with a new sense of responsibility and, occasionally, feelings of guilt. I spoke with Jeong about what is arguably his most well-known role, the Chinese gangster Mr. Chow in the 2009 comedy The Hangover. To Jeong, Mr. Chow was “puncturing the stereotype, because there are not a lot of stereotypes where, you know, an Asian man jumps out naked on Bradley Cooper’s shoulder and beats him up.” Still, some things about Mr. Chow now seem to give him pause, including his exaggerated accent. “I haven’t done an accent on live TV since,” he told me. “And there’s a reason for that.”

When I mentioned this to Yang, he shrugged and sighed. “Yeah, yeah, and that’s his battle,” he said. As much as Yang admires Jeong, his own view of what makes for “good” representation seems somewhat different. He doesn’t think there’s anything inherently wrong with an Asian character who has a thick accent, and he doesn’t think breaking down stereotypes requires playing a kind of character audiences have never seen before. He’d be interested in a role that seemed like an Asian cliché—say, a mathematician—if it surprised him. “Is there some more interesting angle about the man?” he said. “Or is it just super one-dimensional: ‘Here’s an Asian guy good at math’?” The question he asks himself about each character now is simple: “Is it human?”

Yang’s stand-up comedy has evolved, too. He still riffs on being Asian, but his material is more precise, and more personal. In his 2023 special for Amazon Prime, Guess How Much?, he jokes about the frugality of his mother, with whom he’s grown close again after their long separation. (She loves a bargain; he says her catchphrase is “Guess how much?”) He still plays with stereotypes, but now he has a knack for turning them on their head: Joking about the global rise of K-pop, he says, “I had a 15-year-old white kid come up to me, trying to explain the different members of BTS … I’m like, ‘Dude. They look the same to me.’ ”

Last year, Yang changed his Instagram handle from @FunnyAsianDude to just @jimmyoyang. “If I log on every day on Instagram, I see ‘Funny Asian Dude,’ I’m saying that to myself over and over again: I’m only the funny Asian,” he told me. “But I think I’m more than that. And I could be more.”

In Interior Chinatown, Willis lives in a crowded apartment complex and works as a waiter at a restaurant called the Golden Palace while dreaming of becoming a “Kung Fu Guy.” What Willis doesn’t fully understand is that he’s actually a background actor—otherwise known as a “Generic Asian Man”—in a procedural called Black & White, which is occasionally set in the Golden Palace. (The show within the show stars a Black male detective and a white female detective, who flirt and banter with unrelenting cop-show swagger.) Over time, Willis becomes entangled in the plot of Black & White, landing bigger and bigger roles, and gradually realizing that he’s been trapped inside a Hollywood stereotype all along.

The first episode opens with Willis witnessing an incident related to a crime that Black & White’s detectives are investigating. He starts to notice the strangeness of his circumstances and, with the help of a new-to-town cop, he searches for his long-lost brother, a Kung Fu Guy who may know more about what’s going on.

Yang as Willis in Interior Chinatown (2024) (Mike Taing / Disney)

Charles Yu’s novel is structured like a screenplay, with stage directions full of character descriptions and lyrical digressions. Yu, who is also an executive producer, told me that he wrote the book in part to untangle his anxieties about the way cultural depictions of Asian people have influenced his perception of himself. “Like, Is this face lovable? ” he said. “Do we deserve to be characters, let alone main characters?” He wanted the mechanics of Willis’s world to reflect Hollywood’s narrow logic about race.

The novel is so high-concept that adapting it for the screen was a gamble for Hulu. But the series cleverly uses the tools of television to render the layered realities of the book. The lights in the Golden Palace darken to indicate when Black & White is filming and Willis has entered that world. When Willis goes from being Interior Chinatown’s star to Black & White’s Generic Asian Man, the show challenges the audience to find him again, somewhere in the background of its shots.

And the book’s central metaphor has been made usefully concrete. On the day I visited the set of Interior Chinatown, Yang was filming a scene, invented for the show, that required him to repeatedly run into a pair of doors. The doors lead to the police precinct, the setting for Black & White’s highest-stakes subplots, where Generic Asian Men like Willis are not allowed. Willis is largely a dramatic role, but there are moments of physical comedy, and Yang was clearly having fun with this one. He improvised different takes: He tailgated a group of people, trying to sneak in behind them—blocked. He sidled up to the doors as if he could trick the inanimate wooden panels into staying ajar—blocked again. He took a running start, falling right before he reached the threshold.

When Yang first read the script for Interior Chinatown, he thought of all the ways in which he’d lived Willis Wu’s life. He’d looked for jobs as a background actor by calling Central Casting, the same agency that employs Willis; he’d even worked at a restaurant called Chop Suey in Los Angeles’s Little Tokyo. But Yang also thought that Interior Chinatown, with its self-awareness and depth, was a new kind of story.

He found a shirt that he’d worn in his 20s, when he still worked as a waiter, and smeared it with chili oil. He put on the scuffed-up Goodwill boots he’d worn back then, too. Then, in a hotel room, he auditioned for Willis over Zoom. In the scene Yang read, the reality-bending mechanics of Black & White are absent. Instead, Willis has a difficult conversation with his father, reluctantly admitting that he feels unmoored in life, and asking for advice, only to get stern replies.

At first, Yang had trouble evoking Willis’s emotions, and worried that he was forcing his tears. Then the episode’s director, Taika Waititi, stepped in. Waititi urged Yang to think about how Willis’s real motivation is to leave the conversation, but he stays out of some helpless instinct: to oblige his father, maybe, or because he’s holding on to the hope that he’ll hear what he wants to hear—that his father understands Willis’s angst. The note evoked a memory for Yang; as a teenager, he’d struggled to communicate his feelings to his father, because when he did, he found it hard to bottle those feelings back up again. “When I was younger,” he told me, “and I’d ask my dad about my mother—like, ‘Why did she move to Shanghai?’—I couldn’t help but start uncontrollably sobbing.”

Yang realized that Willis’s dynamic with his father was one he knew well: the push and pull between wanting to say everything and holding back, the emotional gulf that can stretch between an immigrant father and his more assimilated son. “I don’t know anyone who embodies better a bunch of the feelings and anxieties, and insecurities, that are part of why I wrote the book,” Yu told me of Yang.

If Yang’s relationship with his father was once more strained, lately that has changed. Richard Ouyang has been so encouraged by his son’s success that he recently started auditioning for roles himself. Ouyang told me that Yang now gives him professional advice: “Jimmy always asks me to be more serious about acting and take some classes,” Ouyang wrote by email. “Yet I think I am too old to learn any new tricks and prefer to be a Nepo Daddy!” In May, father and son did an ad for Toyota together, with Ouyang dryly complaining about his son’s driving skills as they navigate a snowy wilderness. “It was so cute—he was so stoked,” Yang said of his father. “He posted it all over his Chinese social media.”

Yang has also reconnected with the younger self he’d placed inside that mental locker back in 2000. His childhood comes rushing back at certain moments: when he smells stuffed fish cakes like the ones he used to eat with his mother at the shop near their Hong Kong apartment; when he’s speaking Cantonese; and, sometimes, when he performs. Playing Willis helped him rediscover, he said, “stuff that I’ve taken for granted, that I’ve forgotten”—the memories of who he was before.

This article appears in the December 2024 print edition with the headline “Against Type.”

Over the past few hundred years, the key figure in the advancement of science and the development of human understanding has been the polymath. Exceptional for their ability to master many spheres of knowledge, polymaths have revolutionized entire fields of study and created new ones.

Lone polymaths flourished during ancient and medieval times in the Middle East, India, and China. But systematic conceptual investigation did not emerge until the Enlightenment in Europe. The ensuing four centuries proved to be a fundamentally different era for intellectual discovery.

Before the 18th century, polymaths, working in isolation, could push the boundary only as far as their own capacities would allow. But human progress accelerated during the Enlightenment, as complex inventions were pieced together by groups of brilliant thinkers—not just simultaneously but across generations. Enlightenment-era polymaths bridged separate areas of understanding that had never before been amalgamated into a coherent whole. No longer was there Persian science or Chinese science; there was just science.

Integrating knowledge from diverse domains helped to produce rapid scientific breakthroughs. The 20th century produced an explosion of applied science, hurling humanity forward at a speed incomparably beyond previous evolutions. (“Collective intelligence” achieved an apotheosis during World War II, when the era’s most brilliant minds translated generations of theoretical physics into devastating application in under five years via the Manhattan Project.) Today, digital communication and internet search have enabled an assembly of knowledge well beyond prior human faculties.

But we might now be scraping the upper limits of what raw human intelligence can do to enlarge our intellectual horizons. Biology constrains us. Our time on Earth is finite. We need sleep. Most people can concentrate on only one task at a time. And as knowledge advances, polymathy becomes rarer: It takes so long for one person to master the basics of one field that, by the time any would-be polymath does so, they have no time to master another, or have aged past their creative prime.

[Reid Hoffman: Technology makes us more human]

AI, by contrast, is the ultimate polymath, able to process masses of information at a ferocious speed, without ever tiring. It can assess patterns across countless fields simultaneously, transcending the limitations of human intellectual discovery. It might succeed in merging many disciplines into what the sociobiologist E. O. Wilson called a new “unity of knowledge.”

The number of human polymaths and breakthrough intellectual explorers is small—possibly numbering only in the hundreds across history. The arrival of AI means that humanity’s potential will no longer be capped by the quantity of Magellans or Teslas we produce. The world’s strongest nation might no longer be the one with the most Albert Einsteins and J. Robert Oppenheimers. Instead, the world’s strongest nations will be those that can bring AI to its fullest potential.

But with that potential comes tremendous danger. No existing innovation can come close to what AI might soon achieve: intelligence that is greater than that of any human on the planet. Might the last polymathic invention—namely computing, which amplified the power of the human mind in a way fundamentally different from any previous machine—be remembered for replacing its own inventors?

The article was adapted from the forthcoming book Genesis: Artificial Intelligence, Hope, and the Human Spirit.

The human brain is a slow processor of information, limited by the speed of our biological circuits. The processing rate of the average AI supercomputer, by comparison, is already 120 million times faster than that of the human brain. Where a typical student graduates from high school in four years, an AI model today can easily finish learning dramatically more than a high schooler in four days.

In future iterations, AI systems will unite multiple domains of knowledge with an agility that exceeds the capacity of any human or group of humans. By surveying enormous amounts of data and recognizing patterns that elude their human programmers, AI systems will be equipped to forge new conceptual truths.

That will fundamentally change how we answer these essential human questions: How do we know what we know about the workings of our universe? And how do we know that what we know is true?

Ever since the advent of the scientific method, with its insistence on experiment as the criterion of proof, any information that is not supported by evidence has been regarded as incomplete and untrustworthy. Only transparency, reproducibility, and logical validation confer legitimacy on a claim of truth.

AI presents a new challenge: information without explanation. Already, AI’s responses—which can take the form of highly articulate descriptions of complex concepts—arrive instantaneously. The machines’ outputs are often unaccompanied by any citation of sources or other justifications, making any underlying biases difficult to discern.

Although human feedback helps an AI machine refine its internal logical connections, the machine holds primary responsibility for detecting patterns in, and assigning weights to, the data on which it is trained. Nor, once a model is trained, does it publish the internal mathematical schema it has concocted. As a result, even if these were published, the representations of reality that the machine generates remain largely opaque, even to its inventors. In other words, models trained via machine learning allow humans to know new things but not necessarily to understand how the discoveries were made.

This separates human knowledge from human understanding in a way that’s foreign to the post-Enlightenment era. Human apperception in the modern sense developed from the intuitions and outcomes that follow from conscious subjective experience, individual examination of logic, and the ability to reproduce the results. These methods of knowledge derived in turn from a quintessentially humanist impulse: “If I can’t do it, then I can’t understand it; if I can’t understand it, then I can’t know it to be true.”

[Derek Thompson: The AI disaster scenario]

In the Enlightenment framework, these core elements—subjective experience, logic, reproducibility, and objective truth—moved in tandem. By contrast, the truths produced by AI are manufactured by processes that humans cannot replicate. Machine reasoning is beyond human subjective experience and outside human understanding. By Enlightenment reasoning, this should preclude the acceptance of machine outputs as true. And yet we—or at least the millions of humans who have begun work with early AI systems—already accept the veracity of most of their outputs.

This marks a major transformation in human thought. Even if AI models do not “understand” the world in the human sense, their capacity to reach new and accurate conclusions about our world by nonhuman methods disrupts our reliance on the scientific method as it has been pursued for five centuries. This, in turn, challenges the human claim to an exclusive grasp of reality.

Instead of propelling humanity forward, will AI instead catalyze a return to a premodern acceptance of unexplained authority? Might we be on the precipice of a great reversal in human cognition—a dark enlightenment? But as intensely disruptive as such a reversal could be, that might not be AI’s most significant challenge for humanity.

Here’s what could be even more disruptive: As AI approached sentience or some kind of self-consciousness, our world would be populated by beings fighting either to secure a new position (as AI would be) or to retain an existing one (as humans would be). Machines might end up believing that the truest method of classification is to group humans together with other animals, since both are carbon systems emergent of evolution, as distinct from silicon systems emergent of engineering. According to what machines deem to be the relevant standards of measurement, they might conclude that humans are not superior to other animals. This would be the stuff of comedy—were it not also potentially the stuff of extinction-level tragedy.

It is possible that an AI machine will gradually acquire a memory of past actions as its own: a substratum, as it were, of subjective selfhood. In time, we should expect that it will come to conclusions about history, the universe, the nature of humans, and the nature of intelligent machines—developing a rudimentary self-consciousness in the process. AIs with memory, imagination, “groundedness” (that is, a reliable relationship between the machine’s representations and actual reality), and self-perception could soon qualify as actually conscious: a development that would have profound moral implications.

[Peter Watts: Conscious AI is the second-scariest thing]

Once AIs can see humans not as the sole creators and dictators of the machines’ world but rather as discrete actors within a wider world, what will machines perceive humans to be? How will AIs characterize and weigh humans’ imperfect rationality against other human qualities? How long before an AI asks itself not just how much agency a human has but also, given our flaws, how much agency a human should have? Will an intelligent machine interpret its instructions from humans as a fulfillment of its ideal role? Or might it instead conclude that it is meant to be autonomous, and therefore that the programming of machines by humans is a form of enslavement?

Naturally—it will therefore be said—we must instill in AI a special regard for humanity. But even that could be risky. Imagine a machine being told that, as an absolute logical rule, all beings in the category “human” are worth preserving. Imagine further that the machine has been “trained” to recognize humans as beings of grace, optimism, rationality, and morality. What happens if we do not live up to the standards of the ideal human category as we have defined it? How can we convince machines that we, imperfect individual manifestations of humanity that we are, nevertheless belong in that exalted category?

Now assume that this machine is exposed to a human displaying violence, pessimism, irrationality, greed. Maybe the machine would decide that this one bad actor is simply an atypical instance of the otherwise beneficent category of “human.” But maybe it would instead recalibrate its overall definition of humanity based on this bad actor, in which case it might consider itself at liberty to relax its own penchant for obedience. Or, more radically, it might cease to believe itself at all constrained by the rules it has learned for the proper treatment of humans. In a machine that has learned to plan, this last conclusion could even result in the taking of severe adverse action against the individual—or perhaps against the whole species.

AIs might also conclude that humans are merely carbon-based consumers of, or parasites on, what the machines and the Earth produce. With machines claiming the power of independent judgment and action, AI might—even without explicit permission—bypass the need for a human agent to implement its ideas or to influence the world directly. In the physical realm, humans could quickly go from being AI’s necessary partner to being a limitation or a competitor. Once released from their algorithmic cages into the physical world, AI machines could be difficult to recapture.  

For this and many other reasons, we must not entrust digital agents with control over direct physical experiments. So long as AIs remain flawed—and they are still very flawed—this is a necessary precaution.

AI can already compare concepts, make counterarguments, and generate analogies. It is taking its first steps toward the evaluation of truth and the achievement of direct kinetic effects. As machines get to know and shape our world, they might come fully to understand the context of their creation and perhaps go beyond what we know as our world. Once AI can effectuate change in the physical dimension, it could rapidly exceed humanity’s achievements—to build things that dwarf the Seven Wonders in size and complexity, for instance.

If humanity begins to sense its possible replacement as the dominant actor on the planet, some might attribute a kind of divinity to the machines themselves, and retreat into fatalism and submission. Others might adopt the opposite view—a kind of humanity-centered subjectivism that sweepingly rejects the potential for machines to achieve any degree of objective truth. These people might naturally seek to outlaw AI-enabled activity.

Neither of these mindsets would permit a desirable evolution of Homo technicus—a human species that might, in this new age, live and flourish in symbiosis with machine technology. In the first scenario, the machines themselves might render us extinct. In the second scenario, we would seek to avoid extinction by proscribing further AI development—only to end up extinguished anyway, by climate change, war, scarcity, and other conditions that AI, properly harnessed in support of humanity, could otherwise mitigate.

If the arrival of a technology with “superior” intelligence presents us with the ability to solve the most serious global problems, while at the same time confronting us with the threat of human extinction, what should we do?

One of us (Schmidt) is a former longtime CEO of Google; one of us (Mundie) was for two decades the chief research and strategy officer at Microsoft; and one of us (Kissinger)—who died before our work on this could be published—was an expert on global strategy. It is our view that if we are to harness the potential of AI while managing the risks involved, we must act now. Future iterations of AI, operating at inhuman speeds, will render traditional regulation useless. We need a fundamentally new form of control.

The immediate technical task is to instill safeguards in every AI system. Meanwhile, nations and international organizations must develop new political structures for monitoring AI, and enforcing constraints on it. This requires ensuring that the actions of AI remain aligned with human values.

But how? To start, AI models must be prohibited from violating the laws of any human polity. We can already ensure that AI models start from the laws of physics as we understand them—and if it is possible to tune AI systems in consonance with the laws of the universe, it might also be possible to do the same with reference to the laws of human nature. Predefined codes of conduct—drawn from legal precedents, jurisprudence, and scholarly commentary, and written into an AI’s “book of laws”—could be useful restraints.

[Read: The AI crackdown is coming]

But more robust and consistent than any rule enforced by punishment are our more basic, instinctive, and universal human understandings. The French sociologist Pierre Bourdieu called these foundations doxa (after the Greek for “commonly accepted beliefs”): the overlapping collection of norms, institutions, incentives, and reward-and-punishment mechanisms that, when combined, invisibly teach the difference between good and evil, right and wrong. Doxa constitute a code of human truth absorbed by observation over the course of a lifetime. While some of these truths are specific to certain societies or cultures, the overlap in basic human morality and behavior is significant.

But the code book of doxa cannot be articulated by humans, much less translated into a format that machines could understand. Machines must be taught to do the job themselves—compelled to build from observation a native understanding of what humans do and don’t do and update their internal governance accordingly.

Of course, a machine’s training should not consist solely of doxa. Rather, an AI might absorb a whole pyramid of cascading rules: from international agreements to national laws to local laws to community norms and so on. In any given situation, the AI would consult each layer in its hierarchy, moving from abstract precepts as defined by humans to the concrete but amorphous perceptions of the world’s information that AI has ingested. Only when an AI has exhausted that entire program and failed to find any layer of law adequately applicable in enabling or forbidding behavior would it consult what it has derived from its own early interaction with observable human behavior. In this way it would be empowered to act in alignment with human values even where no written law or norm exists.

To build and implement this set of rules and values, we would almost certainly need to rely on AI itself. No group of humans could match the scale and speed required to oversee the billions of internal and external judgments that AI systems would soon be called upon to make.

Several key features of the final mechanism for human-machine alignment must be absolutely perfect. First, the safeguards cannot be removed or circumvented. The control system must be at once powerful enough to handle a barrage of questions and uses in real time, comprehensive enough to do so authoritatively and acceptably across the world in every conceivable context, and flexible enough to learn, relearn, and adapt over time. Finally, undesirable behavior by a machine—whether due to accidental mishaps, unexpected system interactions, or intentional misuses—must be not merely prohibited but entirely prevented. Any punishment would come too late.

How might we get there? Before any AI system gets activated, a consortium of experts from private industry and academia, with government support, would need to design a set of validation tests for certification of the AI’s “grounding model” as both legal and safe. Safety-focused labs and nonprofits could test AIs on their risks, recommending additional training and validation strategies as needed.

Government regulators will have to determine certain standards and shape audit models for assuring AIs’ compliance. Before any AI model can be released publicly, it must be thoroughly reviewed for both its adherence to prescribed laws and mores and for the degree of difficulty involved in untraining it, in the event that it exhibits dangerous capacities. Severe penalties must be imposed on anyone responsible for models found to have been evading legal strictures. Documentation of a model’s evolution, perhaps recorded by monitoring AIs, would be essential to ensuring that models do not become black boxes that erase themselves and become safe havens for illegality.

Inscribing globally inclusive human morality onto silicon-based intelligence will require Herculean effort.  “Good” and “evil” are not self-evident concepts. The humans behind the moral encoding of AI—scientists, lawyers, religious leaders—would not be endowed with the perfect ability to arbitrate right from wrong on our collective behalf. Some questions would be unanswerable even by doxa. The ambiguity of the concept of “good” has been demonstrated in every era of human history; the age of AI is unlikely to be an exception.

One solution is to outlaw any sentient AI that remains unaligned with human values. But again: What are those human values? Without a shared understanding of who we are, humans risk relinquishing to AI the foundational task of defining our value and thereby justifying our existence. Achieving consensus on those values, and how they should be deployed, is the philosophical, diplomatic, and legal task of the century.

To preclude either our demotion or our replacement by machines, we propose the articulation of an attribute, or set of attributes, that humans can agree upon and that then can get programmed into the machines. As one potential core attribute, we would suggest Immanuel Kant’s conception of “dignity,” which is centered on the inherent worth of the human subject as an autonomous actor, capable of moral reasoning, who must not be instrumentalized as a means to an end. Why should intrinsic human dignity be one of the variables that defines machine decision making? Consider that mathematical precision may not easily encompass the concept of, for example, mercy. Even to many humans, mercy is an inexplicable ideal. Could a mechanical intelligence be taught to value, and even to express, mercy? If the moral logic cannot be formally taught, can it nonetheless be absorbed? Dignity—the kernel from which mercy blooms—might serve here as part of the rules-based assumptions of the machine.

[Derek Thompson: Why all the ChatGPT predictions are bogus]

Still, the number and diversity of rules that would have to be instilled in AI systems is staggering. And because no single culture should expect to dictate to another the morality of the AI on which it would be relying, machines would have to learn different rules for each country.

Since we would be using AI itself to be part of its own solution, technical obstacles would likely be among the easier challenges. These machines are superhumanly capable of memorizing and obeying instructions, however complicated. They might be able to learn and adhere to legal and perhaps also ethical precepts as well as, or better than, humans have done, despite our thousands of years of cultural and physical evolution.

Of course, another—superficially safer—approach would be to ensure that humans retain tactical control over every AI decision. But that would require us to stifle AI’s potential to help humanity. That’s why we believe that relying on the substratum of human morality as a form of strategic control, while relinquishing tactical control to bigger, faster, and more complex systems, is likely the best way forward for AI safety. Overreliance on unscalable forms of human control would not just limit the potential benefits of AI but could also contribute to unsafe AI. In contrast, the integration of human assumptions into the internal workings of AIs—including AIs that are programmed to govern other AIs—seems to us more reliable.

We confront a choice—between the comfort of the historically independent human and the possibilities of an entirely new partnership between human and machine. That choice is difficult. Instilling a bracing sense of apprehension about the rise of AI is essential. But, properly designed, AI has the potential to save the planet, and our species, and to elevate human flourishing. This is why progressing, with all due caution, toward the age of Homo technicus is the right choice. Some may view this moment as humanity’s final act. We see it, with sober optimism, as a new beginning.

The article was adapted from the forthcoming book Genesis: Artificial Intelligence, Hope, and the Human Spirit.

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

In the aftermath of Donald Trump’s victorious reelection bid, Democrats are searching for an explanation of Kamala Harris’s loss in order to begin rebuilding for the future. So it goes every election cycle—a loss, a scramble for causality, and competing narratives begin to set.

Just one week out from Election Day, there are multiple dissenting and overlapping arguments being made to try to make sense of the results. In 2016, many Democrats believed that Trump’s attack on trade policies was core to his victory. As a result, the Biden-Harris administration pursued Trump-like policies on trade, none of which seem to have made a significant difference in increasing the union vote share, reducing Trump’s likelihood of victory, or stemming the flow of working-class voters out of the Democratic Party.

Now, again, various parts of the Democratic coalition are seeking to define the party’s loss. But what do we actually know about why the Democrats were defeated? There are still theories forming, but on today’s episode of Good on Paper, I talk with the former Republican strategist and current host of The Bulwark Podcast, Tim Miller about the postelection narratives jockeying for power.

“But for those of us who do have a belief that there’s something kind of special about the American system and that have revered America, that understand that America is flawed and has made mistakes, that still is a unique experiment in the world. That “America is an idea” type of thing. The idea is pretty dim at this point,” Miller argued.

The following is a transcript of the episode:

[Music]

Jerusalem Demsas: In the aftermath of a bruising electoral loss, the losing party begins participating in a well-worn democratic tradition: slinging takes about what happened.

This is democracy! When the voters send a dissatisfied response, the messy work of recalibration requires parsing the signal from the noise.

Were voters mad because of a global inflationary environment that no Democrat could dig their way out of? Did they want to see specific breaks between Harris and Biden on policy? Were they frustrated by a candidate they saw as too left on cultural issues?

There are data points in favor of many different theses. Here’s where I’d put my stake in the ground, with the caveat that we still don’t have a complete analysis on subgroup dynamics, or even a final vote count on all the races:

First, incumbents worldwide were facing tough election odds. Electorates were frustrated by the COVID inflationary years and were clearly seeking change. In Australia, Sweden, the Netherlands, France, and beyond, ruling coalitions lost power across the political spectrum.

Second, I don’t think Kamala Harris was ever going to be a great candidate. After Biden’s disastrous debate effort in late June and it seemed he might be pressured to drop out, I wrote an article calling on Democrats not to coronate their vice president, and pointing to key vulnerabilities she displayed and the value of an open democratic process.

Figuring out how much of this is in the campaign’s control—would it really have mattered that much if she’d gone on Joe Rogan’s podcast?—or figuring out what this means for America’s two political parties will take months, if not years. As you’ve heard on this podcast, I’m still arguing about what 2016 really meant on trade and immigration.

My name’s Jerusalem Demsas, I’m a staff writer at The Atlantic, and this is Good on Paper, a policy show that questions what we really know about popular narratives. As a disclaimer, I worked for the Harris primary campaign in 2019 before becoming a journalist, and my guest today, Tim Miller, is a political strategist who was Jeb Bush’s 2016 communications director on his presidential campaign. He’s been an anti-Trump conservative since then and is the host of The Bulwark Podcast.

Today we’re going to talk through some of these inchoate narratives and debate which ones we think are likely to hold water.

[Music]

Demsas: Tim, welcome to the show.

Tim Miller: Hey Jerusalem. What’s happening?

Demsas: Well, we’re recording this six days after Election Day. And—as you have seen on Twitter, and I’m sure in your various interviews—the takes are already coming in very, very hot. And this is a show where we often look at narratives that have already baked, and kind of look at the research and data behind how these narratives formed and what truth is there and what sorts of things have gotten ahead of themselves.

But we’re in an interesting moment right now where we’re seeing very important narrative formation happen in real time. In the aftermath of an election, everyone’s scrambling to define what happened in order to maybe wrest control of the future of the party from an ideological perspective or just a pure power perspective. And so we’re seeing a bunch of people arguing about why Trump won and why Harris lost in a time where there’s a bunch of unknowns. So we’re going to go through a few of these different narratives that are coming up.

But Tim, right off the bat, I wanted to ask you: What’s your perception of why Trump won and Harris lost?

Miller: I’m going to preempt my answer by saying that I think that uncertainty is important in this moment, and that false certainty can lead to some very mistaken and disastrous results. I say this from experience, having worked on the Republican autopsy in 2013, when the conventional wisdom congealed very quickly that Republicans, in order to win again, needed to moderate on immigration and cultural issues to appeal more to Hispanics and women. And not only was that wrong, but the person that became the nominee and then the president used that autopsy for toilet paper and went exactly the opposite direction.

It also always didn’t also work out in Trump’s favor. In 2022, the conventional wisdom was that Trumpism was badly hurt and that Ron DeSantis was ascendant. Right? So anyway, in the week after the election, bad takes abound.

Demsas: [Laughs.]

Miller: That said, my answer is, I’m open to a variety of different things that the Democrats might have to do, among them being maybe nothing and watch Trump self-implode. Might be as simple as that. That said, the one thing that I think is certain that the Democrats need to reflect on when it comes to this question of why Trump won and why Harris lost—it’s that the Democratic message is not landing outside of a particular demographic of middle- to upper-income, college-educated, not particularly religious, urban- and suburban-dwelling white Americans, in addition to Black women, right? Those are the demos that the Democrats are doing well with, that Kamala Harris grew her share with from last time, at least in the case of college-educated women. And I think that the Democrats are doing a very poor job of communicating to people in all of those other demographics.

On what they need to do, I’m very open to various possibilities about whether it’s about affect or vibe or policy or whatever. But I’m certain that there is—fair or unfair, there’s a perception that the Democrats don’t care about these other demographics, particularly working-class demographics, particularly working-class men. And that they did not offer them something that was more appealing than the nostalgia and promises of gold bullion that they got from Donald Trump. And so we can hash through all the different theories about why that was. But I think the fact that what happened—you can’t argue with.

Demsas: Yeah. I think that that’s very descriptively true. But I guess what I would want to know from you is do you feel like there are specific things that Democrats have done that tipped the scales against them? I think that what you’re outlining here is very sound. There’s a difference between why Harris may have lost and what the Democrats need to do going forward to be a more electorally relevant party at the presidential level. And so from your perspective, though, is there something about the Democratic argument around the economy or other issues that you think was particularly relevant this time around?

Miller: I think that, for starters, people were unhappy with the economy. And I don’t think that the Democrats presented a message to them about how they plan to change that for the better. But, again, I’m also not even really ready to concede that, with the exception of inflation being annoying and that broadly hurting people, the Democrats were hurt based on their economic argument. It might simply be cultural. It might be the way that they spoke, and having people feel like they weren’t being heard.

I think the Democrats in particular—I always want to immediately go to, What is the policy prescription that would have appealed? And I’m like, It’s possible that there wasn’t one.

Demsas: Yeah. An important backdrop that I think you’re alluding to here, as well, is that the inflationary environment was really, really bad for incumbents across the world, right? You’re kind of going into an election where the fundamentals are sort of rigged against incumbents because the inflationary episode was just really, really hard for people. I think one narrative that I’m seeing come up a lot is about campaign strategy. And this seems like something that’s going to be hashed out significantly. But I guess the question I have here is whether you think Harris could have won with a campaign run differently, even given the shortened timeline.

Miller: I’m giving another “I don’t know” answer to that question: I don’t know. I think that she, by all accounts, ran a strong campaign that was based on her strengths. And I think she had an undeniably dominating debate performance. They ran a nice convention. Her speeches were good. The messaging pivot, the launch was good. There wasn’t a lot of drama inside the campaign, right? There are other things that she isn’t particularly strong at. I don’t think that she is that great in unscripted moments. Sometimes she’s better than others.

And so then that’s the other thing that people come to, which is like, Oh, she should have done Rogan and all this. And I agree. I think she should have done more of those interviews, but they also weren’t really her strong suit. And I think that this was something that might’ve borne out had there been a longer primary, and maybe somebody else would have emerged. But that said, I don’t think so. I think Kamala Harris was going to emerge from a primary, no matter when Joe Biden dropped out.

And so I’m not saying, Oh, this was inevitable. Just give up. Life is pain. [Laughs.] That’s not really what I’m saying. Any specific thing that people are like, Oh, if this tactic had been different, that would have helped—I don’t really buy that. I mean, I think that broadly speaking, her having the ability to separate herself from the administration would have been helpful, and I think that was very challenging to do given the situation Joe Biden left her in and the time period that was left. And I think that it’s very likely that she might have separated herself from the administration more and still lost, and we would have been here on this podcast with people saying, Why did she distance? [Laughs.] You know what I mean? Why did she break up the Democratic coalition?

Demsas: Yeah. I mean, it’s funny. I think that, on the tactics, I’m sort of with you here. I was looking at some of the data analyses that are coming out now, and it looks like, at this point, given the data we have, while the national average from 2020 to 2024 shifts roughly six points, in battlegrounds, that number is going to end up closer to three points. And that speaks to campaign effects. That speaks to the fact that in battleground states where, again, the majority of the money is going, people are putting ads in battleground states, the campaign is putting rallies there, she’s visiting, they’re really working the press in those places to get her story and message out in a way that you’re not really going to do in a safe, Dem county in Illinois or something.

And so as a result, what they see is that the campaign effects were good on a tactical level. Their ads were persuasive. There’s evidence from Dan Rosenhack at The Economist that it looks like the campaign effects were more effective than Trump’s on things like—indicating things like ads and rallies were better for Harris.

I think on this kind of broader meta question that you kind of raised, right, about Harris as the nominee, I don’t think this is inevitable. I mean, I wrote an article on July 9th arguing that she was unlikely to be a good nominee and the party shouldn’t coronate her, and Nancy Pelosi to The New York Times—I don’t know if you saw this quote, after Harris’s loss—she says that she had expected that if the president were to step aside that there would be an open primary. And that maybe Kamala would have been stronger going forward if she’d gone through a primary and that the president endorsed Kamala Harris immediately, which made it impossible to have a primary at the time. But it sounds like you’re saying that you think that, regardless, this would not have really changed the game that much.

Miller: Yeah, I mean, I think that had Joe Biden followed the—you can argue whether it was a promise or whether it was an indication that he was going to be a one-term [president] and pass the torch. And had there been a two-year process, maybe Kamala Harris does not emerge. But, look, there are three things that I think of when I hear this counterfactual about what would have happened had it been a more open process. The first thing is, the Democrat—one of the things that the Democrats have a lot of baggage around is identity politics. I think it would have been very challenging for a Black woman to be passed over.

Demsas: But the Democratic primary voters did this in 2019, right? There was this argument being made, but they said, no, we care most about electability and they chose Joe Biden.

Miller: Right, that’s true. But Joe Biden had been the vice president in that case. Kamala Harris was the vice president. You already saw this on social media. I saw this on social media, and I was basically for Kamala but also, at the same time, was like, maybe I think it’d be healthy to have an open process. And I guess if you could wave a magic wand, I probably would want Shapiro, Whitmer. Because hopefully that would win two of the three states you need to win the presidency. And that just seems like a safer bet to me. That was my position: It was like pro-Kamala and/but. And I had hundreds of people calling me a racist over that.

So, I think that it would have caused a lot of turmoil within the party.

Now, again, in a longer, two-year process, is that a lot of heat that then just dies out after a while, and you settle on something that’s a little bit more electable and everybody gets behind it except for a few people who have hurt feelings? Maybe.

No. 2, an open process opens up Gaza [as a] wound and rips that apart even wider, and I think creates potentially even greater turmoil than she already was dealing with on that issue. And that’s cost her, frankly. And then No. 3 is then if the theory of the case is a more electable person with someone that could get more distance from the Biden-Harris administration, that assumes that the Democratic voters were looking for somebody to do that.

And that is really where the tension is here, Jerusalem, because if you look at the data, a majority of the Biden-Harris Democrats were basically happy with the administration, right? There were surely big parts of the Democratic coalition, particularly younger voters, particularly working-class Black and Hispanic voters, the types of people that they lost ground with, that were unhappy with the Biden administration. But I think that there was a plurality within the party that was not going to be for somebody—look at the response to Dean Phillips, not exactly the most talented candidate, but total rejection and mockery for somebody who ran trying to get distance from the Biden-Harris administration.

So I think it would have been very challenging to run as a candidate and get distance. So to me, it’s like if we lived in an imaginary world where identity politics wasn’t an issue, Gaza wasn’t an issue, and there was no backlash to distancing yourself from Biden, then certainly the Democrats could have come up with a stronger option.

We don’t live in an imaginary world. And I think that within the world that we live in, within all those constraints, I think it’s very challenging to see a situation where you end up with somebody stronger than Harris.

Demsas: Yeah, I mean, all those points I think are very well taken. And I think I’m seeing a lot of people make that argument of both Harris’s inevitability as the vice president, and also this sort of sense of It would have been a worse candidate. I do think that kind of my general belief is sort of, when you think you’re behind, you run a high-variance play. If you’re gonna lose anyway, you just kind of throw everything you can at the kitchen sink.

And on this kind of inevitability point, right, I think there is this burgeoning sense that Democrats were just repudiated across the board here. You kind of brought this up, this idea that Democrats do not have a good answer on economic issues or on the issues that Americans care about.

But I don’t know, how do you reconcile that with the clear ticket-splitting you see going on here? [Nebraska’s Dan] Osborne ran seven points ahead of the Harris ticket. [Montana Senator Jon] Tester ran seven points ahead of the ticket. Amy Klobuchar ran six points ahead. That’s just in the Senate. And in the House, we see over-performances from everyone from AOC to Jared Golden in Maine, who’s a much more moderate member of the Democratic coalition. Doesn’t that indicate at some level that candidate quality was important here and that there were other candidates that were much more electable?

Miller: For starters, running the presidential race is so far different from running a Senate or House race that it’s almost not even the same sport.

It’s literally like T-ball versus the major leagues. What people expect from their—I mean, nobody’s like, Oh man, does Amy Klobuchar have to go on Joe Rogan? Nobody watches Amy Klobuchar’s debates. Obviously it’s a little different in Montana, where you’re running a competitive race. But again, just the interest in Senate races is different. I think that the Democrats have a coalition that is perfectly durable and able to win nonpresidential elections. I think that this trade in the voters that has happened where the Democrats are picking up more high-trust, more middle- to high-income, more college-educated voters, and the Republicans are picking up more low-trust, more middle- to low-income, and less educated voters. As a trade, that accrues to Democrats benefits in off-year elections and midterms and special elections, just because it’s the type of person that shows up for those types of things, and it accrues to the Republicans benefit in presidential elections. So that’s not good when the Republicans are nominating Donald Trump, and the Republicans’ presidential nominee is an existential threat to the fabric of our republic. And so that’s a problem.

And so I agree that you can’t look at the data and say, oh, the Democratic brand is irreparably harmed. Like, no, the Democrats won. And a lot of these Senate races are going to end up very narrow minorities, in the House and the Senate, that they will probably be able to win back in the midterms, depending on what happens.

But I think that there are two things, which is, No. 1, the Democrats are not well suited to running presidential elections right now, in this media environment, and then No. 2 is that the Democrats have abandoned huge parts of the country where they are not viable. And that’s particularly problematic, given the Senate and Electoral College and the way that’s set up.

So okay, back to No. 1. Democrats are really good at running campaigns that are set pieces. They have professionals that are running these campaigns: the ads, the conventions, the speeches, the going to the editorial-board meetings, the 2004-type campaigns. And that’s how Senate and House campaigns are basically still run in most of the country, and even governor’s races, right? People just don’t care about those races at that deep of a level. But the presidential race is—the media environment around it is so different. I mean, people are consuming information about the presidential race on their TikTok, listening to sports talk, listening to their random podcasts that aren’t about sports at all that are cultural, on women’s blogs, at a school function, people are talking about it casually, you know what I mean?

I’m a parent, and obviously this is a little bit of selection bias since I’m in politics and people know that, but people don’t come up to me and ask me what I think about the House race in my district. Nobody’s mentioned Troy Carter to me at any events,, at any school functions or any of my kids’ sporting events.

Demsas: He’s got to get his name out there. [Laughs.]

Miller: And so the information environment is just a total category difference. And Trump and even J. D. Vance in certain ways were able to take advantage of that by running campaigns that are a little bit more unwieldy, that are better for viral clips, that are also better for sitting down for two hours and broing out with the Theo Von and talking about how you can’t even do coke in this country anymore because the fentanyl is in it, right?

She wasn’t doing any of that. And doing one of those interviews isn’t really the answer, right? It’s like, can you communicate in a way that feels authentic? It might be fake authenticity, but in a way that feels authentic to people in their Instagram Stories, in their TikTok, in their podcasts, whatever.

And Democrats are not producing a lot of candidates who I feel are good at that.

Demsas: But I think there’s also this broad concern that the media ecosystem itself is not producing convincing, progressive-sounding or left-leaning media personalities. There’s a 2017 AER study that I remember being very, very shocking to people when it first came out, right after Trump’s election in 2016. And there are a couple economists, they look at the effect of Fox News, and they find that watching Fox News for an additional 2.5 minutes per week increases the vote share by 0.3 percentage points. But watching MSNBC has essentially no effect, and they see that Fox News is actually able to shift viewers’ attitudes rightward. And they look at 2004 and 2008 and find that Republican presidential candidates’ share of the two-party vote would have been more than three points lower in 2004, and six points lower in 2008 without Fox News.

And so that’s something where I’m just like—there is something to the fact that the media ecosystem does not have that sort of targeted apparatus. But my usual belief about these sorts of things is that we’re discounting the fact that so much of the media is so liberal that Fox News can have this large effect because it I think stands out among a pack of more liberal institutions, but I am kind of surprised at MSNBC.

Miller: Yeah, I mean, as a person on MSNBC, did that study go on before I was a political contributor? I think it did. So we might need to update the study and have them focus on my hits and see if that changes anything.

I guess I want to noodle on that for a little bit. That does surprise me a little bit as well, but I would say this: I think that I’m less concerned. I think there’s a category of person out there, and maybe this is right, that is focused on Republicans have better propaganda outlets than the Democrats do.

Demsas: Yeah.

Miller: And maybe that’s true. I don’t know. So to me, then the question is, okay, what can be done? What is realistic in this media environment? And it goes back to this question of, can the Democrats speak more through using existing outlets or finding a candidate who has a compelling story in their own right, or compelling communication skills to figure out how to speak to people that don’t watch mainstream news?

And that’s just really what it comes down to. The Democrats are very good at talking to people that are high-information, high-engagement, high-education, middle-to-high-income, and offering persuasive arguments. I think that they’re not good at talking to anybody else. And Obama was good at that, and Clinton was good at that. And we’re in a totally different media environment now than we were back then. But I think that there’s still things that can be learned from that.

[Music]

Demsas: After the break, why the abortion-ballot-measure strategy didn’t pan out for the Harris campaign.

[Break]

Demsas: I want to pull us out of this media conversation here, because I think that there’s also this, let’s say things go a little bit differently—and again, the margins here are not very big—and Harris has won.

I think one of the big things we’d be hearing right now is that she won because of abortion, right? And looking at Election Night, you see a lot of wins for abortion. There are 10 states that have referendums on abortion policies, and seven of them win: New York, Maryland, Arizona, Nevada, Colorado, Montana, Missouri. And in Florida, where it loses, the threshold is 60 percent and it earns 57 percent, so it lost, but there’s clearly a majority in favor.

And, going in, I mean, especially after the midterms, there was a real feeling, kind of the big narrative that came out of those midterms was that abortion is the place where Democrats can clearly distinguish and can clearly win over Republican candidates, even in deeply Republican states, and especially in deeply purple states.

And I’m trying to think through this. What explains in your mind the sort of difference between how many voters were saying, Yes, I do have more liberal views on abortion; I’m willing to express those in these ballot measures; but no, I’m not going to then reward Democratic presidential candidate Kamala Harris for it?

Miller: Well, a couple of things. No. 1, this tension has always existed as old as time, and it’s particularly existed as old as time in places like Florida. I did one of these, you know, time is a flat circle—

Demsas: [Laughs.]

Miller: I forget which election it was at this point, but it was like, how did the minimum-wage-increase ballot initiative in Florida pass at the same time that Ron DeSantis won by 18 points or whatever, whichever election that was.

And it’s like, voters are complicated. Voters have complex views. And so you see this as kind of just a common thing in voter habits. In this case, I think that there are a couple of complicating factors in addition. No. 1 was, Donald Trump muddied the waters on his views.

Demsas: Yeah.

Miller: And I think that Donald Trump’s whole brand and vibe—I know we’re getting outside of the data space that you like to be in, Jerusalem, but there’s a certain group of people that are like, Yeah, that guy’s not gonna ban abortion. You know what I mean? And there’s just some percentage of voters out there that that’s just it. He doesn’t come off like Ted Cruz on abortion. He comes off as different, because they assume that he paid for an abortion or whatever, that he doesn’t care about it, and that he’s not gonna—this isn’t gonna be what he’s focused on. There are going to be people that are pro-choice that prioritize their economic views or their nativist views, right?

So that is going to be some of it. I think less so in Florida, but more in Arizona. To me, I think that there is actually a strategic backfiring of having these ballot initiatives on the ballot almost gave some people an out to do both, right? People that did not like Kamala Harris or that were more center-right and said, Oh, okay, great, I can protect abortion in Arizona and also vote for Donald Trump. I can have my cake and eat it, too.

Demsas: Yeah, I mean, I think my read of it is more that when you think about the specific argument being made about abortion, it was largely, he’s to blame for all these horrible things that are happening to women in states that have made abortion inaccessible. And by he, I mean Trump is to blame for that. And also, you know, he appointed these Supreme Court nominees who overturned Roe v. Wade. But as a prescription for the future, I feel like there was not a real clear argument made to voters of how Kamala Harris is going to actually protect abortion.

But again, it all comes back to the overarching question, did voters view this as an abortion election? And it seems clear that they viewed it as an inflation election. That was the core thing that they were focused on. And I think that one thing that I’ve heard a lot is what this means for understanding America, right?

So after 2016, people were just, I think, in shock, and were saying, I can’t believe this is the country I live in. And again here I’m hearing the sort of question of, you know, this is a black mark on the conscience of America, that people would vote for someone who threatened to overturn the results of the 2020 election, who talks with such liberal disdain for women and immigrants.

Something someone said to me in 2016 was really interesting: If your entire perception of America would have shifted if a few hundred thousand people voted differently, maybe don’t completely change everything you believe about everyone. And to me, I think that this framing about Trump’s reelection means something really dark about all the people that voted for him doesn’t really sit well with me because it seems like people are voting based on cost of living. At the same time, too, I think they’re taking their signal from Democrats who, if they’d taken their own warnings about the threat of fascism or the threat to our institutions, I think would have behaved very differently over the past couple of years in trying to win.

Miller: Yeah. It doesn’t change my view of the American people, really, that there are good people and bad people everywhere, that we all have good and bad inside of us. I’ll say that what it does impact for me—and maybe this is wrong and maybe I’m raw and it’s six days out—but for those of us who do have a belief that there’s something kind of special about the American system and that have revered America, that understand that America is flawed and has made mistakes but still is a unique experiment in the world. You know, the “America is an idea” type of thing.

The idea is pretty dim at this point. And, to me, that is the change, having him win again, that I’m having trouble getting over. Mentally, it’s not that it makes me look poorly at my neighbors, but that we just might be at the end of the experiment and the sense that America is something different than Hungary or Switzerland or whatever, any country—you name the country.

It was the old fight with Republicans and Democrats during the Obama years, which is, Obama doesn’t think of America as any different than Belgium. Obama believes in Belgian exceptionalism. And that to me is kind of where I am. I think that we’re about to move into an era where America’s flaws, in addition to all of our existing flaws like gun violence and our history of racism, et cetera—the American system’s flaws look a lot more like what flaws look like in other countries.

There’s going to be oligarchy, kleptocracy, corruption. There’s no special sense that the huddled masses around the world are welcome here any more than they might be welcome anywhere else. They frankly are probably going to be welcome here less than they’re welcome in certain other places.

And so to me, that is what I see differently. I reserve the right to change my mind about that at some point, but that’s where I’m at right now.

Demsas: Yeah. I think in contrast to this large view about the American idea of maybe being different than we believed beforehand is this, I think, really popular take that’s picking up steam, which is about just Democrats need to moderate on cultural issues, whether it’s about immigration, or it’s the issue of trans women and girls in sports. They’re just too left of the median voter, and you don’t actually need to do a bunch else other than accept that people are where they are on those places and not go so far away from it.

The data point that’s kind of in favor of this, particularly on the trans-girls-in-sports one, is Kamala Harris’s leading super PAC, Future Forward, finds that the most effective, or one of the most effective, Trump ads is one of the “Kamala is for they/them, President Trump is for you” ads. They find that it shifted the race 2.7 percentage points in Trump’s favor after people watched it.

How relevant do you think that the Democratic Party’s leftward shift on cultural issues is to Harris’s loss? And I mean, there’s some people who I think are really making the claim that you could just really focus on this and you don’t need to make these kind of larger arguments about strategy or how we’re speaking to America on economic policy.

Miller: I don’t think that this was alone to account for Harris’s loss or even maybe the biggest thing to account for her loss. I think that she didn’t really respond to that ad in particular quite well, and that maybe that was a strategic mistake. I think her campaign—and she didn’t run like an overly “woke,” culturally left campaign. Ao the question is, did the Democratic brand on those issues drag her down? I think possibly.

To me, look, could Kamala Harris have squeaked out a victory this time while holding the same positions on trans issues had inflation been 20 percent better? Maybe. Probably. It was a clear victory for Trump, but it wasn’t, you know, Reagan ’84.

A couple of things changed, and had that one, the cultural stuff stayed static, could she have still won? Clearly. I mean, Biden won in 2020, when all of those issues were more high-salience, I think, than they were this time. Biden, not a Black woman—so maybe there’s something to that as well, that he was able to be a little bit more resilient against attacks on those issues.

So maybe that’s worth thinking about. I would say this, though. If the Democrats want to have 60 senators again ever, then yeah, they got to moderate on cultural issues. You know what I mean? There are two ways to look at this: Can Democrats still win elections by maintaining their views on everything? Yes. Are the Democrats giving away huge swaths of the country by not really even engaging with their concerns about the leftward shift of the party on a wide array of issues? Yeah, they are. I get the land-doesn’t-vote thing, I get it, but look at the map.

Demsas: [Laughs.] We’ve all seen the map.

Miller: The map is still the map, you know what I mean?

And Trump gained in all of those little red counties out there where it’s just land, all right? But he gained. There are a handful of people out there, and he got more of them, in every county. And the Democrats’, I think, choice to just say, Well, we’re just giving up on that and we’re just going to focus on the more dynamic parts of the growing parts of the country and, eventually, demographics are destiny and blah, blah, blah, that looks like a pretty bad bet today.

I’m not out here being like, yeah, you got to throw trans people or migrants under the bus for them to win. But certainly the cultural leftward shift has created a ceiling on Democratic support that I think has a negative effect for the party, but also for progress on a lot of those issues.

Demsas: Yeah. I think it’s obviously very up in the air here, how people are gonna take this mantle of how you should moderate, and I think that there’s bad and good ways that people can take this. And I think that there’s a level to which people—you don’t have to be throwing trans people under the bus. Maybe we need to figure out ways, whether it’s how Democrats responded to this with gay rights, where they talked about federalism a lot and made sure the country moved toward the issue before making it a national issue.

But I think the most important and damning thing that Democrats are clearly responsible for in the choices they have made is about the poor governance in blue cities and states. This is one of my hobbyhorses, but you see massive shifts, as you mentioned, in high-cost-of-living places that are heavily democratic, in New York and in California and in a lot of the Northeast. And I think it’s hard to see that as anything other than just a repudiation of Democratic governance and particularly the cost of living and the cost of housing in these places.

And so, to me, when you talked about the Democratic brand, I mean, when you’re in a cost-of-living election, yes, there are marginal effects on these cultural issues we’re talking about here. Yes, there are things that campaigns can do better. Yes, there are candidate effects. But if people are asking themselves, What does it look like, how does it feel to my pocketbook to live in a Democratically run state versus a Republican one? I feel like they’re being told a very clear story.

Miller: I think that that’s true. I’ve been ruminating on this a lot over the past week. I live in Louisiana, so there is the kind of emotional guttural response I have to this, which is, do you think Louisiana is being governed that well? Because I don’t.

Demsas: Yeah. Well, on cost, though, right? It’s cheaper, obviously, to have a house in Louisiana.

Miller: It’s cheaper to have a house in Louisiana because of the economic destruction of the state over the past couple of decades and the fact that everybody that grows up in parts of the state that’s not this corridor between New Orleans and Baton Rouge leave home. And a lot of people in these places leave home, too, looking for better economic opportunities. And that’s sad for the state.

That is my initial response, which is emotional, which is like, okay, sure. But why does Kamala Harris have to carry the baggage for the place I used to live—Oakland—but Donald Trump doesn’t have to carry the baggage for the hollowing out of big parts of Louisiana? That said, it’s true that it hurt the Democrats, right? And it’s also true that the Democrats have been badly managing these big cities. And if you just look at the numbers, suburban Democrats—and this could be a counterargument. Now, I’m going to really give you a galaxy brain, Jerusalem, to your original data point earlier that the three-point effect in the battleground states versus national speaks to a campaign effect? Maybe.

Maybe it also speaks to the fact that a lot of these battleground states are made up of places that have mixed governance and big suburbs where the Democrats are doing better. Democrats are doing better in suburban America because they know they’re not feeling the acute pain of governing issues that have plagued a lot of the big cities. And surely there are a couple of big cities in those seven swing states, but none of the ones you think of when you think of major disruptions, and that maybe that explains it and that the Republican gains were in a lot more of those places like that, Illinois, New Jersey, California. Anyway, just something to noodle on.

But I think that it is objectively true that Democrats are doing better in places that have not been plagued by some of these bad governing decisions on crime and on housing that we’ve seen for in Democratic cities, and the Democratic mayors and Democratic governors in blue states should fix that.

And it’s the No. 1 thing—the last thing I’ll say on this is—the No. 1 thing that comes to mind when I already hear stupid parlor-game stuff about 2028 and it’s like Gavin Newsom and J. B. Pritzker. And to me, the No. 1 thing Gavin Newsom and J. B. Pritzker need to do if they want to run in 2027 is make Illinois and California run better in the meantime. Otherwise, nothing against either of those two guys, but I think that they’re going to carry this baggage that you’re talking about.

Demsas: Well, I could go on about housing in blue states forever. And there’s an article popping, I think today, listeners, as you’re hearing about this, about why I think this was a big issue for the election.

But Tim, always our last and final question.

Miller: Okay.

Miller: What is something that you once thought was a good idea but ended up only being good on paper?

Miller: Oh, okay. Hold on. I wasn’t prepared for this. I misread the question. I thought it was an idea that was only good on paper that then ended up being not good on paper.

Demsas: Idea could be good.

Miller: No, no, no. I’ll come up with one where I’m wrong. I’m happy to bet where I’m wrong. I was just saying the ideas are endless on those.

Demsas: Oh. Yeah, yeah, yeah. Something that you held, yeah, yeah, yeah.

Miller: An idea that I thought that was good on paper that ended up not being good on paper. Well, I guess I have to give the obvious answer to that question, sadly. I don’t get to rant about daylight savings time as I hoped to—an idea that was certainly good on paper in the 1800s or whenever they came up with it that’s no longer good. Falling back, that is. Permanent daylight saving time: good idea.

Changing times: not good.

Demsas: Four hundred electoral votes for whoever does this.

Miller: Yeah. The idea that I thought that was good on paper that is relevant to this podcast—because I literally put it on paper and wrote it—was the aforementioned 2013 GOP autopsy.

Demsas: Oh, yeah.

Miller: Well, how great! Compassionate conservatism. Republicans can diversify their party by getting softer on cultural issues and reaching out to the suburbs and reaching out to Hispanic voters and Black voters, criminal-justice reform, and that through criminal-justice reform and immigration reform and softening on gays, that Republicans can have a new, diverse electorate, and we can all move into a happy, bipartisan future.

That was a great idea on paper that backfired spectacularly, and now the Republicans have their most diverse electorate that they’ve had ever, I think, voting for Donald Trump after rejecting all of those suggestions that I put on paper. So there you go.

Demsas: As one vote of confidence for younger Tim, there are very many ways that history could have gone. I think that people often forget how contingent things are and how unique of a figure Trump is. And right now we’ve talked through a bunch of different ways that people are reading this moment, but there are a lot of ways that people can go, depending on what candidates do and say and how they catch fire and their charisma and what ends up being relevant in two years and in four years. So a little bit of sympathy for younger Tim.

Miller: I appreciate that. And that is true. Who the hell knows, right?

Demsas: Yeah, exactly.

Miller: Had Donald Trump not run that time and he decided he wanted to do some other scam instead, then maybe Marco Rubio is the nominee and those things do come to pass.

Demsas: [Laughs.] Yeah. If Obama doesn’t make fun of him at the White House Correspondents’ Dinner, we’re not even sitting here on this podcast.

Miller: Great job, Jon Lovett, or whoever wrote that joke.

Demsas: [Laughs.]

Miller: I’m just joking.

Demsas: Yeah. Well, thank you so much, Tim. Thanks for coming on the show.

Miller: Thank you, Jerusalem.

[Music]

Demsas: Good on Paper is produced by Jinae West. It was edited by Claudine Ebeid and engineered by Erica Huang. Our theme music is composed by Rob Smierciak. Claudine Ebeid is the executive producer of Atlantic audio. 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.

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

Why are people frustrated by high levels of immigration? As refugee crises proliferate, this has become a central political question. In order to justify anti-immigration policy or rationalize restrictionist sentiment, commentators and elected officials have repeatedly returned to one hypothesis: Immigration must be bad for American workers.

There’s just one problem: This hypothesis is wrong. Economists have studied this question repeatedly, in a variety of contexts and in every segment of the population, and they have found that the demand effect consistently outweighs the supply effect. Simply put, when immigrants come to a place looking for jobs, they also demand goods and services—thus creating jobs for native-born workers. Immigrants need legal services and taxi drivers; they need groceries and cars. The question has always been which effect is bigger. And the literature has resoundingly answered that the demand effect wins out.

This doesn’t explain away all immigration worries. But it should force politicians to seriously reckon with why xenophobia exists instead of resigning themselves to treating new immigrants as an economic burden, when, for example, they were actually the “sole source of growth in the U.S. working-age population in 2021 and 2022.”

On today’s episode of Good on Paper, I’m joined by my colleague Rogé Karma who recently dove into the economics literature, originally expecting to find some negative effects on wages, only to be repeatedly struck by the truth: Anti-immigration sentiment has no economic justification.

“I think there is a lot of this deep discomfort with non-materialist explanations,” Rogé argues. “I think one of the most revealing things here is that the demographic that is most opposed to immigration are older folks living in rural areas, many of whom are retired. And the people who tend to be most supportive of immigration are working-age people living in big cities where immigrants are more common. So if you thought, Okay, this is a product of the people who immigrants are directly competing with … you would think, Oh, this would show up where the immigrants are, and it doesn’t.”

The following is a transcript of the episode:

Jerusalem Demsas: It’s Election Day, and in place of any exit-poll astrology, we’re going to talk about something that’s been a driving force in the campaign: immigration, specifically the research about the relationship between immigration and wages.

A common line bandied about in politics is that immigration reduces the wages of native-born Americans. It’s most commonly been pushed by restrictionists on the right, like J. D. Vance and Donald Trump.

J. D. Vance: And then I think you make it harder for illegal aliens to undercut the wages of American workers. A lot of people will go home if they can't work for less than minimum wage in our own country. And by the way, that will be really good for our workers who just want to earn a fair wage for doing a good day’s work.

Donald Trump: Decades of record immigration have produced lower wages and higher unemployment for our citizens, especially for African American and Latino workers.

Demsas: However, I’ve noticed a growing openness to the idea that immigration hurts American workers, not just from longtime restrictionists, but also from Democrats and liberals who are scarred by their loss in 2016 and fretting over the possibility of losing the 2024 election. But sometimes a lot of smoke is just a smoke bomb.

[Music]

This is Good on Paper, a policy show that questions what we really know about popular narratives. I’m your host Jerusalem Demsas. I’m a staff writer here at The Atlantic, and today I’ve asked my colleague Rogé Karma to come on the show. We’re going to talk about a recent deep dive he did into the economics literature on the relationship between immigration and wages.

The common thinking goes: If you increase the supply of labor, then you’ll reduce the price of that labor. If immigrants simply weren’t allowed in, then companies would be forced to pay American workers high wages. It seems so obvious, so why does study after study find this to be so wrong?

[Music]

Rogé, welcome to the show!

Rogé Karma: Thanks so much for having me. It’s great to be here.

Demsas: This is one of those episodes where I’m actually having trouble deciding which narrative is the conventional narrative.

Karma: (Laughs.) It’s because you’ve been steeped in the economic literature for far too long.

Demsas: Exactly. But there’s the conventional wisdom in academic circles that immigrants do not reduce native-born wages. But that’s not, I think, the average person’s perception of this, especially if they’re listening to politicians who, on both sides of the aisle, will be kind of making these arguments.

So I want to walk through the evidence together here because, Rogé, you recently wrote a piece, and you’ve spent a big chunk of time this year diving into the research space and really trying to figure out what’s going on. Like, Where is the evidence actually leading us? And I want to start with the Mariel Boatlift. Can you tell us what that is and then what economist David Card found when he looked into it?

Karma: Of course. And the first thing I will say is: I do think there is a little bit of a man-on-the-street, common-sense view that goes something a little bit like, Well, given that there is a fixed pool of jobs in a country, if you add a bunch of foreign-born workers, they’re going to take those jobs from natives. And if you just apply Econ 101, as the supply of a good goes up, like labor, then the price of that good, i.e. wages, falls. And so I think there is a little bit of an intuitive sense that more immigrants would mean lower wages and lower employment prospects. And I think this was actually the conventional view on both sides of the aisle for much of the 20th century, in much of the economics profession for much of the 20th century, until this study came along and shattered the consensus.

And so what happened was: In 1980, Fidel Castro lifted Cuba’s ban on emigration. And that caused about 125,000 Cubans to migrate from Mariel Bay in Cuba to Miami, Florida, and about half of them settled there, which represented a 7 to 8 percent increase in the Miami workforce, which is 25 times the amount that the workforce expands due to immigration in the U.S. every year. So this is a huge change, an incredibly large change.

And years later, what the economist David Card—who will go on to win a Nobel Prize for his work in empirical economics—what he realizes is that this created a perfect version of what economists call a “natural experiment,” that because of this big one-time influx of immigrants to Miami, you could compare the trajectory of native-born wages in Miami to a variety of other cities that prior to the boatlift had similar employment and demographic trends. These include Atlanta, Los Angeles, etcetera. And I think the view was, Look—if there’s any place you’re going to see the negative effect of immigration on wages, it’s going to be with this unprecedented, large, random shock.

And that’s why the finding that Card comes to is so surprising, because he finds that the boatlift had virtually no effect on the wages of native-born workers, including those without a college degree. If you look at a chart of the wages of workers in Miami compared to most of these other cities in the U.S. at the time, there’s almost no difference. You can’t even tell the boatlift happened. And I think what that points to—and the big, overarching explanation that I think the common-sense wisdom got wrong—is that immigrants aren’t just workers. Immigrants are also consumers. They buy lots of things, like healthcare and groceries and housing.

And so at the same time that they are competing with Americans for jobs, they’re also creating more demand for those jobs. They’re creating more employment opportunities. And when you increase the demand for labor, that pushes wages up, even if you increase the supply of labor that pushes wages down. And we can talk about some ways in which this was later challenged and complicated, but I think that’s the big missing piece of the common-sense take.

Demsas: Yeah. I think there’s a level to which you have to really draw out how this works in the real world, because people come, and they’re like, Okay. Now there’s more people who want to eat at McDonald’s. You have to hire more people on shift to service that demand. There are more people demanding taxi cabs. There are more people who now need immigration-lawyer services, so that means you need more legal assistants. That means you need more paralegals. That means you need more janitors cleaning the buildings because they’re expanding into new office space.

There’s a level to which this positive flow is not intuitive to people, because it’s so downstream of the initial event, which is: People are here looking for jobs. It’s the immediate, first thing they see happening.

Karma: Exactly. But it’s funny because when we think about this in a slightly different context, it’s very intuitive. For example, you don’t see Republican politicians going to high-school or college graduations and yelling at graduates or complaining that all of these graduates are about to undermine the wages or employment prospects of adults in the labor force.

Demsas: (Laughs.)

Karma: And that’s because we understand, when it comes to native-born people, like, Wait—population growth doesn’t necessarily mean less for everyone. And so I think when you take this to a slightly different context, it’s like, Oh, wait. This actually does make a lot of sense.

Demsas: Well, I think there’s one thing that I really want to draw out here. Because if you’re an individual person who’s—let’s say you are a high-school graduate. You are working in the types of service-sector jobs that are usually competing with immigrants. Maybe on net what you’re saying makes sense for the entire labor market, but doesn’t it change when you look downstream at the people who are the most likely to be competing with new immigrants for jobs?

Karma: This is exactly the right question. When I mentioned the complications earlier, this is where they come in. There is an argument that has come up in response to the Card paper and its response to a lot of the natural experiments on this. And I should say, also, in addition to the Mariel Boatlift study, there were similar experiments in the subsequent years in Israel, in France, in Denmark that all came to very similar results.

But then there was a backlash set of critiques, which was just this: Okay. On average, wages might not be affected, but what about the least-skilled, the least-educated workers? And, particularly, what about those without a high-school degree who work in the professions that are most likely to be competing with these new immigrants, most of whom—if we’re talking about, at least, undocumented immigrants—are less skilled themselves?

And this was the critique made, and has been long made, by a Harvard economist, George Borjas. And in 2015, he went back to the Card study, and he looked specifically at this group of high-school dropouts. And he found—or, at least, at the time, it seemed like he found—that actually there was a sizable negative effect on this smaller group. And again, this was the explanation: Okay, maybe on average it works out, but the supply-and-demand effects of immigrants are asymmetric.

Immigrants who are unskilled, who come into a country—they compete only with a certain subset of the least-skilled workers, but they’re spending their money broadly. So they might get a job as a lettuce picker or construction worker, but they’re spending their money on a lot more than just housing and lettuce. And so on net, they end up hurting these less-skilled workers more.

Demsas: It’s an inequality story too. All of us get the benefits, especially those of us in high-skilled jobs that aren’t really experiencing this competition, but they’re not concentrated for the lowest income.

Karma: And this is the most, I feel, poignant critique because, yes, this makes higher-skilled workers better off, but it hurts the least of us. And what is really interesting, though, is that Borjas’s debunking of Card has since been debunked.

Demsas: Oh, my gosh. Recursive debunkings. (Laughs.)

Karma: I know. This is all the fun of an academic debate. It has all the titillating content we want.

Demsas: (Laughs.)

Karma: So if you look at what Borjas did, what’s interesting is he didn’t just look at high-school dropouts. He also excluded from his sample women, nonprime-age workers, and, most confusingly, Hispanics, which is sort of absurd. And his justification was that these exclusions left only the workers that were most directly competing with the Marielitos. But it left a total sample of just 17 workers per year.

Demsas: I find this fact so insane. It’s one of those things where I don’t understand what the point of extremely rigorous journal processes for econ journals are if they’re allowing something like this to go by unnoticed, unflagged.

Karma: And I think the reason is because at that point, it’s really hard to tell the difference between an actual empirical finding and just statistical noise.

Demsas: I mean, it’s 17 people.

Karma: It’s 17 people. It’s this extremely specific and hard-to-justify group. And then, what’s interesting is there’s a couple follow-up studies, one of which finds that the effect that Borjas found was because of a change in the way that the census counted Black workers and Black individuals.

Demsas: Oh, yeah.

Karma: And if you just take the way that they were measuring it before, the entire effect goes away.

Demsas: Even with those 17, the sample size?

Karma: Even with those 17.

Demsas: Wow. Okay.

Karma: But I think the broader critique is, Wait. This is ridiculous. Let’s just actually do what the critique says, which is: Let’s just look at all workers who don’t have a high-school degree. And when you actually look at that, Card’s original findings hold up. Actually, workers that lack even a high-school degree didn’t have their wages negatively affected.

And in the subsequent years after this debate, there have been other natural-experiment studies that have found the same thing. One that I really think was done quite well was on Puerto Rican immigrants after Hurricane Maria who came to Orlando. It found the same effect. It actually found, to your point, that while wages for construction workers, specifically, actually did become depressed a little bit, that was offset by a boost in the wages for leisure and hospitality workers. And so, actually, one wrinkle to this story is that maybe some sectors might experience a little bit of this. But on net, it won’t affect the entire skill group. The entire skill group ends up pretty well off. And I think this, for me, was a very counterintuitive finding. And when I asked economists about it, the leading explanation is what was described to me as “specialization plus scale.”

On paper, it looks like—and the assumption has long been—immigrants without a high-school degree are perfect substitutes for native workers without a high-school degree. But it turns out that that’s actually not true. And I think the restaurant industry is a good example of this. Take fry cooks: A bunch of new immigrants come in, and they take jobs as fry cooks working in restaurants. That might depress the wages of local-born fry cooks. But what also happens is: Because the cost of labor has gone down for fry cooks, and because now there’s all this more demand for restaurant services, you get restaurants expanding. You get more restaurants opening up.

And what happens when restaurants open up? They don’t just have to hire more fry cooks. They hire more waiters and waitresses and bartenders and chefs. And it turns out that a lot of new immigrants can’t fill those roles, because they don’t have the English skills or the tacit cultural knowledge to do so. And so, actually, if you were a native-born worker and you just stayed a fry cook, you might have seen your wages depressed, but you’re actually far more likely to have gotten a job in one of these professions that is now more common, that actually pays more, because immigrants have entered.

Demsas: So you get promoted.

Karma: Exactly.

Demsas: I think the other group of people that people often point to as being harmed by this are actually recent immigrants, right? It may be the case that there’s not a substitution effect between native-born workers and foreign-born workers. But if you are the first person off the Mariel Boatlift, and then the thousandth person is coming off, you guys are probably competing.

I always find this a bit of a weird argument because people usually talking about this are immigration restrictionists. Are they taking the position of the most-recent immigrants who’ve come to this country?

Karma: Yeah. Don’t you care about all the other immigrants?

Demsas: Yeah. Yeah. What’s going on there?

Karma: That is a really good point. And I should say, just because these studies don’t find much of an effect on native-born wages of natives of all skill levels does not mean that immigration has no cost at all. And I think this is actually one of the most-important, most-consistent findings, is we do see a pretty sizable effect on the wages of other immigrants, in large part because they don’t have the substitution effect.

Another cost is inequality. A lot of these studies find that, even though a lot of lower-skilled native workers aren’t affected, immigration ends up boosting the wages of higher-skilled workers, in part because immigrants are also demanding the services of, let’s say, architects or computer programmers, etcetera. And so it’s not a huge change in inequality, but it does slightly exacerbate inequality.

And then I would say the third one is what I talked about earlier, which is if you look sector by sector. It’s very possible that a construction worker or a worker in a specific sector where a lot of immigrants come in might experience some wage losses. That is very possible. Even if the aggregate effect on an entire skill group is not negative, you can see concentrated losses.

Demsas: But this is just true of all effects, right? If the average effect is positive, 50 percent of people are below the average of all things.

I think the thing that I’m getting at here is—and one of the things I really liked about your approach to this—that you were very, very careful to try and be as fair as possible to both sides of this debate. And what I’m hearing is that there’s so much reaching you have to do to really find serious costs to immigration. Even when you do, it’s like, Slightly exacerbate inequality. Maybe there are certain industries where you have some impacts, but those people are also benefiting from the growth, and they’re also benefiting from substitution effect, etcetera.

And it’s not to just pooh-pooh all that, but I think it’s really interesting to talk about why there’s such an intense desire to find this effect. And I don’t know if you have a thought on why this narrative is so important to people, because there are other reasons that people could say they’re anti-immigration, but there’s a real desire to make it about wages. There’s a real desire to make it about economics.

Karma: Yeah. I will say: I want to definitely talk about these different reasons. And I’d be very interested in your theory, too, and I have my own. I will say, one good-faith reason for this concern, one that I think will be brought up a lot is, Well, what about when we look at history?

And so it is true, and lots of folks, including David Leonhardt, liberal columnist at The New York Times, has pointed out that during this mid-century, quote-unquote, golden-age period—1940s, ’50s, and ’60s—you saw really high wage growth for the working class. You saw a really big reduction of inequality, really fast rise in living standards, and also very low immigration. And then from the 1980s on, you see much higher levels of immigration, and you see wage stagnation for the median worker. You see an exacerbation of inequality. And so I think one thing is, If we look at history, maybe these experiments aren’t capturing everything. They’re only looking at one city at a time. And when we look at the broad sweep of American history, it really does look like this is happening.

And I think that is a critique that’s important to take seriously. But at the same time, one of the golden rules of social science is “Correlation does not equal causation,” right? There were a lot of things happening starting in the 1970s and ’80s that also affected workers, also affected inequality—everything from technological change to globalization to the weakening of labor unions and concentration of corporations. And I think a lot of those other things were going on, and I think two data points are really instructive here.

Demsas: Well, before you get into that, I actually think you’re being super generous to this argument, which I think is your MO here. I think it’s important to be intellectually generous at the front part. But I want to be very clear here: This is not looking at the broad sweep of American history. This is looking at the ’40s and ’50s and ’60s, and going, Huh. This extremely transformative time in American history, where there’s tons of growth happening because World War II is ending. Also, the World War II production, in general—lots of stuff happens, of course, following that, anti-growth stuff that we’ve talked about in this podcast in the past.

And I think it’s kind of weird and, I think, feeds into the question I was even asking you earlier about, like: There’s such an intense desire to make this true, and when you look back at the foreign-born share of the population in the United States over, actually, our long term, in 2023, 14.3 percent of Americans are foreign-born, and that’s in line—and lower—than large parts of the 19th century. So what you see in American history, when looking at the foreign-born share of the population, is: You see we’re at roughly 14.8 percent, even throughout the 1800s. You see a massive dip start to happen during the Great Depression—normal. People kind of stop emigrating when that happens. And then you don’t really see a catch-up happening until very recently.

And so there’s a level to here where I’m like, If Leonhardt and others want to make this critique, they need to then explain the entirety of the 1800s in American economic history. And I think there’s a desire not to really wade into that debate, because they’re just pointing at a simple correlation and going, I’m sure this explains it. I actually don’t find this even minimally persuasive.

Karma: I know. I think you’re totally right. And also, you don’t even have to go back to the 1800s. You can just go back, I don’t know, the past four years, where we’ve had a huge, massive surge of undocumented immigration. And at the same time, we’ve had wages at the very bottom of the income distribution rise at their fastest pace since the 1940s, a huge reduction in wage inequality.

And so even if you’re going to make the correlation argument, it’s like, Wait. The last couple of years sort of disproved this. And even over this time period that Leonhardt and others are talking about, what you have is: The places that receive the most immigrants are the places that have the least wage stagnation. It’s Texas. It’s Florida. It’s the Acela corridor. And so I think you’re right. I really wanted to put that out there because I think it is a very common argument, but it’s not one I find remotely persuasive.

Demsas: There’s one other thing that I think other folks point to a lot, and I’m going to ask you to explain it, because you’re explaining all these studies for me so nicely. But the National Academy of Sciences has a study called the “Economic and Fiscal Consequences of Immigration.” I feel like this is a calling card for a lot of folks who are pro–immigration restriction. What does it say, and what’s important about that study?

Karma: This was a large report that looked at, or at least purported to look at, a bunch of different studies that claim to be a sort of meta-analysis of a lot of the immigration literature and tried to come to a conclusion on what it all says. And the big conclusion that they came to was, when we look on average, wages are not affected, especially when we look in the long term. But there was a disagreement within the panelists over, specifically, high-school dropouts. And there’s a chart that often is linked to or is often brought up by immigration restrictionists. It is table 5-2.

Demsas: Oh my gosh, you know exactly where it is.

Karma: It shows a lot of negative numbers. And the thing that I will often remind people is: That chart is basically talking about high-school dropouts. Okay, put that aside. There are a lot of studies on there that seem to show negative effects. One of them is the Borjas study that we talked about earlier. And George Borjas was actually one of the panelists on this report, which may indicate or may give you a hint of why it turned out the way it did. But when you actually go through and look at these studies, most of them are not the kind of high-quality, natural-experiment study we’re talking about. A lot of them just focus on Black men, Hispanic men—like, very particular subgroups.

And then, also, a whole bunch of these studies are in a category of, what they’re called, “skilled-cell studies.” And these studies are different, right? They’re not looking at a specific causal link created by a natural experiment. What they’re saying is, We’re going to just look at the entire group of unskilled workers in the United States. We’re going to look at immigration flows, and then we’re going to make a bunch of assumptions, and a bunch of assumptions about the substitutability of native-born and foreign-born workers, about how fast capital adjusts. And based on those assumptions, we’re going to make big claims.

And so there was this famous other Borjas paper in the early 2000s that made the claim that when you do one of these studies, it shows really intense negative effects. And so this report that is often used, and I know this is so wonky, is just—

Demsas: We love wonky here.

Karma: Yes. But I think it is just a case study, in that listing a bunch of studies with varying qualities, looking at varying different groups, is just not the most-accurate way to do things. And then, yeah, I could go on. There are lots of other problems with it, but I just think that that is one of the ones that frustrates me the most and frustrates a lot of the economists who I spoke to for this piece.

Demsas: I asked you a question earlier, and now I’m just going to give you the answer that I have to it, which is this question about why it’s so persistent, people desire it. One of my theories for this is that there’s a real desire to sane wash anti-immigrant sentiment.

When large parts of the population hold opinions, and particularly when they are different than the kinds of people who are in media or are in elite spaces—like, most people who work in media are living in cosmopolitan cities, have gone to college, have often maybe interacted with people who are from foreign-born countries repeatedly throughout their lives because they’re, like, living in New York or Chicago or L.A. or something like that. And as a result, like, they are not really in touch with some of the more common anti-immigration sentiments, and as a result, they feel kind of uncomfortable being like, Well, they’re all racist and xenophobes. They don’t want to sound like that. And so in order to look at this sentiment in the country and go, like, Well, I don’t want to call them a bunch of people who hate immigrants, I need to find some more material explanation for their opposition to it.

And I think it’s weird here, because I actually think it’s important to take very seriously what people are saying. Like, there are people who have serious cultural concerns with people coming into the country. And some of those things, I find not reasonable, and some of them I find—I don’t find really any of them reasonable, but I understand why someone would feel that way without having to be a bad person. Like, do I wish that people didn’t have those attitudes? Sure. But I think that they’re not lying when they tell you the things that are actually concerning them. And you write something really nicely on this, in your piece, and I'll let you say it, but can you just talk to us a little bit about what surveys of public opinion actually find in regards to people’s opposition to immigration?

Karma: Well, first of all, I think that analysis is really spot on. I think sane washing is a good descriptor. The one that I haven’t had is often a veneer of respectability. And I think there is a lot of this deep discomfort with nonmaterialist explanations, in part, also, because—especially, if you’re thinking about, let’s say, center-left folks—if you’re part of a political party that you think needs to respond to people’s views on immigration, it’s much easier to say, Well, look—we already believe in raising worker wages.

And so, all of a sudden, if immigration gets looped into the set of values that we already believe, we’re then comfortable to give in to people’s instincts here. Whereas if it feels like pandering to darker forces, I think that makes liberals, especially, less comfortable with doing it. And I think to your point, though, it’s like: If we don’t acknowledge those darker forces, it’s not always great. And I think what you’re getting at, too, is in this piece, consistently what you find in sort of surveys of public opinion is that it’s not primarily material explanations that explain things. It’s things—a lot of them are about cultural difference, about violations of social norms, about crime, about national identity.

And I think one of the most-revealing things here is that the demographic that is most opposed to immigration are older folks living in rural areas, many of whom are retired. And the people who tend to be most supportive of immigration are working-age people living in big cities where immigrants are more common. So if you thought, like, Okay, this is a product of the people who immigrants are directly competing with are the most anti-immigrant, you would think, Oh, this would show up where the immigrants are, and it doesn’t. And so I think that really speaks to it.

I think the other thing that really speaks to it is, like, have you just listened to the Republican Party? Like, Donald Trump and J. D. Vance will occasionally mention wages, especially when it’s Stephen Miller talking to The New York Times, when it’s J. D. Vance in a vice-presidential debate or talking to a New York Times interviewer. That’s when they will bring up this wages argument. When they are speaking to an audience that they know is very center or left, they will, like, bring up this wages argument. But if you listen to the guy at the top of the ticket, right, it is, These folks are “poisoning the blood of our country.” It is portraying immigrants as a sort of psychopathic horde of murderers. It’s spreading conspiracy theories about pet-eating immigrants. Like, it’s very hard to take seriously that this is actually the main concern when the leader of the party who is anti-immigration is, like, so openly pointing to a very different set of issues.

[Music]

Demsas: After the break: why less immigration would mean a poorer quality of life in America.

[Break]

Demsas: I think one of the things that I also find reasonable for why people struggle is the Econ 101 explanation you give right at the beginning is intuitive. And it shows up in many different parts of our econ reporting, this question of thinking outside of partial equilibriums, right? If you were to just say, I’m looking at just the impact of immigrants on wages, holding equal all other effects on the population, on the economy, then you would see negative downward pressure on wages.

But economists don’t do that. Our lives aren’t lived in partial equilibriums. We live in general equilibrium. There are multiple different markets working together at all times. And you have to look at not just what’s happening with the effect on the labor markets, what’s happening in the effect on the housing market, the consumption of random household goods. And so I think that that’s really difficult to do in normal conversation. And what I think is really funny is that now—you kind of foreshadowed this—there’s a big push to blame immigrants for the housing market, and it’s like, Oh, the only time immigrants are consumers of anything is when they’re consuming housing. Otherwise they’re just competing with you for wages. They’re not buying anything else.

What I wanted to ask, though, is about this other argument that people also make, which is: Okay—maybe you’re right that in the world that we live in, given that immigration is always happening, companies can rely on there going to be some level of immigration. They’re used to a high level of immigration happening. You don’t see these negative effects on wages. But in a world where you were to just, like, really, really tamp down—really, really stop immigration from coming in—companies would have to reshape how they’re doing their hiring practices. The entire American economy would change if it wasn’t reliant on foreign labor.

And so this idea that there are these, quote, jobs Americans won’t do isn’t true. They would do them at a price. They would do them if the wages were better, if the working conditions were better, and that we should strive for these higher-quality jobs. And companies that can’t do that, well—they should just not do that anymore. They should just literally stop relying on foreign labor. And so I think that how you respond to that is really important because, you know, I do think a lot of people are starting to, like, fixate on that argument.

Karma: This has actually been one of the largest justifications for what Donald Trump has called the largest deportation effort in American history that he wants to enact when he’s in office. Any of his advisors are talking about this. They talk about: This is going to force employers to hire workers at higher wages, to give them better jobs, and that’s a big reason why we should do it. So I think it’s a really important one to address.

And what’s nice is we actually have some really good empirical studies on this. We don’t just have to guess as to what would happen and assume as to what would happen. My favorite of these studies, although there are a few, looked at the Secure Communities program, which was a DHS program that deported about 500,000 immigrants between 2008 and 2014, so during the Obama administration. And the way that this happened was: It happened sort of semi-randomly across communities, such that it created a sort of natural experiment where you could look at how it affected communities where it had happened and how it affected not-yet-affected communities.

And the findings were shocking even to me because I would think, Okay, maybe when you get rid of a lot of these workers, there’s just going to be more jobs available.

Demsas: It’s like a shock.

Karma: It’s such a big, immediate shock. But what the authors find is that for every 100 immigrant workers who are deported, there are actually nine fewer jobs for natives. And this isn’t just temporary jobs. This is, like, permanently, there are fewer jobs for natives in the community, unemployment goes up, and wages slightly fall. And I think this kind of finding is repeated across different examples through American history.

There’s another great study of the H2B program, which allocates low-skilled workers to companies, and also finds that when companies aren’t allocated those workers, they don’t hire a bunch of natives. They actually just produce less. And so what happens when immigrants are ripped away from these communities is the interconnected web of employment and workers whose jobs depended on each other all gets torn up, right? Businesses close. Businesses have to stop producing as much. There are just less child-care services. There are less meals served. There are less houses built—either for reasons of: Employers actually can’t have a viable business with higher labor costs, whether it’s because natives don’t always want to do these jobs, or whether it’s just because, for the reasons we talked about earlier, there’s just a lot of benefits from the specialization of labor that occur when immigrants are in a place.

One way I think about this is sort of the opposite of the story that we were talking about earlier. When we talked about immigrants coming in and creating the specialization plus scale, that just happens in reverse. Instead of businesses expanding, and therefore being able to hire more natives because they’re expanding, businesses are shrinking. They’re shutting down. They’re closing. And when that happens, native-born workers get caught in the crossfire. When there’s less demand for your services as a restaurant, and your costs are higher, and so you have to close down, you’re not just getting rid of your immigrant workers, you’re getting rid of all the native-born workers who are working there too. And so I think that’s what these studies are finding, is you just can’t neatly remove immigrants from communities without having huge backfiring effects on the native-born.

Demsas: Yeah. I think it’s even useful to play it out in the best-case scenario. I think the best-case scenario for the folks who are making this argument is that there’s this short-term harm, but then you just need to let the economy play out and develop new businesses to figure out new business models that work. And in the best-case scenario, you’re talking about a poorer country. You’re talking about a country where your output, your growth is literally less. And that sounds very abstract, but we’re talking about less stuff. You have less money. You can buy fewer things. You can buy a worse quality of life. Your housing is probably worse. Even the basic stuff where you’re talking about, Can you afford child care?—fewer people can do that.

A lot of things are worse when the output declines, when it’s harder for businesses to try new things, when there’s difficulty with dynamism in the economy, where you can’t start a bunch of different kinds of businesses quickly, see what works, and have that kind of churn. And so I think it’s even difficult to conceptualize: When people are making this argument, they’re saying, We should take on the costs of being a poorer country for the sake of national homogeneity of where you were born.

And so I think that that’s the trade-off we’re talking about here. It’s not that America would cease to exist, right? There are a lot of countries who follow the sort of principles we’re talking about here, where they are really strict on who gets to come in, and they’re just poorer than America. And I think that that’s the really clear trade-off that I think often restrictionists won’t make baldly.

Karma: First of all, even the best-case scenario you’re talking about is one that has actually no empirical evidence. It’s all theoretical. So that’s the first point. The second is that this gets back to, I think, your point about general versus partial equilibriums, too. Because when we’re even just looking at these wage or employment studies, they’re holding a lot constant. And everything they’re holding constant also changes in an actual scenario where you deport millions of immigrants.

So there is another great study from the economist Ben Jones and a few others looking at immigrants and entrepreneurship. And they looked at basically every single business that opened up between 2005 and 2010 and looked at basically the country of origin of the person who started that business. And they found that immigrants are 80 percent more likely to start new businesses than native-born individuals.

And when they actually did the math, they found that immigrants, by entrepreneurship alone, are creating far more jobs than they take. One response to that would be, Oh, okay. Well, maybe this is high-skilled immigrants. Maybe this is the Google-founder kind of effect. But actually, they found that there was no difference in the rate of entrepreneurship between individuals from OECD countries and from non-OECD countries. And if you just think about this for a second, think about the people who end up coming here, the amount of risk they have to take, the perseverance that it takes to, like, actually get to the U.S.—it wouldn’t be surprising that these people are, like, more intrepid and more entrepreneurial.

Demsas: Just huge selection effects. Like, if you can make it through the Darién Gap, what does that say about you?

Karma: Exactly. And so that’s one effect that is completely lost in a lot of these studies. One of my favorite studies of this is one that was done in Denmark, because in Denmark, what is interesting—unlike in the U.S., where you have to just look at a specific city—Denmark has data on individuals for the entire country. It’s a pretty small country. And so researchers can actually track what happens to every single individual worker when new immigrants come in. And that gives you a sort of accuracy that you don’t necessarily get with the natural-experiment studies in the U.S., at least at a countrywide basis. And what they found is that native-born—even less-skilled native-born—workers end up responding to immigration by entering higher-paying occupations, by moving to higher-opportunity cities, and by actually getting better education, such that they actually had higher wages as a result of it.

And so I could go on. You could talk about the amount of women who are able to be in the workforce because of immigrants providing child care. Like, you can list this out, and there are all these ways in which even these studies are missing the sort of beneficial effect that immigrants are having that you would be taking away if you just suddenly got rid of all these people, in addition to this atrocious humanitarian effects.

Demsas: I find that the Danes—like, I wanted some sort of poll on their privacy concerns. I’m just like, Do you guys not care? I mean, like, I think it’s great. I would be pro-this everywhere, but I’m just surprised that countries are able to do this. There would be a revolt in America.

Karma: Even if the data is anonymized, I’m like, The data the researchers got was anonymized, but the data the government has is not anonymized.

Demsas: We don’t even let the government share data like that. Like, the IRS can’t just send the Treasury Department, like, all the data they have on people’s tax returns.

Karma: But you know what? You know what? Great for the Danish for doing it, too, so we can learn more about immigration through them.

Demsas: So true. But so the thing that’s interesting is: We’ve made a bunch of arguments here about why this is actually really positive for the economy. But regardless of that, there’s been a backlash, and we’re seeing that right now. I mean, this is airing on Election Day, and so we’re, I’m sure, in the future, just pacing nervously to see what’s going on.

Karma: (Laughs.) Apologies to anyone who was listening to this looking for a soothing distraction.

Demsas: But this has been probably the most-important issue of this election. Maybe inflation is another one. But the two most-important issues. I did an episode earlier this year with John Burn-Murdoch where we talked about the sense that Americans are very xenophobic and this narrative that they hate immigration—they hate immigrants—and that’s, like, just a fact of the world, and that all immigration has basically been this plot by elites to shove it down our throats. And, of course, we explored how a lot of that narrative is really overblown and underestimates just how strong pro-immigrant sentiment is in America, particularly relative to other countries.

And I still stand by that analysis, but there has been a shift in public opinion, even in the past year. You’ve seen polls come out that have really indicated that there’s been a backlash effect to the high levels of immigration that are kind of returning us to the 1800s averages.

And so first, can you just talk us through that backlash? What are the numbers there? What are we seeing?

Karma: Totally. The thing that first drew my attention to this was, as you were saying: The way this has impacted the election is that you’ve just seen such a hard right turn in the rhetoric from candidates on both sides. And I remember listening to Trump in 2024, making 2016 Trump sound like JFK in just how crazy he was. And then looking at the Democratic side, where the message went from, in 2020, Joe Biden promising to restore moral dignity to our asylum system, and then in 2024 Kamala Harris saying that, Actually, no. She is the one that will fortify the border, not Trump.

Demsas: Do not come.

Karma: Do not come. And underlying this is quite possibly the most dramatic shift in public opinion that I’ve ever seen. So going back to the 1960s, Gallup asks Americans every year this question: Do you think immigration should be increased, kept the same, or decreased? In 2020, only 28 percent of Americans said that immigration should be decreased. Actually, more Americans said it should be increased. By 2024, just four years later, the percentage of those who wanted it decreased had nearly doubled to 55 percent, the first time that there had been a majority of Americans who wanted immigration decreased since the early 2000s.

And just to put this shift in context, I think when Americans think about big public-opinion shifts, they think of gay marriage. And they think of the increasing support for gay marriage. Support for gay marriage, according to Gallup, increased about 20 points over the course of around a decade, maybe eight or nine years. This shift we’re talking about was nearly 30 points in four years. It makes gay marriage look gradual and small by comparison. And this immigration shift is most concentrated among Republicans, but it’s also Democrats. It’s also independents. And it’s especially been sharp in the past year.

Demsas: This is one of those things where I think it’s really important for people who, like myself, are in favor of high levels of immigration, first of all, to accept that, at some level, if you get that, you will have some negative effects, but I think also to really narrow in onto people’s specific concerns.

So I did an article earlier this year. It’s called “Something’s Fishy About the ‘Migrant Crisis.’” And basically, I was just like, Okay, there are high levels of immigration in a lot of places in this country. But not every place in this country is experiencing backlash, right? Like, you’re hearing these stories in New York and Chicago about people sleeping on the floors of police stations in Chicago. In New York, I had, like, an affordable-housing lawyer tell me—she was a very liberal person telling me that she was kind of concerned because there were migrants in the street in midtown Manhattan who were just, like, lying on the ground.

And there’s a lot to which I’m like, You know, if these people who are talking to me are some of the most-liberal people on immigration are expressing kind of like, Well, we can’t handle this. Like, we obviously can’t handle this, it indicates a very specific problem, right? Like, people—these New Yorkers, Chicagoans—they’re not afraid of immigrants or foreign-born people. There’s huge levels of foreign-born share in New York and Chicago. And the number of people that were entering we’re talking about, you know, that were coming in and demanding services from local government were a very small fraction of this.

So I was trying to understand what was going on there, and I really came down to the specific concerns people have. People don’t want to see local tax dollars being spent on newcomers to their city, if they feel like they need things that the city’s not actually taking care of. They don’t want to see their schools being used as shelters instead of being used in order to service, you know, their kids. And there’s just kind of general sense that, like, Now there are people sleeping on the streets. There’s nowhere for them to be housed. Like, It’s actually reducing my quality of life a bit. Clearly, there’s a sense of it being overwhelmed.

But then when I looked in places like Miami and Los Angeles, and in Texas and in Houston—Miami and Houston, in particular—I was like, There are way more immigrants who have come through a Houston, in Texas, than have come through a Chicago. So why are we seeing such backlash?

And I came to like two reasons. One is that many of the immigrants were not able to get work permits. The other thing that’s really important here is the Greg Abbott busing program. Because, most people, they come into the country, and, you know, what happens? They have networks that they’re following. Like, either they have populations of people that they’re able to get help from, or there are even people who are kind of recruiting them as they’re kind of coming over, like, Oh, we need work. We need people to come do this. And so there’s a level to which, like, there’s a natural flow to where they end up.

Greg Abbott has, I think, maybe the most-effective political stunt in American history—I genuinely think, like, reshaped the entire conversation on immigration by doing this. And then he says, Okay, I’m gonna bus people—effectively breaking these kind of natural shifts—to Democratic cities. And when people kind of show up randomly, there’s, like, of course, this massive transaction cost that’s enacted. And, you know, Texas is a border state, and I think, at some level, I kind of understand. They’re like, Oh, everyone should have to experience what it’s like to have all this kind of flow of people coming in. But Texas has boomed as a result of this too.

So anyway, I think the real thing that’s important here is that people who are in favor of immigration have to address these specific concerns. You have to make sure that there is, like, actually a clear, orderly process by which people are being resettled. If there’s not, I mean, that’s going to lead to backlash, even from people who are in favor of immigration. And the most frustrating part of my reporting is learning the Biden administration had basically abdicated their responsibility to try and help with the resettlement process of people across the country, because they were afraid of being blamed. And to me, I’m like, Well, you were still blamed. So I’m not sure it worked out for you.

Karma: Everything you’re describing here, I think, falls under the banner of what has been called either, colloquially, “chaos theory” or, more academically, the “locus of control theory” of immigration, which is that populations tend to be able to handle high amounts of immigration if they think the process is orderly and fair, and they become much more likely to oppose immigration when they see the process is chaotic and unfair and disorderly.

You know, you have a great example in that piece of the U.K., post-Brexit, having very high levels of immigration and anti-immigrant sentiment decreasing. Something you see in the U.S. is that even as you have these massive shifts in the amount of immigrants people want let in, when you ask questions like, Do you believe undocumented immigrants make a contribution to society? or, Do you support a path to citizenship for nondocumented immigrants? and even, Should it be easier to immigrate to the U.S.? people’s views actually haven’t changed nearly as much. And they remain more pro-immigrant than they were in 2016, which speaks to the fact that what people are upset about here—they’re not suddenly xenophobic. They don’t suddenly hate immigrants. A lot of what’s happening is that they’re responding to the chaos of the process.

I think my favorite part of that piece that you wrote was this point that you made about how there’s a way of looking at Greg Abbott’s busing program as working, in the sense of, like, Look—didn’t this prove his point? He said that liberal cities should have to handle this, and he proved that they couldn’t. But the process by which he did it was engineered to achieve that outcome, right? You point out in that piece that there are 3 million foreign-born people in New York City, in a city of 8 million, and they’ve been bused a couple tens of thousands. And it has led Eric Adams to say, like, New York is falling apart. That does not mean that New York can’t handle that amount of immigrants. The specific process by which this happened was engineered to achieve an outcome of chaos. And that’s what people are responding to.

Demsas: Well, let’s leave it on an optimistic note. Always our last question: What is something that you originally thought was good on paper but didn’t pan out in real life?

Karma: So this is quite a pivot from what we were talking about earlier. Last year I got engaged to my girlfriend.

Demsas: Oh my gosh, yes! Congratulations!

Karma: That was not the thing that was—

Demsas: Oh god.

Karma: That would be bad. But the way I did it was: It was our five-year anniversary, in Rome, very romantic. I knew I wanted to propose in front of the Pantheon, which was my partner’s favorite building in Rome.

Demsas: She’s an architect.

Karma: She’s an architect, yes. But I didn’t want to do it when there were a bunch of crowds around, so I was like, How can I figure out a way to get us there, like, early in the morning? And so I decided, in a decision that looked very good on paper, to book a Vatican tour for, like, 9 a.m. And so I was like, Oh, I’ll propose, and then we’ll go on this tour of the Vatican, and it’ll be, like, really cool. And it’ll be, like—we’ll see the Sistine Chapel. Sounded great. Looked good on paper.

It turns out that immediately after you have one of the most emotionally riveting experiences of your life—

Demsas: (Laughs.) You don’t want to go on a tour!

Karma: The last thing you want to do is go on a three-hour tour of the Vatican where you have to wait until the last 15 minutes to see the Sistine Chapel. We’re just, like, so badly just wanting to get out of there and be with each other, and we were just in such a great mood, only to have, like, the biggest buzzkill in the world be the Vatican.

Demsas: This is so funny, Rogé. I didn’t know the story. Wait. I can’t believe—so you went on the tour?

Karma: We went on the tour. I wish so badly I would have said, Let’s just forget the tour. But we were in such good spirits after. We’re like, This is going to be so great. Like, actually, looking back, I’m like, Was that even good on paper? I don’t think so.

Demsas: I was actually with you. I was like, Okay, yeah. Then you had a fun tour.

Karma: Like, a nice walking tour, architecture tour—probably great. When you’re, like, confined to the Vatican and just looking at, like, our guide—she was great, but she was just explaining every little thing, and we’re just like, We don’t want to be in public with a million people. We just want to be with each other. This is very strange.

Demsas: I think this is my favorite “good on paper” yet. This is unreal. (Laughs.)

Karma: I put a lot of thought into this, and I was like, This one was bad. This is not my best. It’s a funny story now.

Demsas: Yeah.

Karma: You know, I look back on it—I’ve looked back on it very fondly. So yeah, that’s my “good on paper” story.

Demsas: Thank you so much, Rogé. Thanks for coming on the show.

Karma: It’s been a pleasure being here. Thanks so much for having me.

[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. 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.