Over the past few weeks, Donald Trump has positioned himself as an unabashed bull on America’s need to dominate AI. Yet the president has also tied this newfound and futuristic priority to a more traditional mission of his: to go big with fossil fuels. A true AI revolution will need “double the energy” that America produces today, Trump said in a recent address to the World Economic Forum, days after declaring a national energy emergency. And he noted a few ways to supply that power: “We have more coal than anybody. We also have more oil and gas than anybody.”

When the executives of AI companies talk about their ambitions, they tend to shy away from the environmental albatross of fossil fuels, pointing instead to renewable and nuclear energy as the power sources of the future for their data centers. But many of those executives, including OpenAI’s Sam Altman and Microsoft’s Satya Nadella, have also expressed concern that America could run out of the energy needed to sustain AI’s rapid development. An electricity shortage for AI chips, Elon Musk predicted last March, would arrive this year.

Both Trump and the oil and gas industry—which donated tens of millions of dollars to his presidential campaign—seem to have recognized an opportunity in the panic. The American Petroleum Institute has repeatedly stressed that natural gas will be crucial in powering the AI revolution. Now the doors are open. The oil giants Chevron and Exxon have both declared plans to build natural-gas-powered facilities connected directly to data centers. Major utilities are planning large fossil-fuel build-outs in part to meet the forecasted electricity demands of data centers. Meta is planning to build a massive data center in Louisiana for which Entergy, a major utility, will construct three new gas-powered turbines. Both the $500 billion Stargate AI-infrastructure venture and Musk’s AI supercomputer reportedly already or will rely on some fossil fuels.

If one takes the dire warnings of an energy apocalypse at face value, there’s a fair logic to drawing from the nation’s existing sources, at least in the near term, to build a more sustainable, AI-powered future. The problem, though, is that the U.S. is not actually in an energy crunch. “It is not a crisis,” Jonathan Koomey, an expert on energy and digital technology who has extensively studied data centers, recently told me. “There is no explosive electricity demand at the national level.” The evidence is ambiguous about a pending, AI-driven energy shortage, offering plenty of reason to believe that America would be fine without a major expansion in oil, coal, or natural-gas production—the latter of which the U.S. is already the world’s biggest exporter of. Rather than necessitating a fossil-fuel build-out, AI seems more to be a convenient excuse for Trump to pursue one. (The White House and its Office for Science and Technology Policy did not respond to requests for comment.)

Certainly, data centers will drive up U.S. energy consumption over the next few years. An analysis conducted by the Lawrence Berkeley National Laboratory (LBNL) and published by the Department of Energy in December found that data centers’ energy demand doubled from 2017 to 2023, ultimately accounting for 4.4 percent of nationwide electricity consumption—a number that could rise to somewhere between 6.7 and 12 percent by 2028. Some parts of the country will be affected more than others. Northern Virginia has the highest concentration of data centers in the world, and the state is facing “the largest growth in power demand since the years following World War II,” Aaron Ruby, a spokesperson for Dominion Energy, Virginia’s largest utility, told me. Georgia Power, similarly, is forecasting significant demand growth, likely driven by data-center development. In the meantime, Microsoft, Google, and Meta are all rapidly building out power-hungry data centers.

But as Koomey, who co-authored the LBNL forecast, argued, that forecasted growth does not seem likely to push the nation’s electricity demands past some precipice. Overall U.S. electricity consumption grew by 2 percent in 2024, according to federal data, and the Energy Information Administration predicted similar growth for the following two years. A good chunk of that growth has nothing to do with AI, but is the result of national efforts to electrify transportation, heating, and various industrial operations—factors that, in their own right, will continue to substantially increase the country’s electricity consumption. Even then, the U.S. produced more energy than it consumed every year from 2019 to 2023, as well as for all but one month for which there is data in 2024. An EIA outlook published last month expects natural-gas-fired electricity use to decline through 2026. John Larsen, who leads research into U.S. energy systems and climate policy at the Rhodium Group, analyzed the EIA’s power-plant data and found that 90 percent of all planned electric-capacity additions through 2028 will be from renewables or storage—and that the remaining additions, from natural gas, will be built at two-thirds the rate they have been over the past decade.

None of this discounts the fact that the AI industry is rapidly expanding. The near-term electricity-demand growth is likely real and “a little surprising,” Eric Masanet, a sustainability researcher at UC Santa Barbara and another co-author of the LBNL forecast, told me. More people are using AI products, tech companies are building more data centers to serve their customers, and more powerful bots may also need more power. Last year, Rene Haas, the CEO of Arm Holdings, which designs semiconductors, attracted much attention for his prediction that data centers around the world may use more electricity than the entire country of India by 2030. Some regional utilities have projected much higher demand growth into the late 2030s than nationwide estimates suggest. And chatbots or not, building enough electricity generation and power lines for transportation, heating, and industry in the coming years will be a challenge.

Still, tremendous uncertainty exists around just how power-hungry the AI industry will be in the long term. State utilities, for instance, are likely exaggerating demand, according to a recent analysis from the Institute for Energy Economics and Financial Analysis. That might be because utilities are overestimating the number of proposed data centers that will actually be built in their territories, according to a new Bipartisan Policy Center report that Koomey co-authored. And AI still could not turn out to be as world-changing and money-making as its makers want everyone to believe. Even if it does, the energy costs are not straightforward. Last month, the success of DeepSeek—an AI model from a Chinese start-up that matched top American models for lower costs—suggested that AI can be developed with lower resource demands, although DeepSeek’s cost and energy efficiency are still being debated. “It’s really not a good idea” to look beyond the next two to three years, Masanet said. “The uncertainties are just so large that, frankly, it’s kind of a futile exercise.”

If AI and data centers drive sustained, explosive electricity demand, natural gas and coal need not be the energy sources of choice. For now, utilities are likely planning to use some fossil fuels to meet short-term demand, because these facilities are more familiar and much quicker to integrate into the grid than renewable sources, Larsen told me. Plus, natural-gas turbines can operate around the clock and be ramped up to meet surges in demand, unlike solar and wind. But clean energy will also meet much of that short-term demand, if for no reasons other than cost and inertia: Solar panels, wind turbines, and batteries are becoming cost-competitive with natural gas and getting cheaper, while a growing number of industries are turning to renewable energy sources. The tech firms leading the AI race are major purchasers of and investors in clean energy, and many of these companies have also made substantial investments in nuclear power.

Using natural gas, coal, or oil to power the way to an AI future will not be the inevitable result of the physics, chemistry, or economics of electricity generation so much as a decision driven by politics and profit. AI proponents and energy companies “have an incentive to argue there’s going to be explosive demand,” Koomey told me. Tech firms benefit from the perception that they are building something so awe-inspiring and expensive that they need every possible source of energy they can get. Any federal blessing for data-center construction, as Trump granted Stargate, is a boon to production. Meanwhile, oil and gas companies want to sell more energy; utilities earn higher profits the more they spend on infrastructure; and the Republican Party, Trump included, has a pretense to satisfy demand to ramp up fossil-fuel production.

Of course, AI needn’t precipitate a national energy shortage to add to a different crisis. Microsoft and Google, despite promising to significantly reduce and offset their carbon footprints, both emit more greenhouse gases across their operations than they did a few years ago. Google’s emissions grew 48 percent from 2019 to 2023, the most recent year for which there is public data, and Microsoft’s are up 29 percent since 2020, an increase driven substantially by data centers. These companies want more power, and the fossil-fuel industry wants to supply it. While AI’s energy needs remain uncertain, the environmental damages of fossil-fuel extraction do not.

In the early 1500s, an unknown wealthy patron is said to have commissioned Leonardo da Vinci to produce the Salvator Mundi, a striking ecclesiastical masterpiece in which Jesus is shown blessing humanity with his right hand while holding an orb representing the Earth in his left. The patron’s identity has been lost to history, and whether da Vinci actually painted it is still debated among scholars, but such commissions were common during the medieval and Renaissance periods: Medici-like benefactors, uncomfortable with the potential sinfulness of their extravagant wealth, sought to offset their guilt and enhance their prestige by sponsoring magnificent works of art and architecture for the public to enjoy.

Da Vinci’s Salvator Mundi changed hands countless times through the ensuing centuries. Mistaken as a comparatively commonplace artwork, it was owned by a 17th-century heir to the Scottish crown who was later beheaded, passed to the illegitimate son of an 18th-century duke, and then languished in obscurity for more than a century. An unknown buyer acquired the painting at auction for £45, or about $1,300 today, and it ended up in Houston. The painting later passed to Basil Clovis Hendry Sr., who ran a Baton Rouge, Louisiana, sheet-metal company. Then, in 2005, on suspicion that there was more to the painting than met the untrained eye, an art consortium bought the painting for just over $1,000. Years of restoration, cleaning, research, and speculation yielded a shocking announcement: The painting was Da Vinci’s lost Salvator Mundi.

What happened to the tableau after that is a good illustration of just how little today’s superrich resemble the public-spirited patrons of the past. Yves Bouvier, an art dealer who is currently accused of evading $800 million in taxes, bought the work for $83 million, then sold it the following day for nearly $50 million more to Dmitry Rybolovlev, a Russian oligarch and superyacht enthusiast who, according to the leaked Panama Papers, set up a shell company to hide artwork assets from his ex-wife during divorce proceedings. Finally, in a 2017 auction at Christie’s, Saudi Arabian Crown Prince Mohammed bin Salman used a little-known proxy to purchase the Salvator Mundi for $450.3 million, the highest price paid for a single artwork in history.

In 2021, The Wall Street Journal broke the news that the priceless painting had been kept on private display aboard bin Salman’s superyacht, Serene, a 439-foot-long, half-billion-dollar boat that had recently run aground in a navigational accident. A fragile, irreplaceable object of significance to the shared cultural history of all humanity was being kept in a hot, humid environment for the private enjoyment of one royal billionaire and his ultrarich guests. (In another room, Serene was also equipped with state-of-the-art snow machines that could produce four-inch-deep flurries on demand.)

The journey of one painting charts a profound shift in modern societies. The role of the ultra-wealthy has morphed from one of shared social responsibility and patronage to the freewheeling celebration of selfish opulence. Rather than investing in their society—say, by giving alms to the poor, or funding Caravaggios and cathedrals—many of today’s plutocrats use their wealth to escape to private islands, private Beyoncé concerts, and, above all, extremely private superyachts. One top Miami-based “yacht consultant” has dubbed itself Medici Yachts. The namesake recalls public patronage and social responsibility, but the consultant’s motto is more fitting for an era of indulgent billionaires: “Let us manage your boat. For you is only to smile and make memories.”

In 1908, the English writer G. K. Chesterton observed that “the poor man really has a stake in the country. The rich man hasn’t; he can go away to New Guinea in a yacht. The poor have sometimes objected to being governed badly; the rich have always objected to being governed at all.”

Chesterton’s observation was astute for the modern era, but for much of Western history, it was only half true. In his 2023 book, As Gods Among Men, the economic historian Guido Alfani chronicles the role of the ultra-wealthy from antiquity to the era of cryptocurrency. The superrich have always wielded inordinate economic and social power and, as such, have plenty of historical ills to answer for. But the affluent of many past periods also invested in the shared betterment of society, understanding that doing so helped justify the existence of wealth inequality. Today’s ultrarich, by Alfani’s telling, are uniquely selfish, and by abdicating the philanthropic role, they are “fuelling resentment and leaving their place in society uncertain.”

The social contract that imposed certain civic responsibilities on the rich emerged after the Black Death decimated nearly half the population of medieval Europe. The dominant Christian theology of the Middle Ages held that wealth was inherently sinful in a world where most people toiled in terrible poverty. But as the continent recovered from the plague, a new, pragmatic arrangement emerged. The surviving wealthy would be expected to use their wealth to provide public goods. This echoed the norms of antiquity; the historian Paul Veyne has noted that in ancient Rome, for example, the belief was widespread that any defects in civic life directly reflected on the virtues of the city’s elite.

[Read: Cancel billionaires]

The barons of medieval society would serve two important functions, Alfani recounts: “making the city splendid in everybody’s interest by means of their ‘magnificence’ and acting as private reserves of financial resources into which the community could tap in times of crisis by means of taxation or of extraordinary contributions.” In other words, wealth inequality was tolerated because it provided a useful social function. The wealthy were expected to spend lavish sums on transforming cities by building shared public spaces. They were also meant to come to the rescue with their largesse in the case of a public crisis or calamity. Cosimo de’ Medici did precisely that, saving Florence from bankruptcy in 1434.

Benefactors did not necessarily serve these functions out of uncomplicated generosity. From their “magnificence” they could expect personal glory, political favor, and perhaps a pathway to eternal salvation. The savvy used patronage to expand, not drain, their wealth. Some patrons participated in history’s great crimes, from the Crusades to the slave trade. Nonetheless, as Alfani convincingly argues, even the most self-interested and amoral among them often wound up doing some good.

The 15th-century archbishop of Florence called this norm the “public theology of magnificence.” But it also required enforcement. When the rich refused to fulfill their social obligations, governments imposed taxes, extraordinary war levies, or, in the case of 16th- and 17th-century Spain, mandatory loans (called secuestros). The wealthy were not allowed to simply hoard their wealth, park it in an offshore haven, and escape catastrophe by sailing away from a collapsing society on a superyacht.

Even in Gilded Age America, with all its injustices, and where the pursuit of wealth was hardly condemned as sin, society’s richest members were expected to use their riches to benefit the public in times of crisis. J. P. Morgan bailed the United States out in 1907, acting as a banker of last resort. A decade later, the U.S. government pressed financiers and tycoons to invest in Liberty Bonds for World War I, a bad bet that worsened their financial positions considerably. During World War II, the top American marginal tax rate reached an eye-popping 94 percent.

Over time, however, the norms eroded. An ethos of what historians call “munificence,” a belief that the rich should be generous, but only if they wish, replaced the theology of obligatory magnificence. This subtle difference had profound implications: Patronage and public benefit were no longer assumed to be duties, but bonuses that wealthy individuals could provide out of the goodness of their hearts. The coronavirus pandemic ushered in even more grotesque inequality. Elon Musk’s net worth surpassed $400 billion. The world’s economies ground to a halt, and public coffers were crushed with debt, but superyacht sales surged by 46 percent. The public, especially the poor, suffered; the rich, like those of G. K. Chesterton’s caricature, escaped.

During the early coronavirus lockdowns, the billionaire media mogul David Geffen hunkered down on his 454-foot-long superyacht, Rising Sun, which included a private basketball court and wine cellar among its 82 rooms. He posted a stunning photo of a Caribbean sunset to Instagram at the height of the pandemic, with the caption “Isolated in the Grenadines avoiding the virus. I hope everybody is staying safe.”

[Read: A yacht owner’s worst nightmare]

What people spend their money on, beyond hard-nosed investments, sends a social signal. For example, the drivers of Priuses and Cybertrucks are sending rather different signals, as their visible purchases likely reflect different underlying behaviors and beliefs. At the extremes, the wealthy may go to great lengths to display their affluence. For example, in potlatch ceremonies among Indigenous communities of the Pacific Northwest, individuals showcase their riches by engaging in the ritualized destruction of expensive objects.

In The Patron’s Payoff, Jonathan K. Nelson and Richard J. Zeckhauser argue that historical patronage was a form of pragmatic signaling, conveying virtues such as religious devotion and civic duty. But it also provided benefits to the wealthy, who became part of an elite club and were able to use their patronage for personal glory and social advancement. Today the signaling of wealth has shifted from public-facing duties to efforts to provoke private envy.

When the Saudi crown prince pays half a billion dollars for an invaluable artwork and then displays it for ultrarich elites on his private superyacht, he is engaging in a form of signaling, but not one aimed at the public. Instead, the “haves” and “have yachts” play a status game only for the benefit of the rich themselves. Cathedrals were beautifying public icons that often served the poor; yachts are designed to hide their splendors from the prying eyes of the riffraff. One of the great wealthy villains of modernity, Martin Shkreli, didn’t just buy a coveted piece of cultural heritage—the notorious Wu-Tang Clan album—for his own private consumption. He said he would destroy it, potlatch-style. This kind of signaling is a far cry from the one that centered on civic virtue and religious devotion.

The U.S. government has facilitated the ultrarich in their abdication of social responsibility. For example, Charles B. Johnson—the former CEO of Franklin Templeton Investments, a Republican megadonor, and a part owner of the San Francisco Giants—purchased the opulent Carolands chateau, a 46,000-square-foot Gilded Age estate with 98 rooms. According to a ProPublica investigation, Johnson received a $38 million tax break because he pledged to turn Carolands into a museum open to the public 40 hours a week. He didn’t keep that promise—Carolands allows small tours only on Wednesdays at 1 p.m.—but he got the tax break all the same.

Some among the American ultrarich openly deride philanthropy as an ineffective use of money. The tech mogul Marc Andreessen has argued that charitable giving is less useful than investing in tech companies, because “technological innovation in a market system is inherently philanthropic.” (One in 12 people globally lives in extreme poverty, defined as earning less than $2.15 a day; it’s hard to imagine how they have benefited more from Ning, Andreessen’s social-media platform, than they would from, say, food and medicine.) The billionaires Larry Page and Peter Thiel have expressed similar views. Thiel concentrates his philanthropy on what he designates to be breakthrough technology. He has donated to the Seasteading Institute, which says it is “reimagining civilization with floating communities” and “significant political autonomy”—as though the superyachts and offshore tax havens aren’t enough. Why not live offshore, bobbing in a libertarian commune free from burdensome social obligations, such as taxes?

[Read: Space billionaires, please read the room]

Some billionaires have maintained the notion of magnificence by pouring money into solutions to social problems, such as treatments for malaria or children’s hospitals. The Gates Foundation, for example, has made tremendous progress against scourges of public health. But by and large, the notion that wealthy individuals will marshal their resources to alleviate social crises has come to seem quaint in today’s world. In 2008, President Barack Obama proposed that the income of private-equity-fund managers be treated as ordinary personal income rather than capital gains—and ran afoul of the billionaire Stephen Schwarzman, who later compared the president’s proposed tax-policy changes to “when Hitler invaded Poland in 1939.”

Evading social responsibility, even during crises, carries risks for the ultrawealthy. Their opulence compared with the standard of living of their co-citizens becomes harder to justify, and widespread resentment, seemingly inevitable. Some appear to understand that such inequality is unsustainable, but that hasn’t inspired them to become keepers of social wealth for times of shared crisis. Rather, if society collapses, billionaires may just escape onto the waves. A select few are making their contingency plans in rocket ships—as though no longer seeking favor with God but hoping to abscond to the heavens just the same.

Project 2025, the conservative governing playbook produced by veterans of the first Trump administration, has an entire chapter on how to overhaul USAID. Its authors urged the next president to “scale back USAID’s global footprint,” “deradicalize” its programs, and throttle its funding.

Before the election, Donald Trump disavowed Project 2025 because it veered so far to the right. But now he’s making the plan look downright timid. Project 2025 did not call for freezing all foreign aid or locking USAID employees out of their headquarters. Nor did the treatise suggest shutting down the $40 billion agency and subsuming it into the State Department—all without a single vote in Congress.

As the chair of Trump’s quasi-official Department of Government Efficiency, Elon Musk has razed USAID with shocking speed. He’s called it “evil,” “a radical-left political psy op,” and “a criminal organization.” The rampage seemed to come out of nowhere, but the 64-year-old agency has long been one of the government’s most vulnerable conservative targets.

[Read: Why Trump can’t banish the weirdos]

Although foreign aid accounts for less than 1 percent of the federal budget, right-wing politicians began attacking it well before Trump. In the 1990s, the late Senator Jesse Helms of North Carolina likened the disbursement of American money abroad to shoving taxpayer dollars “down a rathole.” Conservatives have even tried to abolish USAID—most notably Helms in the late ’90s and early 2000s. But the scope of those attempts pales in comparison to what Trump and Musk are doing now, George Ingram, a former USAID official in the Clinton administration, told me. “This,” he said, “is fundamentally different.”

At Musk’s urging, the Trump administration has placed nearly all USAID employees on administrative leave and recalled thousands from overseas postings with virtually no notice. (At the same time, the president declared that the U.S. would “take over” the Gaza Strip—a mission that would presumably require a sizable American deployment.) Trump designated Secretary of State Marco Rubio as USAID’s acting administrator. In one of his first moves, Rubio wrote to senior members of Congress—not to ask for their help in reforming the agency but merely to notify them that the government might reorganize it.

“It’s ridiculous,” Andrew Natsios, a former USAID administrator, told me. He ran the organization for the first five years of the George W. Bush administration and describes himself as “the most conservative administrator in the history of the agency.” Natsios has his share of problems with USAID, including his sense that its staff is often unresponsive to political leadership, a critique that Project 2025 echoes. But Natsios, who’s now a professor at Texas A&M University, is aghast at the Trump administration’s purge of USAID. (He began our conversation by comparing it to the Russian Revolution.) For days, he’s been fielding calls from panicked contacts at the agency. “They are not reviewing each project,” he said. “They’re eliminating entire bureaus, whole programs, simply deleting them without even looking at what they’re doing.”

USAID was created in 1961 to consolidate programs that had grown out of the Marshall Plan, said Ingram, who is now a senior fellow at the Brookings Institution. Congress considered putting USAID in the State Department but kept it separate so that it could operate more nimbly—like a business, Ingram told me, rather than a bureaucracy.

Presidents of both parties have supported foreign aid, including Ronald Reagan and the second Bush, who weren’t enthusiastic about it as candidates. “Once they got into office, they saw that it was a very important tool of U.S. foreign policy,” Ingram said. Even one of the Project 2025 authors acknowledged that foreign aid has helped America check global adversaries; a former USAID deputy administrator, Max Primorac, credited the agency with countering China’s Belt and Road Initiative. Indeed, authoritarian regimes have long denounced American aid, and now some of them are praising Musk’s efforts. Musk himself promoted a laudatory post on X from a top aide to Hungarian Prime Minister Viktor Orbán. With that adulation in mind, Natsios questioned whether Musk’s campaign against USAID might be “motivated by his desire to please the Kremlin.”

[Read: America can’t just unpause USAID]

Sending taxpayer funds abroad has never been particularly popular, a reality that Trump seized on during his first term by attacking foreign aid as part of his “America First” agenda. In 2017, administration officials reportedly drafted proposals to merge USAID with the State Department, but they never went anywhere. Polling has found that Americans dramatically overestimate the amount of money the government spends on foreign aid, and in a survey released this week, most respondents backed cuts to foreign aid. Natsios faulted the Biden administration for making USAID an even more inviting target for Trump 2.0 by trying to export progressive values such as LGBTQ and abortion rights, especially to countries where they are unpopular. “They brought part of this on,” he said.

By and large, Republican lawmakers have simply watched as Musk and his allies shut down an agency that, according to a paper published on Monday by the nonpartisan Congressional Research Service, cannot be abolished, moved, or consolidated without authorization from Congress. A few have issued mild protests. Senator Bill Cassidy of Louisiana criticized the pause on distributing HIV/AIDS drugs through the President’s Emergency Plan for AIDS Relief, a George W. Bush–era program that enjoys wide bipartisan support domestically and internationally. “It is a Republican initiative, it is pro-life, pro-America and the most popular U.S. program in Africa,” Cassidy wrote on X. “This must be reversed immediately!!”

Representative Michael McCaul of Texas, who until last month served as chair of the House Foreign Affairs Committee, said the Trump administration was “right to scrutinize and revamp” USAID, but he strongly defended its purpose and urged the president to eventually resume sending aid abroad. “U.S. foreign-assistance programs not only feed starving women and children in some of the most destitute parts of the world, but they also promote democracy, help stabilize fragile nations on the brink of collapse, and counter our adversaries’ attempts to shift the global balance of power,” McCaul told me.

By contrast, McCaul’s successor atop the committee, Representative Brian Mast of Florida, cheered the administration unreservedly and released a four-minute video “exposing radical, far-left grants” supposedly issued during the Biden administration. His list included $15 million for “condoms for the Taliban,” money to expand “atheism in Nepal,” and various line items promoting LGBTQ rights. (The contraceptives were for Afghan citizens, not members of the Taliban; the Nepal grant promoted religious freedom.)

When I asked Natsios, a lifelong Republican, what he made of the response from GOP lawmakers, he scoffed: “The Republican Party in Congress is a disgrace.”

[Listen: Purge now, pay later]

Advocates for USAID now have little choice but to place their hopes in Rubio, who as a senator defended foreign assistance as “critical to our national security.” In his new role, however, he has characterized USAID as a rogue agency whose leaders misspent taxpayer money and refused to cooperate with Trump’s directives during his first few days in office. “There are a lot of functions of USAID that are going to continue,” Rubio told reporters in El Salvador on Monday. “But it has to be aligned with American foreign policy.”

Natsios used to enthusiastically support Rubio. He told me that he once saw Rubio give “the strongest speech for foreign aid” he had ever heard. He contributed to Rubio’s presidential campaign in 2016—when Rubio was a GOP rival to Trump—and said the then-senator had told him that, had he won, he would have brought him into the White House. Now, Natsios told me, Rubio has a choice to make: “He is going to accept the ideology” of Trump and Musk, “or he is going to get fired.”

While Rubio and other Republicans decide whether, and how much, to fight for U.S. foreign aid, the ripple effects of the firings and funding freeze at USAID are quickly growing. Many policy decisions in Washington take weeks or even months to be felt overseas. Not this one, Ingram said. The moves threaten the jobs of thousands of people connected to the aid industry inside the U.S., and they jeopardize the livelihood of potentially hundreds of thousands of people—or more—in the developing world, who rely on USAID for health care, food, fertilizer, and other crucial supplies. Ingram was stunned: “I have never seen a government action have such an immediate impact.”

Ron Johnson may be the most anti-vaccine lawmaker in Congress; he’s the kind of guy who says he’s “sticking up for people who choose not to get vaccinated” while claiming without valid evidence that thousands have died from COVID shots. This morning, at the Capitol, Johnson walked over to his Senate Finance Committee colleague Bill Cassidy, a doctor and a passionate advocate for vaccination, and gave him an affectionate pat on the shoulder. The two of them had just advanced Robert F. Kennedy Jr.’s nomination to lead the Department of Health and Human Services to the Senate floor.

The committee vote, which was held this morning in a room crammed to capacity with what appeared to be roughly equal numbers of Kennedy’s skeptics and devotees, certainly fit with the behavior of a compliant GOP. But it was still surprising in its way, if only because, until this morning, Cassidy had been so clearly wary of giving the nation’s highest role in public health to a prominent anti-vaccine activist. At last week’s confirmation hearings, he seemed like he might even be prepared to cast his vote with the opposition. That didn’t happen.

Whether you like Kennedy or not, the hearings showed that he lacks the basic qualifications to hold this office. He knows very little about the nearly $2 trillion behemoth that he would be tasked with running. He flubbed the basics of programs such as Medicare and Medicaid, and seemed wholly unaware of an important law that governs emergency abortions. The hearings also called attention to a passel of health-related conspiracy theories that RFK Jr. has floated in the past, including that Lyme disease was developed as a bioweapon, that COVID is “ethnically targeted” to infect Caucasians and Black people (and spare Ashkenazi Jews and Chinese people), and that standard childhood vaccinations are damaging or deadly.

As of last Thursday, Kennedy appeared to have unwavering support from the committee’s Republicans, who occupy 14 of its 27 seats—with one notable exception: Cassidy. Prior to taking office, the Louisiana senator had personally led a campaign to vaccinate 36,000 kids against hepatitis B. In an interview with Fox News last month, he said that RFK Jr. is “wrong” about vaccines. And in early 2021, Cassidy joined six other GOP senators in voting to convict Donald Trump on charges of “incitement of insurrection.” The doctor had voted his conscience before. It seemed possible that he would do so once again.

Cassidy made no attempt to hide his skepticism of RFK Jr. during Thursday’s hearing. He spoke up at one point to correct the record after his Republican colleague Rand Paul worked up the crowd of pro-Kennedy spectators by disparaging the practice of vaccinating babies for hep B. Later on, he paused to cite a meta-analysis disproving Kennedy’s often-stated belief that childhood vaccines may be a cause of autism. (Cassidy also explained the concept of a meta-analysis for those in the room and people watching at home.) When RFK Jr. cited his own evidence for being skeptical of vaccines, referring to a paper from a little-known journal, Cassidy put on his reading glasses, peered at his iPad, and reviewed the evidence firsthand. At the end of the hearing, he reported that he’d found “some issues” with the paper, and then implored Kennedy to disavow mistruths about vaccine safety. “As a patriotic American, I want President Trump’s policies to succeed in making America and Americans more secure, more prosperous, healthier. But if there’s someone that is not vaccinated because of policies or attitudes you bring to the department, and there’s another 18-year-old who dies of a vaccine-preventable disease [...] It’ll be blown up in the press,” he warned. “So that’s my dilemma, man.”

Cassidy’s “dilemma” hardly went unnoticed by RFK Jr.’s supporters. Calley Means, a proponent of Kennedy’s Make America Healthy Again campaign, said last weekend on The Charlie Kirk Show that MAHA moms are now “camping out at [Cassidy’s] office.” (I did not see any tents or sleeping bags outside his door this morning.) Other MAHA leaders, including the anti-vaccine activist Del Bigtree, have also issued political threats to any lawmakers who might try to stop Kennedy’s confirmation. “Anyone that votes in that direction, I think, is really burying themselves,” Bigtree told me and a group of other reporters last week.

Cassidy, for his part, wasn’t saying much about his personal deliberations. His only official social-media post from the weekend quoted a Bible verse from the Book of Joshua: “Do not be afraid; do not be discouraged,” it read in part. “Be strong and courageous.”

When he arrived at the committee room this morning, Cassidy was somber. He stared straight ahead, his brow furrowed. He’d been verbose at last week’s hearings, but now he said only a single word—“aye”—and left the room. In a social-media post that went up this morning, Cassidy explained that he’d received “serious commitments” from the Trump administration that made him comfortable with voting yes. Speaking later on the Senate floor, he added that RFK Jr. had promised to “meet or speak” with him multiple times a month, that the Trump administration would not remove assurances from the CDC’s website that vaccines do not cause autism, and that the administration would give his committee notice before making any changes to the nation’s existing vaccine-safety-monitoring systems. “It’s been a long, intense process, but I’ve assessed it as I would assess a patient as a physician,” Cassidy said. “Ultimately, restoring trust in our public-health institution is too important, and I think Senator Kennedy can help get that done.”

Even if Cassidy had voted no, his vote may not have mattered in the end. Under normal circumstances, a nomination that got voted down by the Senate Finance Committee would be dead in the water—but these were not normal circumstances. Majority Leader John Thune could still have scheduled a vote by the full Senate, at which point Kennedy would have been kept from office only if at least three other Republicans had joined Cassidy in opposition.

It’s still not a sure thing that Kennedy will be confirmed by the full Senate. Other Republicans, including Senator Mitch McConnell of Kentucky, have raised concerns about Kennedy’s anti-vaccine activism. But the odds of RFK Jr.’s defeat are shrinking, and Cassidy’s thumbs-up may one day be remembered as the mirror image of John McCain’s thumbs-down from 2017, when that independent-minded senator doomed Trump’s efforts to repeal the Affordable Care Act. Faced with an opportunity to make the same sort of stand, Cassidy folded. Now the American public is at the whims of the administration’s promises.

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

There are two Chicago neighborhoods that are, on the surface, quite similar. They are both more than 90 percent Black; the median age of both is roughly 38. About the same share of people have college degrees, and the median income of both is roughly $39,000.

But one experiences about twice as many shootings per capita as the other.

The University of Chicago economist Jens Ludwig opens his forthcoming book, Unforgiving Places, by describing the neighboring places of Greater Grand Crossing and South Shore, both minutes away from the elite university where he teaches. Ludwig’s argument begins by reframing the problem of gun violence away from the demoralizing story of American exceptionalism and toward the more granular variation that differs state by state, city by city, and yes, block by block.

“Whatever you believe about the causes of gun violence in America, those beliefs almost surely fail to explain why Greater Grand Crossing would be so much more of a violent place than South Shore,” Ludwig writes. “How, in a city and a country where guns are everywhere, does gun violence occur so unevenly—even across such short distances, in this case literally right across the street?”

Talking about gun crime almost always turns into talking about gun-control legislation, a debate that has been happening my entire life and I’m sure will continue past my death. But on today’s episode of Good on Paper, Ludwig and I spend little time on that topic, focusing instead on policy levers that could reduce gun violence but don’t require national gun-control legislation.

The following is a transcript of the episode:

Jerusalem Demsas: In 2022, Louisiana had the second-highest rate of gun deaths in the country. I’m just back from a reporting trip to the Lake Charles area, and I had a few people remark rather pointedly to me that my home of Washington, D.C., is a violent place, seemingly unaware that D.C. has had a significantly lower rate of gun deaths than Louisiana for many years now.

Why do some places see higher rates of gun violence than others? It’s an incredibly important question to answer rigorously. Homicide is a leading cause of death for young adults, and the vast majority of those homicides happen with guns. But this is a topic where the politics rarely line up with actionable solutions.

After the COVID-19 crime wave, politicians have scrambled as they place crime at the top of the agenda again and are searching for public-policy tools to address violence in their communities.

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. My guest today is the economist Jens Ludwig, from the University of Chicago, who has spent his career studying the economics of crime. He has a book coming out in a few months called Unforgiving Places: The Unexpected Origins of Gun Violence.

Jens and I talk about the classic explanations for why gun violence happens in some places and not others. He pushes back against the classic right-wing explanation that the problem is bad people and the classic left-wing argument that solving the problem of gun violence requires ending mass social inequalities first.

One note about the show: We’re going to begin adding the studies and articles and books we reference in the show notes, so you can easily access them for further reading. A link to Jens’ book will be there, too, if you’d like to investigate his argument further.

Okay. Jens, welcome to the show.

[Music]

Jens Ludwig: Thanks so much for having me. It’s such an honor to be here.

Demsas: Jens, you have a book coming out in April called Unforgiving Places. What’s it about? What are you arguing?

Ludwig: The book basically makes two arguments. One argument is that we’re despairing about the problem of gun violence because we’ve thought about it as just all being about gun control, and I think that’s not true. I think the problem of gun violence in America is partly about guns, and it’s partly about violent behavior. And if we can’t do anything about the guns, we can at least try and do something about the violent behavior. And the experiences of L.A. and New York over the last 30 years show us that there’s real progress that you can make there.

And then I think the other core argument of the book is that violent behavior is not what we’ve thought. I think most people have thought of violent behavior in America as being about thoughtful, deliberate action that leads you to focus on incentives, like bigger sticks or more enticing carrots. And in fact, I think most shootings in America are instead fast-thinking, reactive—it stems from arguments. And that leads us away from relying exclusively on incentives and towards a very different type of policy that we just haven’t been talking about or thinking about.

Demsas: When I was reading your book, there was a stat that just has been rattling around in my brain since I read it. You write that shootings account for fewer than 1 percent of all crimes but nearly 70 percent of the total social harm of crime. What does that mean? And how is that even measured?

Ludwig: Yeah. So the way that economists think about that sort of thing is very analogous to how environmental economists think about environmental harm. If you go back to the Exxon Valdez oil spill in Alaska a million years ago, there’s the tangible cost of cleaning up the bay or whatever it is, and then there’s the sort of social costs that don’t show up on any sort of budget spreadsheet anywhere. That’s the “harm to this pristine place now being ruined forever” kind of thing.

And so environmental economists have come up with ways of quantifying those sorts of intangible costs. And we can use the same sort of approach to measure the harm for crime as well. It basically comes down to what people are willing to pay to avoid exposure to different types of crime.

And so what you can see is people really don’t like disorder. They really don’t like having their bicycle stolen, their car stolen. I lived in cities for the last 30 years. I’ve had almost every sort of property crime that you can imagine happen to me. But the thing that people really, really are petrified about is staring down the barrel of a gun. And I can tell you that from firsthand experience. I was held up at gunpoint myself on the South Side of Chicago, going to pick up my older daughter from her piano lesson about five years ago.

My University of Chicago colleague Steve Levitt did a study where he showed that every serious crime that happens in a city reduces the city’s population on net by one person—so fewer people moving in, more people moving out. Every murder that happens in a city—the overwhelming majority of murders in the United States, unfortunately, are committed with guns—every murder that happens in a city reduces the city’s population by 70 people. And I think that’s another way to sort of see exactly how much the gun-violence problem in America is driving the crime problem.

Demsas: I also think it’s just remarkable to really think about this in perspective of how much effort we spend in trying to eliminate certain types of crime. I mean, if 70 percent of total social harm is shootings, then the vast majority of our efforts should just be focused on guns. And property crime should take a backseat, all this sort of thing. Intuitively, we understand that, obviously, murder is worse than other forms of crime, but I think the degree to which that is driving America’s violence problem and crime problem and the harms that ricochet out into communities is, I think, not well understood.

Ludwig: Yeah, I one million percent agree. And I think it also sort of helps you see a path to a criminal-justice system, a law-enforcement system that kind of sidesteps a lot of the current political fights that we’re having. I think everybody agrees that gun violence is a hugely serious problem, that we should be holding people accountable for this.

Even the mayor of Chicago, who I think within the political distribution is one of the more progressive elected leaders in the United States—he’s going around talking about the need to improve the odds that shooters get arrested and wind up behind bars. And so I think this much stronger focus on gun violence would be a way to concentrate everything on the thing that the American public really cares the most about. It sidesteps a lot of the fraught political debates about how we do enforcement over lots of other things that the public doesn’t like, but it’s not the first-order thing that they’re worried about.

Demsas: So there’s familiar pattern that I think most people are aware of when it comes to the gun-policy conversation in the United States, and it’s: There is a tragic mass shooting—maybe at a school, maybe at a nightclub—and then there’s this intense rallying to pass gun legislation.

And economists have quantified this. There’s a study that showed that a mass shooting leads to a 15 percent increase in the number of firearm bills introduced within a state the year following that shooting. Interestingly, in states with Republican-controlled legislatures, those are often laws that loosen gun restrictions. But even when looking at Democrat-controlled legislatures and laws that tighten gun restrictions, studies often struggle to find significant impact of these laws on reducing gun violence, reducing deaths, reducing mass shootings.

In your book, you also seem kind of pessimistic about the potential for gun legislation to have a large impact on reducing gun deaths. Why is that?

Ludwig: Yeah. Let me respond in two ways. The first is: Federal gun laws set a floor, not a ceiling, on what cities and states can do. And so lots of cities and states around the country, including my home city of Chicago, have enacted gun laws that are more restrictive than what you have under the national law. And the problem with that is that we live in a country with open city and state borders. So what Gary, Indiana, is doing about air quality affects the South Side of Chicago, and vice versa, right?

And in the same way, like, my family for the last 18 years has lived in Hyde Park, on the South Side of Chicago. Our favorite ice-cream place in the area is Dairy Belle in Hammond, Indiana. So we spend 20 minutes driving down there every summer, like, way too often. And when we come back from Indiana into Chicago, nobody stops us at the city border to check what we have in our trunk.

And when you look at where the crime guns are coming from in Chicago, almost none of them come from a gun store in Chicago. They come from places like, you know—there are gun stores quite close to Dairy Belle in Indiana that are big sources of crime guns in the city. So I think the way that you want to be thinking about gun regulation, I think, is very analogous to how you would do something like regulate air quality. And that’s to think about regulation at the national level in a world in which you’ve got what an economist would call lots of externalities across jurisdictions in their own laws.

Demsas: But even federal gun-control legislation has often felt, at least from my overview of the economics literature, like it hasn’t had a massive impact, whether it’s assault-weapons legislation or other forms of gun-control legislation that’s passed over the past few decades. Is that just a reflection of the fact that these laws are pretty modest in what they’re attempting to do? Or does that indicate that we can’t really attack this problem legislatively?

Ludwig: What I would say is: Most of the national gun laws that we’ve enacted in the United States are very modest, as you said. I think the biggest problem with the gun laws that we have in the United States is: Most of the laws regulating gun acquisition—you know, gun sales—only apply to gun sales that are, basically, carried out by a licensed gun dealer.

And that’s something like 50 or 60 percent of all gun sales in the U.S. And the other 40 percent are almost completely unregulated under federal law. Some states try and regulate that, but that’s not a loophole—that’s like a chasm that you can drive a truck through. And you know, when you look at where the guns used in crime come from, you wouldn’t be surprised to see that’s the most important source of crime guns that you see in Chicago and other cities around the country.

But you know, I think the difficulty of cities and states regulating their way out of the gun-violence problem, and the difficulty of substantially changing national gun laws, has led a lot of people to conclude that gun violence in America is a hopeless problem, because we can see that the gun-control politics are stuck.

So one way that I’ve come to think about this is that that’s too pessimistic a view. And the reason for that is that gun violence is not just about guns; it’s about guns plus violence. So it’s having lots of guns around, but also having people who use them to hurt other people. And if we can’t make much progress on the gun-access part of things, the good news is that there’s a second path to progress, which is to try and change the willingness of people to use guns to hurt other people.

We have something like 400 million guns in the United States, in a country of about 330 million people. And I think the existence proof that shows us that you really can make a huge difference on the gun-violence problem by figuring out how to control violence comes from the Los Angeles and New York City experience over the last 30 years.

So in 1991, the murder rate per 100,000 people in L.A. and New York was very similar to Chicago, actually, at that time. It was something like 30 per 100,000. So to give you a sense of what that means: In London, the murder rate is something like one or two per 100,000. So the United States is just totally off the charts. Almost all of those extra murders here are committed with firearms.

And in the 30-year period following that—so 1991 (the peak of the crack-cocaine epidemic), 30 years after that, up through 2019 (the last year before the pandemic)—the murder rate in Los Angeles declined by 80 percent; the murder rate in New York City declined by 90 percent. And those are cities that are swimming in the ocean of, you know, hundreds of millions of guns in America. And I think that speaks to a more optimistic take, that it is not a hopeless problem—not just that something can be done but that something substantial can be done.

Demsas: The other variation you point to in your book that is what really intrigued me is that Canada and Switzerland also have above-average rates of gun ownership, but they don’t have particularly high rates of murder in line with what we would expect if you just took America’s experience. And I think I had this kind of model in my head that it’s just like, If you have this many guns, there’s nothing you can do. Like, that’s the situation. There will be variations based on other things, like whether the economy is doing well or whether we’re incarcerating people or not, or how many cops there are on the street and what they’re doing. You’d still see variations in crime, but you would always have some kind of baseline level of criminality.

But I want to get to the core argument of your book, which I think is maybe encapsulated by a pretty provocative question on the back cover, which says, “What if everything we understood about gun violence was wrong?” This is a very bold claim, and I’m excited to explore it with you. But I think that the first part of that is unpacking what it is that you mean by “everything we understand about gun violence.” You lay out two competing theories that Americans hold about the causes of gun violence. One is the “root causes theory” and one is the “wickedness theory.” Can you just walk us through what those two are?

Ludwig: Yeah, the conventional wisdom in America right now says that violent behavior is thought through, right? So it’s either bad people who aren’t afraid of whatever the criminal-justice system is going to do to them, or it’s people in bad economic conditions who are desperate in doing whatever they need to do to survive. And both of those conventional wisdoms on the right and the left actually have something in common, which is: They think of gun violence as being sort of a deliberate behavior, and that leads us then to focus on incentives to solve the problem. You know, We need bigger sticks, if you’re on one side of the aisle, or if you’re on the other side of the aisle, We need more enticing carrots.

I think the thing that’s so striking is that it just doesn’t fit with what all of the data tell us gun violence in the United States is. Most shootings are not premeditated, and most shootings are not motivated by economic considerations. They’re not robbery. They’re not drug-selling turf. That’s all what psychologists would call “System 2” slow thinking.

Most shootings, instead, stem from arguments. They’re reactive, or what psychologists would call “System 1” thinking. And the fact that so many shootings stem from these sorts of in-the-moment conflicts that go sideways and end in a tragedy because someone’s got a gun, that helps explain why deterrence is imperfect. Someone acting very reactively is not thinking through a jail sentence. And it also helps explain why a social program that’s intended to reduce poverty—like give somebody a job, give somebody cash, whatever—that also isn’t solving the violence problem.

Demsas: I want to hold here a bit because I think this question, Are people making rational calculations? is both at the heart of a lot of economics and also the heart of what we’re going to talk about for the rest of this episode. And I accept that I do not think that I or anyone else is constantly doing a benefit-cost analysis about every action that I take, even if it is as important as whether you pull out a gun and shoot someone.

But I wonder whether that undersells the rationality that still exists, right? Because we know that deterrence is possible. We know that when we increase the certainty of capture—if you know you’re going to get caught for shoplifting, if you know that you’re going to go to jail if you shoot someone—that significantly decreases crime incidents. And what that indicates to me is that there is a level of benefit-cost analysis happening, even if people aren’t fully using that System 2 part of their brain.

Ludwig: Yeah, I one hundred percent agree that deterrence is really a thing. I’m a card-carrying economist. I work at the University of Chicago. I totally believe that incentives matter and that deterrence is a thing. But I think that this really connects very importantly to where we started, that gun violence is the part of the crime problem that is the thing that drives the total social cost of crime.

So in many ways, crime is an unhelpfully broad term. It’s almost like disease. What would you do about disease? I mean, I don’t even know how to think about answering that. Like What are we talking about? Like, pneumonia or cancer? And crime is a similarly unhelpful, super-broad umbrella.

And there was a study, for instance, done in Sweden a few years ago where they looked at what happened when you put cameras up in the subway system. And what you could see is that property crimes go down when you camera-up the trains, but violent crime doesn’t go down, right? And I think what that tells you, partly, is that different behaviors are shaped differently.

The key breakthrough of behavioral economics and behavioral science over the last couple of decades is to realize that our minds work in two different sorts of ways. There’s the deliberate, sort of rational benefit-cost calculation that psychologists call System 2, and a sort of very reactive, automatic, below-the-level-of-consciousness cognition that psychologists call System 1—or fast thinking and slow thinking.

And different behaviors are driven by different types of cognition. And so stealing a loaf of bread to feed your family is much more System 2 than what you do in an argument. Let me just point the finger at myself, first and foremost here. I’m not saying anything about other people’s behavior that is not true of my own behavior.

I’ve lived for 18 years in Hyde Park. It’s a little University of Chicago village in the middle of the South Side of Chicago. Every Wednesday morning, I take my dog, Aiko, out for a little walk. One day, I’m walking down the street, and about three or four doors down from me, there’s a neighbor whose dog is off leash, runs down the driveway, and attacks my dog.

Demas: Oh God. I hate that.

Ludwig: No, exactly. And this guy, the neighbor—his kids are literally in the same classroom as mine at the lab school. He lives four doors down from me. I have every incentive in the world to handle that gracefully and constructively. And that’s exactly what System 2 rational thinking would have done.

It turns out: That is exactly not what I did in that case. I assume this is a podcast where people don’t curse, but you can only imagine the stream of four-letter, seven-letter, and twelve-letter words that came out of my mouth at this guy who I’m going to be seeing for years into the future. I’m going to be seeing him at the parent potluck at school.

And so it really speaks to this idea of: In these super high-stakes moments, where people just don’t have very much bandwidth and they are relying on sort of very fast thinking to navigate, we are not always our best selves. We are not thinking about benefits and costs and things off into the future. We can make mistakes. All of us can make mistakes.

And in my case in Hyde Park, I was very lucky that neither one of us had a gun. But in a country with 400 million guns, you know, lots of people are in situations like that and behave the way I did and, unfortunately, they or the other person’s got a gun, and it ends in tragedy. And those tragedies, really, I would just point out, claim two lives. Somebody does something stupid in a moment and, you know, you spend the rest of your life in prison, and somebody else winds up dead. It’s multiple tragedies stemming from that.

Demsas: First, is your dog okay? Was everything fine?

Ludwig: Yeah, she’s a big chicken. She’s, like, a 70-pound shepherd mix who decided, rather than to try and defend herself or whatever, she would—I don’t want to throw my dog under the bus here. Everything turned out fine. She’s a lover, not a fighter. (Laughs.)

Demsas: (Laughs.) Your dog also is in System 1 thinking.

Ludwig: Yeah, exactly.

Demsas: Well, first, we’ll shout out the late Danny Kahneman here and his Thinking, Fast and Slow book, which provides much of the foundation of the System 1, System 2 model that you’re talking about here.

But I want to push here a bit because I think one of the common objections people have to this line of argument is that, yes, it is the case that, whether someone’s coming at you or you’re worried about your dog, and you don’t react the in the way that you might if you used your logical brain to react if you had time to think—but given that if you place every single American in the exact same conditions, you still see large variations in how people choose to respond, right? Like, all the people who are in conflicts in the South Side of Chicago do not shoot each other. A very small minority of people are choosing to shoot each other, even if they have access to a gun.

And so doesn’t that push against this idea that the problem is this System 1 thinking? Like, there is something particular about the choice to pull out a gun and kill someone in that moment. And it’s not just, Well, anyone can make that mistake, because even if you think about this demographically, we’re seeing mostly young men make this mistake and make this choice. There is something going on here that is not just, You’re not able to think under stress.

Ludwig: Let me take your question and sort of turn it on its head for a second. One of the things that I point out in the book is like a version of an observation that Jane Jacobs made 60 years ago in her book The Death and Life of Great American Cities, which is: When you look at similarly poor neighborhoods in American cities, you see huge variation in crime rates, especially violent crime.

And as I mentioned, I lived for a long time in Hyde Park, on the South Side of Chicago. There are two neighborhoods just south of Hyde Park. There’s Greater Grand Crossing and South Shore that are socio-demographically, historically almost identical in terms of their racial and ethnic composition, their socioeconomic composition. They’re adjacent neighborhoods, so they’ve got exactly the same gun laws; they’ve got exactly the same social policies. When people get caught, they get sent to exactly the same court system. So all the incentives that conventional wisdom would say would matter are identical. And yet the shooting rate per 100,000 is, in most years, about twice as high in Greater Grand Crossing than literally across Dorchester Avenue in South Shore.

Demsas: Wow.

Ludwig: So that’s sort of taking the premise of your question and noting that the incentive explanation certainly doesn’t explain all of the variation that you see in gun violence either.

So what could it be then? I one million percent agree with you that—at its core, the argument here is: People are people, and a lot of what determines the outcome of this interpersonal conflict is the situation that someone finds themselves in. But if it’s not socioeconomics, and it’s not the characteristics of the criminal-justice system, what else would it be?

And I think in many ways, Jane Jacobs was really onto something 60 years ago in thinking about what that thing would be. To sort of connect an experience that I had in Chicago a couple years ago to Jane Jacobs’ insight, I was in the juvenile-detention center on the West Side of Chicago, I’m talking to a staff leader there, and he says, I tell all the kids in here, “If I could give you back just 10 minutes of your lives, none of you would be here.”

And one of the insights that Jane Jacobs had 60 years ago is: If the problem here is people do things in these 10-minute windows that they later regret, you could almost sort of think of fraught social interactions as like a high-wire act. And one of the ways that you can help people is by—what do they do in the circus for high-wire performers? They have a safety net there.

And one of the safety nets that you have much more of in some neighborhoods than others is essentially what Jane Jacobs called “eyes on the street”—prosocial adults who are around and able to step in and deconflict things when it happens. And you could see exactly that when you look at South Shore versus Greater Grand Crossing.

So there is, for instance, much more commercial development in South Shore than in Greater Grand Crossing. And what that means, in practice, is that there’s just lots more foot traffic in the community in South Shore than Greater Grand Crossing. And so if a group of teenagers is getting into an argument, there’s more likely to be, like, a neighborhood adult around to step in.

It’s also the case—so my friends Sendhil Mullainathan and Eldar Shafir have a wonderful book that came out a couple of years ago, called Scarcity, where they point out that one of the many challenges of being poor in the United States is living in day-to-day circumstances that tax mental bandwidth. It’s just very stressful, right? And people with limited bandwidth wind up relying much more on System 1 than people who are less bandwidth taxed.

So when you look at the data, you can see all sorts of indicators that there’s much more stress and bandwidth tax for people living in Greater Grand Crossing than South Shore. And what that would lead you to conclude is that the people who are in Greater Grand Crossing are going to be more likely when they’re in these difficult, 10-minute, fraught interactions with somebody else to rely on System 1 to navigate that than their more deliberate, rational benefit-cost-calculating selves.

So I think the sort of left-of-center perspective that there are root causes that matter is definitely right. I think it’s totally right for property crime—you know, crimes shaped by economic considerations. I think it’s just a little bit incomplete with respect to the part of the crime problem that the public cares the most about, which is gun violence. And so I think we just need to expand our lens about what aspects of the social environment we want to be prioritizing for our public policies.

Demsas: I’m a housing person, so I’m a big fan of the Jane Jacobs book and the argument that she kind of draws out, and I think people can imagine this if they’ve been in streets and communities like this before, is when you have kind of mixed-use development—you have a coffee shop, and above that coffee shop, you have apartments, and across the street, there’s also a park, and there’s also a school nearby—is that that means that throughout the day, there are many different kinds of people watching the streets.

Versus if you had just a fully residential area, and then during the day, everyone’s basically gone because they’re either at school or work, so it really empties out of people to watch things. Or if you have a fully industrial area, where when people go home for the day, there’s nobody there. Or commercial area, same thing. And so when you have these kinds of mixed-use-development areas, it feels a lot safer because you can just always feel like there’s someone around doing normal business or taking their kids to school or whatever.

So I would love for housing policy to be the key. But is your argument, then, that the differences between neighborhoods that have similar socioeconomic problems, similar legal environments, etcetera but a large variation in gun violence is largely a function of their urban form?

Ludwig: I just—I absolutely adore that this is a sort of empirical, data-intensive, data-nerd podcast, and so in that spirit, I do think one of the big challenges for making progress on the sort of the crime and criminal-justice problem is: A lot of it is editorializing rather than guided by data. And so I think one of the key things that I tried to do in the book is really stick to the data and see what the data are telling us.

And so does the built environment matter? There was a wonderful study by Mireille Jacobson and Tom Chang that looks at what happens in Los Angeles when marijuana dispensaries open or close as a result of some regulatory change and when food places open and close.

That’s like the natural experiment of Jane Jacobs, like, let’s put in more mixed use—and what you can see is that when a retail establishment closes and foot traffic goes down, crime goes up.

There was a wonderful study by a great team at the University of Pennsylvania that worked with the City of Philadelphia to do a randomized experiment where they picked a bunch of rundown, vacant lots all over the city and picked half of them to redevelop and turn into little pocket parks. And what you can see is that the pocket parks then wind up bringing more people out of their homes and spending time there in public. And you can see that people feel safer, and they are safer. Gun violence goes down as a result of that.

My research center, the University of Chicago Crime Lab, we did a randomized trial with the City of New York a couple years ago where we helped put increased street lighting in some public-housing developments and not others. And one of the things that that would do is also potentially get more people out in public. We see violence decline there as well.

And then one other thing that I would just add—actually, two other quick things that I would add to this is: I think it gives you another way to understand all of the research and economics that suggests more police reduce crime. I know you had Jen Doleac on recently; you guys were talking about this.

I think most people would say, Oh that’s, like, deterrence or incapacitation. But when I look at the Chicago Police Department, for instance, the average Chicago cop makes about three arrests—not per week, not per month—per year. Three arrests per year.

Demsas: Wow.

Ludwig: So it’s, like, not a gigantic arrest machine that is generating all of this massive deterrence. What are police doing? Well, one of the things that they might be doing is helping interrupt these 10-minute windows. It’s something preventive, right? And I think that is a potentially important part of it.

And the thing that I would add to this, as well, is that sociologists believe that one of the most important determinants of a neighborhood’s violent-crime rate is what they call “collective efficacy”—this is research from the 1990s—the willingness of neighborhood residents to sort of step in and do something when there’s some sort of problem in the neighborhood. And I think that also is very consistent with this kind of behavioral-economics view of the gun-violence problem and what to do about it.

[Music]

Demsas: After the break: the problem with focusing on the “root causes” of gun violence.

[Break]

Demsas: Someone listening to this will say, How is this different from the root causes analysis that you kind of critiqued? Right?

Because there’s a really great quote that you have in your book, which is that we “treat gun violence as something that will get better once we fix everything else that’s wrong with society.” And I think that’s a frustration that a lot of people have, is that they are sympathetic to the idea that if we invested more in education, or if we invested more in social-welfare programs and UBI (universal basic income), expanded health care, that there would be reduced crime in 20 years, in 30 years.

But that doesn’t really respond to the specific concern of, Tomorrow when I walk to school, am I going to get shot? Can you help distinguish between your analysis and that root cause analysis?

Ludwig: What I hear in Chicago is something that you hear in lots of cities around the United States, is like, Gun violence is just a symptom of poverty, and we’re never going to solve the gun-violence problem until we solve the poverty problem.

And let us all hope that’s not true, because, as you know even better than I do, we’ve been working really hard for decades to try and solve the poverty problem in the United States, and it’s proven to be very difficult. I think the key optimistic observation or suggestion that we get from this behavioral-economics perspective on the gun-violence problem is: We can make massive changes in the gun-violence problem by changing parts of the social environment that are much easier to change than poverty and segregation and all of these other super big, super important social problems.

If I could wave a wand and I could end poverty and segregation in Chicago, believe me—I’d be the first person to wave that wand. And so I’m not arguing against any of the policies trying to do that. They’re super important. It’s more like, What else can we do on top of that to really start to make a meaningful difference on the gun-violence problem?

And I can’t wave a wand and end poverty in Chicago, but what I can do is: I can make it easier to have commercial development in Greater Grand Crossing than we currently have here on the South Side of Chicago. I can strategically deploy money to turn a bunch of vacant lots that are littered with empty broken beer and tequila bottles and turn that into a little pocket park that people are willing to be in. I can put money into things like block clubs. I can do some version of what the University of Chicago does, like put unarmed private security guards on some key corners to make sure that there’s an eye on the street because of that. So there’s a bunch of pragmatic things that you can do that can really make a difference that sort of complement these other efforts to address these really big root causes.

And maybe the one other thing I would just add: You might look at that sort of strategy and say, To some people, that’s going to feel unsatisfying that it is addressing a symptom, not the underlying cause. Like, we’re leaving the root causes there, and we’re just treating the symptom of the root causes. But I actually think what that concern or that perspective misses is that the causal arrow runs in both directions between gun violence and root causes, if that makes sense.

And you can sort of see a lot of these communities are in vicious cycles right now, where it’s like: You’ve got a lot of gun violence. People and businesses leave—fewer eyes on the street, fewer community resources to build the kind of public infrastructure that helps address this problem, even more gun violence, even more people leaving. There are lots and lots of neighborhoods, lots and lots of cities that are trapped in that sort of vicious cycle.

But if you can get the gun-violence problem under control. I think you can see that you can turn those vicious cycles into virtuous cycles. I think of gun violence, you know, not as a symptom of some deeper thing but in many ways as the social problem for cities that sits upstream of so many of the other social problems that cities are trying to wrestle with.

Demsas: To give your model in layman’s terms: Gun violence and shootings happen because there’s a large availability of guns and because people are not interrupted in pulling those guns out in the midst of a heated moment. So as you point out in your book, the vast majority of shootings are happening in the course of an argument—not in a premeditated sense but in [the sense] that someone bumps you on the sidewalk, or they insult you, or something like that—and that violence, that shooting happens because there’s no one to step in and say, Hey. Let’s calm things down. Is that kind of the overview that you’re giving us?

Ludwig: Yeah. The highest-level version of this is: All of our policies have conceived of gun violence as a problem of System 2 slow thinking, when I think it’s, actually, mostly a problem of System 1 fast thinking.

And so for starters, we just need a big reorientation to understand differently what the problem actually is to be solved. And once you have that reorientation—once you sort of think of gun violence as a problem of not bad people unafraid of the criminal-justice system, not people in bad economic circumstances stealing to feed their families, but normal people making bad decisions in fraught, difficult, 10-minute windows—one thing that you start to do then is start to think about, How do I change the social environment so there are more people, more eyes on the street to sort of step in and interrupt? And the other thing that you start to think more seriously about is, like, How do I focus my social policies more on helping people understand their own minds better and anticipate what they’re going to do in these difficult 10-minute windows?

And one of the ways that we can do that is through a very different type of social program than we’ve typically thought of in the U.S.—these behavioral-economics-informed programs like Youth Guidance’s Becoming a Man or Heartland Alliance’s READI program or YAP and Brightpoints’ Choose to Change program. These are all things that we’ve subjected to randomized controlled trials in Chicago.

And what they basically are doing is: They’re helping people understand that they’ve got fast thinking as well as slow thinking and recognize that their fast thinking can get themselves into trouble in these fraught moments, and helping them anticipate that and sort of better navigate those 10-minute windows. And you can see in randomized experiments that that reduces risk of violence involvement by, depending on the study and the time period, like, 30 to 50 to 60 percent. How you scale that, I think, is the frontier scientific and policy challenge, but at least now we can sort of see the direction that we’ve got to go.

And the other thing I would just add is: I think this behavioral-economics perspective also helps us understand why education is so important for solving the violence problem, but not in the way that people have historically thought. Most people would say, Yeah, of course, education is so central to solving the crime problem, because education improves people’s earnings’ prospects, and blah, blah, blah.

And it’s true that education is hugely important for people’s earnings prospects, and education is good for making better citizens. It is good for lots and lots of reasons. But the other thing that the data tell us education does is: It helps people learn to be more slow thinking and skeptical of their own minds in high-stakes moments. That turns out to be sort of a key byproduct of everything that schools ask people to do.

And I think of education as, like, in many ways, the most important sort of crime-prevention, gun-violence prevention tool that we have. I think things like rote learning are not what we want either for educational purposes or from the perspective of making schooling as sort of crime preventive as possible. And so I think there are other ways of reimagining what school does, which would be good for making school sort of more helpful for a world in which things like problem-solving are increasingly important for economic outcomes, but also super valuable for making education more helpful in addressing the gun-violence problem.

Demsas: You alluded to this a couple of times now, but it’s interesting that there’s one way to interpret your result as just, as like, We need to put a bunch more cops on the street, and those can be the eyes on the street. And that is kind of consistent with the literature we explored in the Jen Doleac episode around why increasing numbers of police officers can reduce crime, and violent crime, in particular. And the other avenue—I mean, these are complementary—is that there needs to be more attention on how to improve people’s System 1 thinking. And the Becoming a Man program, which I think is now really popular, is a great example of that.

But scaling these sorts of things is really, really difficult, as you mentioned. Are you indifferent between these two policy avenues, like an increased number of police officers, versus investing in programs that improve people’s ability to understand their own System 1, System 2 thinking? Or is it that you really want people to do one of those over the other? And in which case, it does seem very difficult to scale Becoming a Man and other programs. We have not been able to do that, despite years and years of positive coverage of that program.

Ludwig: For starters, I would say, we should be pushing on every possible front to solve this problem. It’s a huge humanitarian problem, one of the key drivers of Black-white life expectancy disparities in the United States, hugely important for the future of our cities that are the key economic engine for the whole country. So I wouldn’t say, like, Let’s do this or this. If we have multiple things that could be helpful, I’d say, Let’s push on every front.

On the eyes-on-the-street stuff, I would say, There’s tons of scalable stuff there, and it’s not just hiring more cops. So you can hire more cops in cities that like cops. You can put unarmed security guards on the street. You can fund community-violence-intervention nonprofit groups. You can clean up vacant lots and turn them into parks. You can improve street lighting. You can change zoning laws and permitting rules and whatever to make it easier to have stores interspersed with residential in a neighborhood. Tons of different things there that you could do, depending on the local political environment in your city, all of which are super scalable, all of which would be super helpful, all of which would increase the chances that there’s some sort of prosocial adult around who can sort of step in and de-escalate something.

On top of that, I think then there’d be huge value in trying to figure out how to scale the social programs that also help people better understand their own sort of thinking. And I think one of the most exciting visions for the future here comes from artificial intelligence, weirdly. My University of Chicago colleague Oeindrila Dube did a fascinating study with Sandy Jo MacArthur, who used to be at LAPD for many years, and my friend Anuj Shah, at Princeton.

They basically did Becoming a Man for cops. And what was so interesting about it is: Becoming a Man works with teenagers in middle school and high school. And it’s, like, an adult working with these kids, and that’s super hard to scale, because the program counselor is expensive, and they vary in skill, and How do you hire enough people? and everything that makes a social program hard to scale.

But the Becoming a Man for cops—what they did is they had this artificial-intelligence-driven force simulator thing, where they give cops feedback to see when their System 1, their fast thinking, is leading them to an unhelpful response, through a bunch of simulation exercises that the AI can do. And you look at the randomized control data, and it seems to have remarkably helpful impacts.

And I think the thing that’s so exciting about that is: Thinking about AI as a human-capital development tool lets you see, Oh I see. Once you’ve got the software, the marginal cost for rerunning software is super low. And the great thing about software is that it basically runs the same way over and over again. So we might be looking at a future where AI can be a super valuable way to enhance human capacity in ways that include addressing one of the most important social problems facing cities, which is gun violence.

Demsas: We’ve gotten a little bit into this, but trying to compare all three theories that are kind of existing out there: When we’re thinking about the root causes theory, that leads us to believe that we should invest a ton in antipoverty measures and expand healthcare, job-training opportunities, UBI, whatever. And then the wickedness theory kind of indicates that we should just try to root out and incarcerate bad people for as long as possible to prevent them from doing crime. Your theory, the “unforgiving places” theory—what do you want policy makers to take from that?

Ludwig: The first thing I want policy makers to take from this is to recognize that the gun-violence problem itself is different from what we think. Again, it’s not a problem of System 2 deliberate, slow thinking, people responding to incentives. Gun violence is mostly driven by System 1, reactive, fast thinking. That’s the most important thing.

From there, I would say we need to do two types of things. We need to change those aspects of the social environment that reduce the risk that conflict escalates. And related to that is, too, just in the safety net, is whatever your position on the Second Amendment, I think this is also why guns out in public are particularly worrisome. If people want to have 500 guns in their basement locked up, that’s one thing. But when people are taking guns out on the street, that’s the thing that makes interpersonal conflict on the South Side of Chicago so much more dangerous than interpersonal conflict in the south side of London or whatever. So people around to deconflict conflict when it happens, and anything that we can do to get guns off the street would be super helpful.

And then I think policies that help people, you know, both K–12 education and things like, you know, Becoming a Man to try and help people better anticipate and navigate those 10-minute windows. And that’s a policy agenda that really doesn’t make much sense under either the conventional wisdom of the left or right, right now. Those things aren’t about changing people’s incentives, so it’s like, Why in the world would they possibly work? But I think they’re really central to making huge progress on the problem. And I think if you look at the experiences of L.A. and New York over the last 30 years, they validate that view, or they’re certainly very consistent with that view, at least.

Demsas: Jens, always our last and final question: What is an idea that you once thought was great and ended up being only good on paper?

Ludwig: Great—so we launched a big research project with the superintendent of the Chicago Public Schools a couple of years ago. The huge priority of this superintendent was truancy. So Chicago used to have something like 150 truancy officers for its 600 schools in 1991, and with budget cuts, they got rid of all of them. And then you look at the data and, like, there are tons of kids who are missing three or four weeks of school a year.

And so you look at that, and the superintendent is like, This surely is a key reason that kids are not doing well in school. So Jon Guryan and I launched this big research project with CPS, and we worked really hard to try and figure out how to get kids to come to school more often, without the punitive whatever of truancy officers. With a bunch of partners, we managed to figure out a way to get kids to come back to school more often. And then we look at the data, and we see it does not boost their learning at all.

Demsas: Oh wow.

Ludwig: So weird, so counterintuitive. You would think, If you don’t go to school, you can’t learn. It’s super intuitive. And yet, you get kids to come to school more often, and they don’t learn.

Demsas: Wait. What’s going on? Doesn’t that kind of conflict with a lot of ed-policy research?

Ludwig: Yeah. So super weird, right? And so it was only very recently that Jon and I were looking at data right after the pandemic, and what you can see in the data is, for instance, if you look at eighth graders in Chicago, the average eighth grader in Chicago academically is like a sixth grader. And something like a third-ish of Chicago eighth graders academically are, like, closer to fourth graders.

Demsas: Wow.

Ludwig: And the eighth-grade teachers—their feet are being held to the fire to teach eighth-grade content. And so then you ask yourself, Why is it the case that sending a kid who, academically, is at the fourth-grade level to school to be taught eighth-grade content doesn’t improve their learning? Like, to ask the question is to answer it.

Demsas: So it’s like, basically, the kids who are missing a bunch of school are more likely to be the kids who are way behind in school. And so they’re going to benefit less from being in school.

Ludwig: Exactly.

Demsas: Oh wow. That’s a very depressing answer.

Ludwig: Yeah, we were confusing, you know, What is a cause, and what is effect? And so it seemed good on paper. Now we realize that there’s a very different underlying problem that we’re working hard to fix. But that’s my depressing answer to leave you with.

Demsas: Well, thank you so much for coming on the show. This was fantastic.

Ludwig: Thanks so much for having me on. It was great.

[Music]

Demsas: Good on Paper is produced by Rosie Hughes. 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.

All eyes are on New Orleans this week, as the Kansas City Chiefs and the Philadelphia Eagles arrive in Louisiana for Super Bowl LIX.

Read more...

After Donald Trump visited the Los Angeles area late last month to observe the damage from recent fires, he made a nakedly political demand: As a condition of releasing federal aid to stricken areas, he wants California to make voters show ID at the polls. Trump is reportedly convinced that he would have won the state if it had such a law, but that has nothing to do with fire safety.

If Trump wants California to mend its ways before receiving federal disaster relief, he could make some reasonable requests: The state should stop encouraging suburban sprawl in fire-prone areas, for example, and start pushing property owners to take more precautions. To the exasperation of emergency-management experts and budget hawks alike, California fires are like many other disasters all around the country. They lead to massive insurance settlements and outflows of government aid, which in many cases pay for rebuilding the same physical environment that left people and their homes vulnerable in the first place.

[M. Nolan Gray: How well-intentioned policies fueled L.A.’s fires]

State governments typically manage the aftermath of natural disasters; when they are overwhelmed, they appeal to Washington for additional money. For years, people in my field have been urging Congress to put strings on that relief. Unfortunately, Trump’s voter-ID demand—along with his insistence that the state should also change its water policies, which he did not appear to fully understand—triggered a righteous response among prominent California Democrats. “Conditioning aid for American citizens is wrong,” Governor Gavin Newsom declared in a statement. No, it isn’t, but Trump is setting back the cause of reform.

The present federal disaster-relief system, built over decades, involves multiple pots of money from a variety of agencies: the Federal Emergency Management Agency, the Small Business Administration, the Department of Housing and Urban Development. The 1988 Stafford Act, which governs the distribution of many of these funds, is built on the presumptions that major disasters are random, rare acts of God, and that communities hit by them need to be made whole again. But as climate change repeatedly exposes certain regions to the same disasters—fires in California, hurricanes along the Gulf Coast, tornadoes in the Great Plains—rebuilding the status quo looks less and less defensible. “Okay, that was a 40-year-old building; let’s rebuild a 40-year-old building,” one recovery official in Louisiana memorably said in a 2009 PBS report, capturing widespread frustration with federal rules governing New Orleans’s long recovery after Hurricane Katrina.

Some local jurisdictions have responded to disasters by taking steps to avoid a repeat. In 2018, the Camp Fire destroyed most of the buildings in Paradise, California, and killed 85 people. To build there now, homeowners must abide by local regulations, known as defensible-space requirements, that require them to remove vegetation that would otherwise help a fire move more quickly. In 2013, a tornado in Moore, Oklahoma, killed 24 people, including seven children in an elementary school. The community responded in part by imposing stringent new residential building codes.

When I visited Moore last year, after a devastating tornado season in Oklahoma, a builder named Marvin Haworth walked me through a home that requires sheathing, nail shanks, and hurricane straps under the new regulations. He was originally concerned about the changes, but the added cost to home purchasers is minimal, and the matter is settled now. “It is not a part of the discussion anymore,” he told me. “It’s the code. This is the way we are building and are going to build.”

[Nancy Walecki: The place where I grew up is gone]

Moore and Paradise both had the foresight to acknowledge the risks they face and take it upon themselves to change. States generally do not require such steps. Sweeping policy changes are difficult to enact immediately after a disaster. People are hurt and in need; political considerations demand that immediate distribution of money. That’s why FEMA administrators under presidents of both parties have proposed some version of conditional relief funding. Project 2025, the blueprint for Trump’s policies, calls on states to pay a “disaster deductible” before securing federal aid—a requirement that might motivate governors and legislators to demand better preparation for disasters. As The New York Times reported, that idea originally came from Craig Fugate, President Barack Obama’s FEMA chief, who insisted that Washington needed a mechanism to force states to do better advance planning. But states want money unconditionally, and substantive reform proposals such as Fugate’s have not survived political pushback.

Nevertheless, a lot of little changes—stronger nails, cleared yards—can add up to a more resilient society. Trump’s focus on political payback is unfortunate, because the current system needs an overhaul. Disasters are no longer random or rare. When disaster strikes, we should rebuild accordingly.

From inside the room, Robert F. Kennedy Jr.’s confirmation hearings felt at times more like an awards show than a job interview. While the health-secretary nominee testified, his fans in the audience hooted and hollered in support. Even a five-minute bathroom break was punctuated by a standing ovation from spectators and cheers of “We love you, Bobby!” There were some detractors as well—one protester was ushered out after screaming “He lies!” and interrupting the proceedings—but they were dramatically outnumbered by people wearing Make America Healthy Again T-shirts and Confirm RFK Jr. hats.

The MAHA faithful have plenty of reason to be excited. If confirmed, RFK Jr. would oversee an agency that manages nearly $2 trillion, with the authority to remake public health in his image. “He just represents the exact opposite of all these establishment agencies,” Brandon Matlack, a 34-year-old who worked on Kennedy’s presidential campaign, told me on Wednesday. We chatted as he waited in a line of more than 150 people that snaked down a flight of stairs—all of whom were hoping to get a seat for the hearing. Many of them were relegated to an overflow room.

Of course, Matlack and other RFK Jr. fans could be in for a letdown. Whether Kennedy will actually be confirmed as health secretary is still up in the air. There was less raucous cheering during the hearing on Thursday, as Kennedy faced tough questions from Republican Senator Bill Cassidy of Louisiana, whose vote could be decisive in determining whether Kennedy gets confirmed. But the realization of RFK Jr.’s vision for health care in America may not hinge on his confirmation. The MAHA movement has turned the issue of chronic disease into such a potent political talking point that people like Matlack are willing to trek from Pennsylvania to Washington, D.C., just to support Kennedy. Politicians across the country are hearing from self-proclaimed “MAHA moms” (and likely some dads too) urging them to enact reforms. They’re starting to listen.

Kennedy is best known as an anti-vaccine activist who spreads conspiracy theories. His more outlandish ideas are shared by some MAHA supporters, but the movement itself is primarily oriented around improving America’s diet and the health problems it causes us. Some of the ways MAHA wants to go about that, such as pressuring food companies to not use seed oils, are not exactly scientific; others do have some research behind them. Consider artificial food dyes, a major MAHA rallying cry. Multiple food dyes have been shown in animal studies to be carcinogenic. And although a candy-corn aficionado likely isn’t going to die from the product’s red dye, the additive is effectively banned in the European Union. (Kennedy claims that food dyes are driving down America’s life expectancy. There’s no evidence directly linking food dyes to declining human life expectancy, though one study did show that they cut short the life of fruit flies.)

Until recently, concerns over food additives, such as artificial dyes, were the domain of Democrats. In 2023, California became the first state in the nation to ban certain food additives: red dye 3, potassium bromate, brominated vegetable oil, and propylparaben. But now food-additive bans are being proposed in Donald Trump country, including West Virginia, Arkansas, and Kentucky. Some of the efforts, such as the Make Arkansas Healthy Again Act, are an explicit attempt to enact Kennedy’s agenda.

In other states, something a bit more Schoolhouse Rock is happening: The purported dangers of food ingredients are riling up Americans, who are then going to their lawmakers seeking change. Eric Brooks, a Republican state legislator in West Virginia, told me that he decided to copy California’s food-additive bill after being prodded by a constituent. “I said, Well, I don’t normally look out West, especially out to California, for policy ideas, to be honest with you,” he told me. “But once I had done the research and I saw the validity of the issue, I said, Okay.” Although his bill, which was first introduced in January 2024, did not move forward in the previous legislative session because, in his words, “there were bigger fish to fry,” Brooks hopes the national interest coming from the MAHA movement will motivate West Virginia legislators to take another look at the proposal.

Other red-state efforts to follow California’s lead have similarly not yet been passed into law, but the fact that the bills have been introduced at all signifies how motivated Republicans are on the topic. The MAHA moms may be enough to propel Kennedy to the job of health secretary. One Republican, Senator Jim Banks of Indiana, told Kennedy on Thursday that he plans to vote to confirm him because doing anything else “would be thumbing my nose at that movement.” Republicans are not only introducing bills to ban food additives; they’re also taking up Kennedy’s other MAHA priorities. Arkansas Governor Sarah Huckabee Sanders recently asked the federal government for permission to forbid food-stamp recipients in her state from using their benefits to buy junk food—a policy Kennedy has repeatedly called for.

Over the past two days, senators seemed shocked at just how loyal a following Kennedy has amassed. One pejoratively called him a prophet. MAHA believers I spoke with made clear that they want RFK Jr.—and only RFK Jr.—to lead this movement. Vani Hari, a MAHA influencer who goes by “Food Babe” on Instagram, told me that, if Kennedy fails to be confirmed, there “will be an uprising like you have never seen.”

But at this point, it seems that the movement could live on even without Kennedy. On Wednesday, the Heritage Foundation hosted a press gaggle with MAHA surrogates. Heritage, an intellectual home of the modern conservative movement, is now praising the banning of food dyes. This is the same pro-business think tank that a decade ago warned that should the FDA ban partially hydrogenated oils, trans fats that were contributing to heart attacks, the agency “would be taking away choices,” “disrespecting” Americans’ autonomy, and mounting “an attack on dietary decisions.”

Republicans have generally been a bulwark for the food industry against policies that could hurt its bottom line. Democrats and Republicans in Washington today seem more aligned on food issues than ever before. “When you have members on both sides of the aisle whose constituencies are now extremely interested in understanding fully what’s in their food products, it does shift the game a bit,” Brandon Lipps, a food lobbyist and a former USDA official under Trump, told me. After all, even Bernie Sanders, despite chastising Kennedy multiple times during this week’s confirmation hearings, professed his support for aspects of the MAHA movement. “I agree with you,” Sanders told Kennedy on Thursday, that America needs “a revolution in the nature of food.” During the same hearing, Cassidy said that he is struggling with Kennedy’s candidacy because, despite their deep disagreements on vaccine policy, on “ultra-processed foods, obesity, we are simpatico. We are completely aligned.”

Although MAHA has shown itself to be a unifying message, the real test will come down to the brass tacks of any political movement: passing actual legislation. How willing lawmakers—especially those with pro-business proclivities—will be to buck the status quo remains to be seen. Politicians have proved eager to speak out against ultra-processed foods; in that sense, MAHA is winning. But policy victories may still be a ways off.

To judge by the shelves of America’s vice merchants, the nation is in the grips of a whipped-cream frenzy. Walk into any vape store or sex shop, and you’ll find canisters of nitrous oxide showcased in window displays—ostensibly to catch the eye of bakers and baristas, who use the gas to aerate creams and foams. At the bodega near my apartment, boxes of up to 100 mini-canisters are piled up to eye level, next to Baby Yoda bongs.

In fact, culinary professionals generally don’t shop for equipment at stores with names like Puff N Stuff or Condom Sense. The true clientele inhales the gas to get high. A dangerous and technically illegal drug, nitrous oxide is widely available as long as everyone pretends it’s destined for use as a food product. Indeed, a whole industry appears to have built its business model around exploiting this loophole. Large distributors brand and flavor nitrous in ways that attract young inhalers, stock it with retailers catering to other vices, and sell it in quantities that are implausible for culinary use but ideal for huffing. The gas can even be ordered from Walmart, Amazon, and eBay. Without meaningful regulation, getting high on nitrous will remain as convenient as picking up a bag of chips.

Nitrous oxide has been recreationally inhaled since the 1800s. It induces a euphoric head rush and tingling in the user’s fingers and toes, often followed by giddy laughter. Almost as soon as it starts, it’s over. The effects of a single hit typically last for less than a minute.

The modern version of the drug, known colloquially as a “whippet” or “whip-it,” has recently climbed the youth-popularity ranks. (These products are distinct from the Reddi-Wip cans found at the supermarket, which contain cream and nitrous oxide together. The canisters proliferating now contain the gas alone.) Although pandemic-era lockdown protocols hampered illicit drugs’ supply chains, nitrous oxide remained broadly available by comparison. In 2018, about 12.5 million Americans over age 12 reported having ever used it, a number that rose to nearly 14 million in 2022. Social media is full of clips of young people ripping hits before falling on their faces (so much so that TikTok eventually banned nitrous-related search terms). Fans of the drug have created gas-tank accessories in the video game Roblox; one rapper’s song “Whippet” features him and his entourage ripping hits of nitrous from tanks tucked in their waistbands between verses. A recent Columbia graduate told me that, back in college, one friend’s birthday party featured a salad bowl “full to the brim” with used nitrous canisters. (He spoke on condition of anonymity in order to discuss illegal drug use.)

[Malcolm Ferguson: Marijuana is too strong now]

As you might imagine, depriving your brain of oxygen in favor of laughing gas is not wise. Heavy nitrous users can suffer severe health consequences, including, occasionally, death. Varun Vohra, an emergency-medicine professor at Wayne State University, in Detroit, told me that heavy users experience symptoms including irregular walking, bodily weakness, and severe limb pain. (Nitrous-oxide-related emergency-room visits in Michigan more than doubled from 2022 to 2023.) Inhaling the gas deactivates vitamin B12, which harms nerves in the brain and the spinal cord. Among chronic users, this can eventually induce paralysis. Users also report depression, anxiety, mood swings, and even hallucination. A senior at Tulane University, who also spoke on condition of anonymity, told me that he’d quit the drug after one night when he took a deep inhale and blacked out. He awoke on the floor a few seconds later, unable to remember what had happened; his friends told him he’d had a seizure.

Perhaps the greatest danger arises from what people do while huffing the drug. Ed Scott, a city-council member in Rialto, California, told me that his son, Myles, died in a car accident after his friend inhaled the gas while driving and passed out at the wheel. This inspired Scott to investigate the drug’s use in California. He told me that he found many other fatal car accidents in which nitrous-oxide products were discovered at crash sites but police—who thought they were helium cannisters—did not register the crashes as DUIs.

Given the health and safety risks, the sale and use of nitrous oxide for recreational purposes is technically illegal. The key word here is technically. Businesses can generally get away with selling nitrous oxide over the counter as long as they say it’s for culinary use—and pretend not to know what customers are really doing with it. (Some states prohibit sales to anyone under 21 or 18.) If, that is, they even bother pretending. The Tulane senior told me that the clerks at his local smoke shop greeted him with shouts of “whip-it boy!” when he visited to restock.

A major industry has grown out of the regulatory vacuum. In 2020, Marissa Politte’s family sued United Brands, the Silicon Valley–based company that distributes the brand Whip-It!, after she was killed by an unconscious nitrous-huffing driver. Documents revealed by the lawsuit suggest that the players involved know they’re benefiting from a legal loophole. A former warehouse employee estimated that three-quarters of United Brands’ customers were smoke shops, not bakeries or cafés. The client list named retailers including Mary Jane’z Novelties, Herban Legend Smoke Shop, Smoke 420, and Precious Slut 1. In a seven-year span, documents show, the company sold about 52,000 “chargers,” or miniature nitrous capsules, to Kaldi’s Coffee—and more than 1 million to the It’s A Dream smoke shop.

Internal emails between United Brands and retailers, uncovered by the lawsuit, suggest a certain cynicism around legal compliance. One United Brands employee emailed a retailer requesting that the Whip-It! chargers “are used properly and legally.” The retailer responded: “yah man we know the deal we put a disclaimer… you know we all got to cover our asses, better safe than sorry.” (United Brands did not respond to multiple requests for comment.)

Meanwhile, many other companies market nitrous in ways that seem conspicuously ill-tailored to professional pastry chefs. A variety of gas tanks are sold with names that evoke marijuana strains, not whipped cream—Monster Gas (slogan: “Become easy, Become happy!”), Hippie Whippy, Baking Bad, Cosmic Gas—and in flavors such as mango smoothie and tropical punch. Colorful labels feature fruits, unicorns, women in bathing suits, or sports cars. (Hervé Malivert, the director of culinary affairs at the Institute of Culinary Education, told me that many food professionals are loyal to iSi chargers, which they order from the company’s website—not at a gas station. Malivert said he has never heard of a chef using flavored gas or gas tanks.)

One of the most prominent nitrous companies is Galaxy Gas, which was founded in 2021 by three brothers who ran Cloud 9, an Atlanta-based smoke-shop chain. A Cloud 9 executive told New York magazine that Galaxy Gas at one time made up nearly 30 percent of all nitrous sales nationwide. Asked by CBS to explain why anyone would need to buy a tank with enough nitrous oxide to make thousands of servings of whipped cream, a Galaxy Gas spokesperson claimed that the product was for customers seeking an “erotic culinary lubricant.” (The company’s rising profile in the world of whip-its has brought legal scrutiny. According to New York, its trademark was recently sold for $1 to a newly registered corporate owner, and it has paused direct sales. Galaxy Gas did not respond to requests for comment.)

[Mike Riggs: Congress accidentally legalized weed six years ago]

In response to the spread of nitrous-oxide use, states and the federal government have begun taking steps toward more effective regulation. The most notable example is Louisiana, where the drug used to be outlawed only if sold for the purpose of being inhaled. Legislation enacted last May made its sale presumptively illegal, with carveouts for genuine industrial and culinary use. Jeanette Brick, the president of iSi North America, told me that her company does not oppose the Louisiana law. “These laws are intended to prevent misuse, and they have not negatively impacted our ability to serve the culinary community in the state,” she said in an email. She also noted that “iSi does not sell large tanks of nitrous oxide, as they have no culinary application and are increasingly associated with misuse. We strongly advocate for additional restrictions on these large tanks to help curb their growing misuse among teenagers.”   

Outside Louisiana, however, legislative efforts have yet to deliver significant change. In 2017, for example, then–California State Senator Jim Nielsen proposed a bill to ban nitrous-oxide sales in stores selling tobacco or tobacco-related products. Industry lobbyists opposed the bill, and it eventually failed to pass.

In the absence of effective regulation, litigation has emerged as the best tool to achieve accountability. In 2023, a Missouri jury found United Brands liable for conspiring to sell nitrous oxide as a drug. The company was ordered to pay $720 million to Marissa Politte’s family for her wrongful death. The novel jury verdict and the large court-mandated payout might set a lasting precedent. “You have companies whose full-time scheme it is to pour this stuff onto our streets,” Johnny Simon, the Politte family’s attorney, told me. “That’s who we need to go after.”

Still, case-by-case litigation can push only so hard against countervailing market forces. Nor should it have to, when the blueprint for lifesaving regulation exists. Had something like Nielsen’s California bill been federal law in 2020, Politte might still be alive. The driver who killed her bought his nitrous oxide from a smoke shop.