No heads in the clouds

Could covid-19 shake up air travel for the better?

Governments’ re-entry into airlines may spur more competition


Economy passengers taking one of the few international flights still running have had an unusually pleasant experience of late. 

Exasperated cabin crew battling to close overhead lockers full to bursting with wheelie-bags, duty-free booze and laptop cases have been replaced by masked attendants presiding over planes two-thirds full at best and often with only a handful of passengers. 

Some report sleeping across empty rows of seats. 

The collapse of passenger numbers and revenues will damage the industry. 

Yet previous disruptions have shaken up the airlines to the benefit of the flying public. 

It could happen again.

The advantages of the previous big interruption to air travel, the second world war, are debatable. 

The experience of servicemen crammed into uncomfortably spartan transport aircraft, argues Eric Zuelow in his book, “A History of Modern Tourism”, created a generation without pre-war expectations of air travel as glamorous and luxurious. 

When those same aircraft were converted for civilian use, travellers accepted being packed in like sardines. 

The cattle-truck economics worked in passengers’ favour when deregulation brought competition to a previously highly regulated and often state-controlled industry. 

Air fares tumbled everywhere. Between 1995 and 2014 they fell by half in real terms, according to iata, a club of the world’s airlines.

World wars aside, past crises have passed speedily. The terrorist attacks of 9/11 and the financial crisis of 2009 were “nasty shocks”, says Brian Pearce, chief economist of IATA, but “small bumps in the road” compared with covid-19. 

At its height in April 2020 flying was down by 94% over a year earlier, measured by total revenue passenger kilometres (rpks). 

Current estimates are that rpks in 2020 will be down by 66% on 2019, when (including domestic flights) 4.5bn passengers took to the skies. 

Only 1.8bn people buckled up last year. By the end of it, some 30% of the global fleet was still grounded, according to Cirium, an aviation-data firm. 

And the sense of optimism that had slowly grown took a fresh hit early in 2021 as more transmissible and dangerous variants of the virus emerged, leading to renewed border closures and capacity cuts.

The plunge comes after years of bumper growth. In each 15-year period since 1988 rpks doubled. They were expected to do so again between 2018 and 2033, according to Airbus, the European half of the duopoly that builds the world’s biggest passenger jets. 

Boeing, its American rival, notes an acceleration over the past decade, when growth averaged 6.5% per year, above the long-term average of 5%. 

Darren Hulst of Boeing is confident of an eventual return to a growth trend of 5% a year, even if over the decade from 2019 the number may be closer to 3.7%. 

In the meantime it will be a rough ride for airlines. 

Only 2.8bn passengers are expected to take to the air this year.



Signs of recovery are scant. 

Domestic flying in China and Russia has largely recovered, but in Australia it is still down by 86% and in America, the world’s biggest domestic market, it has fallen by 60%. 

International routes are suffering the most. 

Despite huge capacity cuts (only a quarter of the international flights scheduled a year earlier were still running in January), load factors are still low. Planes are often taking off a third full at most.

The opportunity for travellers to stretch out comes at a heavy cost. 

Full service will not resume for a long while. 

Mr Pearce says the gradual deployment of vaccines merely makes him more confident about his baseline forecast of a return to 2019 levels by 2024. 

Not all air travel will recover at the same pace. Domestic flying will continue its rebound, followed by regional international traffic. 

The last part of the industry to refill its seats will be long-haul flying. This has a disproportionate impact on legacy airlines that still rely most on long-haul international routes.

All airlines face a bleak period. 

Rather than increase by 4% on 2019 levels to $872bn, air-transport revenues in 2020 have plummeted to $340bn, reckons iata. 

In a good year the industry makes an operating profit of $50bn. Losses for 2020 are forecast at $118bn and, even if the second half of this year sees a recovery, they will still be around $38bn in 2021.

Airlines have responded by slashing costs, getting rid of staff and cutting fleets, yet carriers continue to burn cash at a rate of $5bn-6bn a month. 

Looking at the cash and liquid assets of carriers in mid-2020, the median airline had enough funding to last just eight-and-a-half months (or half way through this month), according to iata. 

Some big names have succumbed. 

Norwegian Air Shuttle, one of Europe’s largest low-cost carriers, sought bankruptcy protection in November 2020. 

Dozens of smaller airlines including FlyBe, a British low-cost carrier, Virgin Australia and Avianca, based in Latin America, have gone to the wall.

Two types of carrier have the best chance of survival. 

Those with sound business models and strong balance-sheets, such as Ryanair in Europe, Southwest in America and AirAsia, are all low-cost carriers ready for a rebound. 

And legacy carriers, mostly propped up by governments, will keep their grip on long-haul flying when it returns.

Subsidies, subsidies

Support for airlines has been prodigious. 

Governments are used to helping airlines for reasons of national pride and to maintain international connections. Many countries in the Middle East, India and Africa cling to ownership of loss-making national carriers. 

Others are just returning to the orbit of governments. 

Bruno Le Maire, France’s finance minister, has promised to “do what is necessary to guarantee the survival of Air France”. 

Germany’s Lufthansa has had a bumper aid package. 

America’s airlines, never shy to extend an upturned palm, benefited handsomely from its cares stimulus act, signed into law in March 2020, which set aside $25bn in grants for airlines and the same in loans and guarantees.

The result will be an industry wallowing in debt and beholden once again to the state. 

Aviation Strategy, a consultancy, puts the total in 2020 of government support plus loans from banks, other investors and aircraft-leasing companies at $600bn, an increase of $170bn since the end of 2019. 

Keith McMullen, a partner at the consultancy, points out that most government aid has come in the form of loans, the terms of which incentivise rapid repayment. But the fragile state of the industry will make that tricky. 

European governments may then be obliged to swap debt for equity.

A return to the days before deregulation and the rise of low-cost carriers, when governments controlled national flag-carriers, set (high) fares and decided routes based on national goals rather than demand, is the fear of Andrew Charlton of Aviation Advocacy, another consultancy. 

America’s big carriers, Asian giants such as Singapore Airlines, and Lufthansa and Air France-klm may be subject to government meddling. 

The terms of bail-outs for Air France stipulate that it must no longer compete with high-speed rail on some routes; Lufthansa’s decide what planes it may purchase.

That could give the low-cost carriers an advantage. Michael O’Leary, boss of Ryanair, admits to mixed feelings about state aid. 

Although it is “hugely distorting”, it is also likely to “sustain high-cost employment”, he says. 

And high-cost competitors are what he and his kind thrive on. 

Legacy airlines, already having to compete harder on short-haul routes and often weighed down by heavy pension-fund deficits, could struggle to adapt to an uncertain future that will require flexibility and nimbleness as well as lower costs.

Moreover, legacy carriers rely especially heavily on lucrative business travellers, a sector that may never recover fully, to subsidise seats at the back of the plane. 

If business customers and feeder routes dwindle, the complex web of domestic and international flights becomes harder to sustain, reducing connections and forcing up long-haul prices.

Not all will suffer. 

British Airways has slashed costs and raised capital and is fast-tracking its previous restructuring effort to compete with low-cost rivals. Other models look riskier. 

The Gulf carriers—Emirates, Etihad and Qatar Airways—that have grown swiftly in recent years may struggle. 

They lack a large domestic market to fall back on, unlike their Chinese competitors, and passengers may prefer not to mix with people from all over the world as they change planes at vast airports in the Gulf. 

If their model falters, the sky will be clearer for China’s state-controlled behemoths, which are already eating away at long-haul routes.

Low-cost carriers such as Ryanair and Wizz Air, based in eastern Europe, sniff opportunities. Mr O’Leary has promised to slash fares to rekindle his business. His ability to compete with airlines under the heavy hand of the state will be helped by several factors. 

The availability of bargains from embattled planemakers and leasing companies will help carriers in decent health or with new business models. 

Ryanair confirmed an order for another 75 Boeing 737maxs in December, taking its total to 210. 

The plane’s grounding after two fatal crashes had led to cancelled orders, so a big vote of confidence from Ryanair was surely worth a hefty discount. 

Wizz Air, which relies on more robust parts of the market such as younger flyers and east Europeans going west for work and returning to visit family, also has big growth plans. 

Southwest, America’s biggest low-cost carrier, is in expansion mode as the country’s legacy carriers retreat. 

Tony Fernandes, boss of Air Asia, based in Malaysia, is “quietly optimistic” about 2021, and hopes to start three new airlines in the region.

A surplus of pilots where there was a dearth and unemployed cabin crew looking for work could also boost low-cost carriers. 

Once-scarce landing slots at busy airports may become more available. 

And, if prices rise for long-haul seats, this may rekindle interest in low-cost long-haul services, which have a patchy history. 

The demise of Norwegian was largely down to its inability to make profits on long-haul routes. 

The business model of short-haul carriers does not easily translate to long-haul flights, where fuel is a bigger share of costs and schedules are compromised by time differences and airport curfews that do not affect short-haul travel. 

But a new plane, Airbus’s a321xlr, a single-aisle jet with the ability to fly long-haul routes, could change the equation again.

The commercial-aviation business will be smaller for years to come. Yet opportunities abound, especially if legacy carriers fail to adapt. 

Higher fares and lower costs will attract competitors that threaten the old guard. 

In 20 years’ time air travel will be back on track and growing again but the old names may no longer be on the tarmac. 

Just as Pan Am and TWA are distant memories, so a clutch of now familiar airlines may become relics of the past. 

The Fed needs to call time on cheap money

Jay Powell, of all people, is haunted by the ghost of past monetary policy mistakes

Gillian Tett 

    © Ingram Pinn/Financial Times


If in doubt, double down.

That is one way to frame the current message from Jay Powell, Federal Reserve chair.

At the start of the week, there was lively speculation about whether the Biden administration’s proposed $1.9tn stimulus plan could unleash inflation — and thus force the Fed to rein in its loose monetary policies soon.

However, in a speech to the Economic Club of New York on Wednesday, Powell emphatically ruled this out. Instead, he stressed that the Fed needed to maintain a “patiently accommodative” policy stance until the economy hits “full employment” and inflation has been running above 2 per cent for some time.

He also said that neither would appear soon. In plain English, this means virtually free money is here to stay for a long time, never mind the inflation chatter. Think years, not months before policy changes.

Is this a mistake? I think so. That is not because I object to the idea of radical monetary policy, or ultra-low rates in general. 

The Fed was correct to employ quantitative easing after the 2008 crisis to ward off the risk of a depression. So too when Covid-19 struck last year. 

And while I would prefer to see slightly higher rates now, the current stance can just about be justified by low inflation; in January, consumer prices were unchanged for a second month.

What worries me, however, are the signals that the Fed is sending about the future. 

Previous generations of central bankers thought it was foolish to give excessively precise signals about future policy as that would either tie their hands or encourage market complacency, or both. 

However, over the past decade, central bankers started to offer a framework for how they would respond to future economic data, hoping this would reduce the chance of market shocks by creating more predictability.

That was sensible. But the Fed has now taken this so-called “forward guidance” to a new level that seems dangerous. Most notably, over the past year Fed officials have indicated that policy will stay unchanged until late 2022, if not beyond. 

That means policy looks time-dependent, not just data-dependent — and since a central bank cannot be both (without magical powers of perfect foresight), this sense of time dependency is shaping investor attitudes.

As it happens, Powell seems minded to correct this. 

He has said before that he wants to be data-dependent and, on Wednesday, did not mention a precise date when policy would change, instead citing goals such as “full employment” and inflation above 2 per cent as the triggers for a policy shift.

But these are fairly vague targets. 

They are also moving ones. On Wednesday, Powell also stressed that 2 per cent is no longer an upper limit for inflation, but an average target, meaning he wants inflation to exceed this for a while.

For investors, data dependency is thus discretionary, and on a par with time dependency. And since investors always crave certainty about the future — even more than low rates — this has fostered speculation. 

Hence the sky high price of assets such as bitcoin or the boom in market vehicles such as special purpose acquisition companies.

Doves may retort that this market mania is a price worth paying to reflate the economy — which, as Powell stressed on Wednesday, helps the poor. 

Fed officials also fear that pre-emptive efforts to prick the market bubble would recreate a “taper tantrum” — the 2013 market ruction unleashed when Ben Bernanke, then Fed chair, hinted that QE might end.

Powell did not mention “taper” on Wednesday. Yet the ghost of Bernanke’s experience clearly haunts him. As he said three years ago, with touching honesty: “The taper tantrum left scars on anybody who was working at the Fed at that time.”

One bitter irony about current Fed policy is that even if Powell hopes to help America’s poor, cheap money is actually increasing inequality by elevating the value of assets held by the rich. Another irony is that Powell’s efforts to avoid a taper tantrum now, could create a bigger one later. Powell, of all people, seems well placed to recognise this.

Back in March 2013, for example, when Powell was a new-ish member of the Fed’s policy setting board, he warned Bernanke that the Fed’s failure to signal an end to ultra loose policy was prompting investors to take dangerously risky bets.

Then, in June that year, he argued that while an abrupt change in policy could spark market turmoil, it would also be damaging “if investors conclude that serious consideration of reducing [QE] is postponed indefinitely”. 

As the transcript reports him saying: “We find ourselves in a situation here where we’re on the roof, and there is no risk-free path. We’ve got to jump.”

He should heed that advice. I would not argue that the Fed needs to “jump” by tightening immediately, or threaten to do so; the 2013 taper tantrum shows the danger of abrupt and clumsy talk.

Even so, Powell does need to find a ladder off the “roof”. He should puncture assumptions that cheap money is here indefinitely, or that Fed policy is a one-way bet. 

Otherwise his attempts to ward off the ghosts of 2013 will eventually unleash a new monster in the form of a bigger market tantrum, far more damaging than last time — especially if investors have been lulled into thinking the Fed will never jump.

Meet the Undercover Anti-Fascists

Embedded with the team of anti-fascist researchers and activists who infiltrate and expose Proud Boys, neo-Nazis, militias, and other members of the violent far right

By ANDY KROLL 

         Christian Exoo photographed by Damian Battinelli in February 2021 / Damian Battinelli for Rolling Stone


On the morning of Wednesday, January 6th, as supporters of Donald Trump gathered near the White House for a last stand to “Save America,” Molly Conger said goodbye to her two dachshunds, Otto and Buck, tossed a wig into her car, and began the two-hour drive from her home in Charlottesville, Virginia, to Washington, D.C.

A journalist and online researcher, Conger specializes in infiltrating and exposing the violent far right. Using dummy accounts and pseudonyms, she lurks in private chat rooms and invitation-only forums used by neo-Nazis, militias, Proud Boys, and other right-wing extremists. 

When she sees someone make threats or plan for violence, she screenshots the person’s messages, digs up the person’s real identity and employer, and publishes her findings on her Twitter account, @SocialistDogMom, where she has more than 110,000 followers.

Because of her work, Conger is well known among the people she tracks. Members of the online far right have made grotesque sexual comments about her and expressed a desire to physically harm or kill her. She knows this because, in some cases, she’s observed them saying these things without them realizing it. 

She anticipated some of those people being in the crowd on January 6th, and with her buzz-cut blond hair and slight build, Conger feared she would be easily recognizable. When she got to Washington, she put on the wig and a pair of sunglasses and met up with a friend who is a trained medic.

“The Proud Boys had instructed their members to beat me to death on sight,” she told me. “I traveled with a medic in case I got stabbed.” She planned to take as many photos and videos as she could, gathering evidence for future investigations.

The crowd was larger than she’d expected — she saw enough people to fill a professional football stadium. Some ate from cans of tuna as they listened to Donald Trump Jr. (“The people who did nothing to stop the steal, this gathering should send a message to them”), Rudy Giuliani (“Let’s have trial by combat!”), and other Trump allies rail against the “stolen” election.

When Trump addressed the crowd around midday, people climbed into trees to try to catch a glimpse of him. As Trump egged on his supporters — “if you don’t fight like hell, you’re not going to have a country anymore” — Conger watched the people around her, masked guys in tactical gear next to dads with kids and senior citizens with Trump flags, whip one another into a frenzy. “There was this crusade mentality,” she says.

Since the 2020 election, Conger had monitored a variety of social-media platforms, including Facebook; Parler, the Twitter-like platform funded by at least one member of the conservative Mercer family; and Telegram, an app favored by white supremacists, racists, and anti-Semites. 

She was disturbed by what she saw. In Facebook groups filled with MAGA fans, the tenor of the conversation was moving toward the language she saw in the more hardcore groups. 

“These suburban dads are talking about murdering journalists, murdering Nancy Pelosi, hanging Chuck Schumer,” she told me. “Their rhetoric was starting to sound like the race-war guys’.”

She followed the MAGA march as it marched to the Capitol, and was struck by how the light the police presence was in the heart of America’s capital. As her Twitter handle would suggest, Conger identifies as a socialist and an anti-fascist. She’s marched for racial justice and against police violence in dozens of cities. 

“I’ve been pushed down, beaten, choked, and gassed by a variety of cops,” she says. 

Now, as the pro-Trump crowd reached the outer security perimeter of the Capitol, she braced for the police to use force on the marchers. “My body is reacting based on past experience: This is where it gets ugly. And then it just … didn’t happen.”

At the base of the Capitol, she finally peeled away from the crowd and watched as the mob fought past the outnumbered police force and ransacked the seat of American democracy. She struggled to process what she was seeing. 

For years, researchers and journalists like her had warned about the thin line between online hate and real-world violence, and about “stochastic terrorism,” the idea that a drumbeat of hateful rhetoric and political demagoguery leads to unpredictable acts of violence. Still, she had never imagined this.

It was “a logical end point for something I’ve been watching for a long time,” she told me a few days later. “But it’s still a hard thing to absorb watching a group of MAGA dads beat a cop.”

“Seeing it coming and seeing it happen,” she said, “are two very different things.”

A FEW DAYS BEFORE the 2020 election, I was sitting in my apartment when a dozen unfamiliar faces, Molly Conger’s among them, appeared on my laptop screen. The people on the Zoom call were in their twenties and thirties; most knew each other online by their Twitter handles, but for many it was the first time they had seen one another’s faces or heard one another’s voices. 

Nearly all of them identified as anti-fascists, and all had years of experience infiltrating and exposing the online far right. They had been brought together by a progressive not-for-profit to watch for and prevent election-related violence, for which they would receive a modest stipend for several weeks of work.

The group was so new, it didn’t yet have a name; it later settled on Deplatform Hate. 

“It was very ‘Antifa avengers assemble!’ ” says Talia Lavin, a member of the group who had recently published Culture Warlords, her chronicle of investigating online white supremacists and neo-Nazis.

I listened as different people described what they were seeing in different corners of the internet. (I was allowed to embed with the group on the condition that I not name the not-for-profit that helped form it or anyone involved who didn’t want to be identified.) 

The group’s de facto leader is an activist and writer named Christian Exoo. On the Zoom call, Exoo reported back on a nightly planning call hosted by the Oath Keepers, a dues-paying national militia group popular among current and former law enforcement officers. 

An Oath Keepers leader had talked on the call about defending the White House in the event “Antifa rushes it” and even doing battle with the Secret Service until it “runs out of bullets.” Exoo said he thought this was mostly bluster, but asked the team to be on guard for similar rhetoric among other groups. 

Another member said militia groups were plan possible operations using an online map of “Protect the Results” demonstrations that had been assembled by a number of progressive groups. Liberal activists and Oath Keepers colliding on election night had all the makings of a disaster; Exoo made a note to alert the liberal groups about what was happening so they could prepare accordingly.

For as long as hate groups have used the internet to advance their cause, a loose network of journalists and activists have committed themselves to disrupting them. 

Devin Burghart, executive director of the Institute for Research and Education on Human Rights who has tracked the Ku Klux Klan and neo-Nazis since the early 1990s, says one of the biggest challenges of this work is keeping up with the far right’s constantly changing tactics, whether by moving to different platforms or devising new ways to recruit members. 

Today, according to a recent analysis by the Center for Strategic International Studies, far-right extremists “operate and organize to a great extent online, challenging law-enforcement efforts to identify potential attackers,” adding that “right-wing terrorists have used various combinations of Facebook, Twitter, YouTube, Gab, Reddit, 4Chan, 8kun (formerly 8Chan), Endchan, Telegram, Vkontakte, MeWe, Discord, Wire, Twitch, and other online communication platforms.” 

The need for researchers and infiltrators familiar with all of these platforms, skilled in online surveillance, and fluent in internet culture was greater than ever.

Anti-fascist activists and journalists are far from the only ones working undercover online. The FBI does it, as do state and local law-enforcement agencies. Last fall, the FBI arrested a group of men who dubbed themselves the Wolverine Watchmen and were planning to kidnap Gov. 

Gretchen Whitmer of Michigan; in a criminal complaint in the case, the feds alleged that they first took notice of two of the group’s leaders on an unnamed social-media platform, where the men had joined a group “discussing the violent overthrow of certain government and law-enforcement components.” 

Established, well-funded nonprofits such as the Anti-Defamation League and the Southern Poverty Law Center also monitor online hate. Overseas, the London-based Moonshot CVE specializes in tracking online extremism in dozens of languages and staging interventions to prevent people from getting drawn into extremism.

But unlike the case in many European countries, the United States’ speech protections and criminal laws make it more difficult to investigate or arrest someone for spewing threatening and dangerous rhetoric online. 

“We have extremely few tools to appropriately deal with people who express violent hateful things unless they’re violent toward particularly public figures, or violent with an extremely specific place and time,” says Rachel Kleinfeld, a senior fellow in Democracy, Conflict, and Governance at the Carnegie Endowment for International Peace.

“We fight the police. The police are absolutely a white supremacist institution” — Christian Exoo

For Christian Exoo, the problem isn’t whether law enforcement has enough tools to stop white supremacism and fascism; it’s that the FBI and the police have a long history of actively enforcing the country’s racist hierarchy and suppressing progressive activists and movements. 

Exoo told me he saw Deplatform Hate as a part of a decades-long anti-fascist and anti-racist movement to fight fascism and white supremacy in America. He traced the fight back to the 1980s, when leftists and socialists in places like Minneapolis used dossiers, public flyers, and sometimes physical violence to keep neo-Nazis from infiltrating the punk-music scene. 

And like those original anti-fascists, researchers and activists like Exoo see themselves in a struggle against both fascism and governmental institutions. Unlike the Anti-Defamation League or other more mainstream groups, he and his colleagues won’t work with law enforcement. “We fight the police,” Exoo told me. “The police are absolutely a white supremacist institution.”

It’s the combination of a political ideology that sees the police as a part of the problem and a fluency with online culture that puts Deplatform Hate at the leading edge of the fight against fascism and hate. 

“It makes perfect sense,” says Devin Burghart at Institute for Research and Education on Human Rights, “to have folks who are peers of the next generation of white nationalists do their best to disrupt and expose the work that they’re doing.”

WHEN I ASKED members of Deplatform Hate how they learned the skills needed to infiltrate and expose the online far right, many of them brought up Exoo. While there is no rigid structure among the loose network of undercover anti-fascists, Exoo is probably the closest thing to a leader. 

Part den mother, part organizer, he connects up-and-comers with others in the movement and runs workshops several times a month on how to use open-source intelligence (OSINT) — social media, newspaper databases, forums, property records, tax records, voter records, DMV records, and leaked or breached data sets — to run investigations and identify individuals.

On a Friday afternoon in December, I joined one of Exoo’s workshops. A pale-skinned 39-year-old with a flashy haircut (long on top and buzzed short on the sides; yes, he sees the irony), he wore a black T-shirt, chunky Buddy Holly glasses, and a pair of stereo headphones. Talking into a Larry King-style microphone, he had a polite demeanor — he says, “That’s so sweet of you,” and, “You’re so kind,” unnervingly often — which made for a jarring effect when it was interspersed with a detailed explanation of how to credibly impersonate a neo-Nazi.

Exoo grew up in upstate New York. His dad was a professor and his mom a lawyer who served on the board of the local rape crisis center. He remembers his mom battling with the district attorney’s office and the local police department for doing little to investigate and prevent sexual assault. 

The experience radicalized him, he says, instilling in him a distrust of law enforcement and a belief in the need for community defense. He dabbled in freelance journalism after college, but he found the work stifling, and editors weren’t interested in his pitches about the alt-right or white nationalism. Around 2014 or 2015, he told me, “I just said, ‘Well, fuck it,’ and went full vigilante.”

Most of the people in Deplatform Hate started out in journalism, labor organizing, or local politics. None had gone out in search of neo-Nazis; more often than not, the hate had come to them — to their communities, their schools, their workplaces. 

“A lot of people got into this work because of their farmers market Nazi,” Molly Conger told me. (In 2019, a controversy embroiled Bloomington, Indiana, after a farmers market vendor was revealed to have made frequent posts on a message board for Identity Evropa, a white-nationalist group. 

The vendor denied being a white supremacist and called herself an “identitarian,” which the Anti-Defamation League links to white nationalism in the U.S.) “A lot of people didn’t experience racial terror until the Trump administration, but that doesn’t mean that it didn’t exist. It’s foolish to think that.”

In 2017, Conger was living in Charlottesville when Identity Evropa, neo-Nazis, the Proud Boys, and other hate groups converged on the city for the Unite the Right rally. 

A friend of Conger’s named Emily Gorcenski, who had attended the rally as a counterprotester, was attacked and later filed criminal charges against Christopher Cantwell, a far-right figure better known as the “crying Nazi,” for assaulting her with pepper spray; Cantwell responded by suing Gorcenski and another counterprotester for malicious prosecution, demanding $1 million in damages. 

Conger began listening to Cantwell’s podcast, noting discrepancies between what he was saying on his show and what he was saying in court, and sending them to her friend’s lawyer. 

Cantwell later pleaded guilty to assaulting Gorcenski and was banned from Virginia for five years. (Last year, a different jury found him guilty of extortion and threatening to rape a rival white nationalist’s wife. He faces up to 22 years in prison and will be sentenced this month.)

CHARLOTTESVILLE, USA - AUGUST 11: Neo Nazis, Alt-Right, and White Supremacists march through the University of Virginia Campus with torches in Charlottesville, Va., USA on August 11, 2017. (Photo by Samuel Corum/Anadolu Agency/Getty Images)


Having glimpsed this online underworld of hate, Conger couldn’t look away. Over the next few years, she taught herself open-source intelligence through trial and error. If she could do it over, she told me, she would have done a better job of hiding her personal information online, including her name and where she lives. 

She has been doxxed, received death threats, and had at least one neighbor post harassing flyers outside her house. Abner Hauge, the editor of the news outlet Left Coast Right Watch, is another member of Deplatform Hate. (Hauge uses the gender-neutral pronouns they and them.) 

Hauge recently told me that after they’d been doxxed by the far right, they now keep a gun on their desk for self-protection. As Conger put it, “All of us doing this work live with the knowledge that maybe some guy that doesn’t like you is going to show up at our house and kill you.”

Christian Exoo began his workshops so people wouldn’t have to experience the harassment that Conger and Hauge have. In the workshop I joined, Exoo started with the basics of infiltrating far-right online spaces: Use a virtual private network, or VPN, to hide your real IP address, which can be used to identify you. 

Start multiple dummy email accounts and use those accounts to create sock-puppet accounts, or “socks” — fake accounts created for the purpose of going undercover — for whatever platform you’re trying to infiltrate. 

Deplatform Hate members say they’ll have active socks in dozens and sometimes hundreds of online groups, across numerous platforms, at any given time.

Exoo told the workshop that, when going undercover, it’s important to craft personas to inhabit. He said he keeps Post-it notes around his laptop screen to remind him about various details of the “characters” he uses. “A lot of it is just learning the handshakes of these different worlds,” he told me. 

“For the Boogaloo Boys, you have to know their memes and talk about guns. In militias, you have to talk about guns and what fleck patterns you like.” (Flecktarn is a camouflage pattern.) To get into some chat rooms, you had to post photos of yourself with your guns or dressed in tactical gear; for others, you went through an interview process, sometimes by phone, other times in writing. 

These included questions like “What is your race?,” “Define fascism in your own words,” “What are your thoughts on Jews, the alt-right, America, fags, and Trump?” and “Are you ready to become active in real life and not LARP (live-action role play)?”

Those specific questions, Exoo told the workshop, had come during a written interview to join an accelerationist neo-Nazi group based in Eastern Europe. Accelerationists are a strain of neo-Nazis who want to start a global race war. 

Exoo said he decided to infiltrate and disrupt the group to thwart its ongoing efforts to expand into North America. After passing his interview, he said he noticed one user talk about attacking a leftist podcaster named Daniel Harper. Exoo typed the user’s screen name into Google. 

One of the first hits was a public Instagram account in Kansas that belonged to a William Smith. The account featured a selfie of a young man in Army fatigues, the name “SMITH” visible on his uniform. Exoo said he found a William Smith in Kansas’ voting records, compiled all his research, and sent it to Harper. 

A few days later, The New York Times reported the FBI had arrested a 24-year-old Army soldier based at Fort Riley named Jarrett William Smith who had discussed plans to attack a major news network and shared bomb-making instructions. (A judge later sentenced Smith to two-and-a-half years in prison.

There was another person in the accelerationist chats Exoo was after, he told the workshop. It was the group’s leader, a user who went by the screenname Commander. Exoo said he noticed that Commander had said he lived on an Estonian island with a population of about 32,000 people named Saaremaa, but there was little else about him online. 

So Exoo took a different approach, stoking the paranoia often found in far-right groups that the platforms they used — in this case, a chat service called Wire — were insecure and liable to be hacked or accessed by law enforcement. Exoo said he persuaded Commander to move the group to the popular encrypted-messaging app Signal. 

Exoo had a Signal account with a dummy phone number ready to go, but Commander slipped up: He used his actual phone number. Exoo said he traced the phone number to a Snapchat account opened in Commander’s name, plugged the name into Facebook, and got a hit: a profile for a pimple-faced, smiley, 13-year-old who lived in Saaremaa’s largest city.

Exoo said he sent his findings to a European nonprofit that fights anti-Semitism and extremism. As it turned out, the Estonian police were onto Commander as well, having confronted the teenager, who claimed he cut ties with the neo-Nazi group after he was discovered.

The reality, though, is that what Deplatform Hate does isn’t always so clear-cut; and sometimes, the biggest threats aren’t the leaders or the most threatening voices. A few years ago, Molly Conger was tracking an online troll named Daniel McMahon who for months had harassed and stalked her. (McMahon used the pseudonym “Jack Corbin.”) 

One day, Conger noticed McMahon interacting online with someone named Robert Bowers. Bowers also made threatening comments about her, and when Conger checked out Bowers’ profile, she saw posts that were unhinged and repugnant, but no more so than what McMahon and many others were saying.

She didn’t think about Bowers again until about six months later, when the police identified him as the man who had killed 11 people at the Tree of Life synagogue in Pittsburgh in the most violent anti-Semitic attack in U.S. history. “The second they said his name — Robert Bowers, it’s a generic name, there are probably lots of people with the name Robert Bowers — I said, ‘Oh, no, I remember him,’ ” Conger told me.

There was nothing in Bowers’ online footprint, she went on, to lead her to believe he would commit an atrocity. “I’m very familiar with what rises to a threat,” she told me. “I looked at (Bowers’ posts) and thought, ‘This guy is pretty crazy.’ But there was nothing he said that was a crime.” 

And as she would later learn, the FBI had also been monitoring Daniel McMahon, the same troll Conger was tracking, and so the feds would’ve seen the same exchanges between McMahon and Bowers that she had. (McMahon was later arrested and convicted for making online threats and election interference. His mother told the police that she thought her son had exhibited characteristics of a mass shooter.)

Still, she felt survivor’s guilt that on a rational level made no sense. Her voice caught as she talked about Bowers. 

In our conversations, she often leavened her stories with a dark humor; that humor was now gone. “This is one of those things that knocks the wind out of me every time it comes up,” she said. 

“I’ve thought about those people who died every day for the past two and a half years.”

A FEW DAYS BEFORE CHRISTMAS, Conger received a series of screenshots taken by someone who was monitoring the Proud Boys on Parler. As the conversation around the then-upcoming January 6th rally was heating up, a user named @WeThePeopleWarrior had made threatening statements about former president Barack Obama, doctors administering Covid-19 vaccines, and Chief Justice John Roberts. 

Responding to a comment about Roberts’ lifetime appointment, @WeThePeopleWarrior responded: “Sounds like he needs to have his life shortened.” 

The user went on to write, “If this is our time to refresh the Tree of Liberty then let’s not half ass our sacrifice, the tree of liberty must be refreshed from time to time with the blood of patriots and tyrants.” 

The user posted about having a “badge,” suggesting he worked in law enforcement, but at the same time he also wrote that he was not beholden to “a Governor, Mayor, County Executive, or any other commie that thinks the [sic] rule over the people.”

Conger noticed that @WeThePeopleWarrior’s profile picture was an illustration of an animal styled like Rambo, complete with a red headband and a rifle held across its chest. 

Conger did a reverse-image search of the illustration and found it was a custom piece of art purchased by someone who identified himself as “Grant Tucker.” 

She tracked down a Facebook profile for Tucker, scanned Tucker’s photos and found a shot of him in a sheriff’s uniform. Within minutes, she had connected @WeThePeopleWarrior to a deputy in the Prince William County, Virginia, sheriff’s office named Aaron Hoffman.

On Christmas Day, Conger published her findings on Twitter. 

The next day, the Prince William County sheriff’s office announced it had investigated the matter and terminated Hoffman; a county later told me the investigation had discovered a “direct violation” of the office’s social media policy. (Attempts to reach Hoffman through his former employer were unsuccessful. Hoffman told The Washington Post his Parler account had been hacked and he didn’t write the posts in question. “I am in no way a threat to the public. This disturbs me as well.”)

Deplatform Hate’s members use various tactics to try to prevent possible violence. One is basic information sharing, passing intel about the movement and operations of various far-right groups to racial-justice and anti-fascist organizers working on the ground. 

Another tactic is using an undercover sock to join an online conversation that’s edging toward violence and to attempt to turn down the temperature. Members of the team told me they take different approaches to de-escalation. 

The most effective is stoking paranoia. Often, a carefully worded question or two about the operational security of an upcoming demonstration or the possibility that law enforcement has infiltrated the group is enough to derail the plan. 

“You mostly just want to use their language of military tactics to sow confusion and dissension and paranoia, which is not hard,” Talia Lavin says. “These are very paranoid spaces.”

Public exposure — what Conger did by outing Aaron Hoffman — is the most controversial strategy. Critics call it just another form of doxxing, posting someone’s home address, phone number, or other personal information online in an effort to harass, intimidate, and endanger them. 

Daryle Lamont Jenkins, a longtime anti-racist activist who runs a watchdog group called One People’s Project and who advices Deplatform Hate, says anti-fascists adopted the tactic of exposing their targets in response to anti-abortion activists who published the names and addresses of abortion providers, believing it was only fair to use their own version of this tactic. 

“We never looked at doxxing as a weapon,” he told me. “We called it reporting.”

Exoo and other members of Deplatform Hate told me they never go further than publishing a target’s name and employer, and that they insist on rigorous fact-checking and a peer-review process before releasing any information. 

They’ve studied books on open-source intelligence by experts such as former FBI official Michael Bazzell to avoid exposing the wrong person. 

And while some on the team balked at the word “doxxing” to describe this tactic, they all believed in its usefulness as a way to de-escalate, and also a necessity in situations when law enforcement can’t — or won’t — intervene. 

“When these guys are pseudonymous or anonymous, the rhetoric gets so loud,” he says. 

“By identifying them publicly, we’re letting them know that if they’re going to talk like that, there’s going to be accountability.”

Other anti-extremism experts voice doubts about this approach. “I understand why people want to dox the far right,” says Rachel Kleinfeld of the Carnegie Endowment for International Peace. “I also think that what you’re basically asking for is mob rule — and that’s worse.”

Michael German, a retired FBI special agent who worked undercover to disrupt domestic terrorism and who has been one of the loudest voices demanding the federal government pay more attention to white supremacists and other violent extremists, says citizen outfits such as Deplatform Hate have resorted to protecting themselves out of necessity because law enforcement has failed to do so. “It shouldn’t be that way, but it clearly is,” German says.

Yet the potential pitfalls he sees with these citizen groups comes down to oversight and accountability. “Nobody has deputized these individuals, so whose authority are they working under and who is holding them accountable?” German, now a fellow at the Brennan Center for Justice at New York University, told me. 

“There is one end of the spectrum in which what individuals could do might be extremely dangerous. The best of them on the other end of the spectrum have reputations to uphold, and the reason people follow them is because they’re honest and accurate and accountable to their audiences.”

WHEN MOLLY CONGER RETURNED to her car on January 6th, she had gotten what she’d come for. Her iPhone was filled with photos and videos of the MAGA mob that had stormed the Capitol, faces and hairlines and militia patches and tattoos that could help her later identify members of the crowd, ground truth to use in future investigations. 

But almost immediately she realized how pointless those photos and videos would be. Hundreds of people who participated in the Capitol Riot insurrection had their phones out as they broke through windows, assaulted cops, forced their way into the Capitol, and ransacked the Senate floor; some of these people had even livestreamed the entire thing, documenting their crimes in real time.

In the days afterward, the staggering volume of content to come out of the Capitol insurrection fueled a nationwide dragnet for anyone who had participated. The FBI released hundreds of potential suspect photos on Twitter, asking for the public’s help with identification; a dedicated FBI tip line received more than 50,000 calls in three days. 

Digital forensics specialists used facial recognition technology to ascertain some of the most visible participants in the riot. One woman told the feds her ex-husband was one of the mob members who stormed onto the Senate floor. Everyone, it seemed, was now researching and identifying far-right extremists.

Weirdly, it made little sense for Deplatform Hate to join this spot-the-Proud-Boy pile-on. Instead, the team discussed how the events of January 6th would reshape the larger landscape of online hate. 

On their weekly Zoom call the Monday after the attack, Exoo said that far-right groups like the Oath Keepers militia saw the failed insurrection as a huge show of power and a potent recruiting tool. 

“This was the biggest galvanizing day for the far right since November 3rd, 2016,” Talia Lavin said on the call. A few weeks later, the Department of Homeland Security issued an advisory warning that the country would continue to see incitements or actual violence by “ideologically-motivated violent extremists with objections to the exercise of governmental authority and the presidential transition.”

     Christian Exoo photographed by Damian Battinelli in February 2021 / Damian Battinelli for Rolling Stone


The members of Deplatform Hate were already noticing new mutations in the landscape of far-right hate groups. 

In the groups he monitored, Exoo told me he was seeing more coalescing around hashtags and conspiracy theories — QAnon, Stop the Steal, fear of 5G telecommunications networks — than explicit racial ideologies, as well as a shift away from established conservative media like Fox News toward fringe outlets including InfoWars and One America News Network. 

He had also seen some early organizing under the banner of the Patriot Party, which Donald Trump has reportedly considered launching as a third party.

It used to be that regional affiliations were key to far-right organizing: you joined the militia or KKK chapter closest to where you lived. Now, a culture of celebrity defined the far right. 

Nick Fuentes, a white Christian nationalist and Holocaust denier who calls himself and his followers “groypers,” is one such figure who has amassed a large following. Republicans Marjorie Taylor Greene and Madison Cawthorn also fit the mold of the far-right celebrity except, unlike Fuentes, they now operate from within the halls of Congress. 

Whether on the inside or outside, these people used social media to push extreme ideas and disinformation, and then monetized their audiences via livestreaming platforms and online donations. What they all share, Exoo told me, is a belief in the idea that violence was necessary and more important than democracy. 

“This idea of militant action has absolutely entered the mainstream of the Republican Party,” he says. (Cawthorn, who spoke at the January 6th rally, later said he condemned the attack at the Capitol.)

By late January, the liberal group that helped launch Deplatform Hate had raised more funds, and team members were making plans to ensure its long-term sustainability. 

They put together a list of de-platforming targets that included Nick Fuentes and GiveSendGo, a Christian fundraising site used by the Proud Boys to raise legal defense funds. In the group’s three-month existence, Exoo told me, the team had achieved some notable successes: Zello, the walkie-talkie app, removed more than 2,000 channels “associated with militias and other militarized social movements”; email and fundraising services banned Fuentes from using their services; and PayPal and Venmo said they would block MyMilitia.com from raising money on their platforms.

Is this what success looks like? How do you even measure it? I recently put these questions to Exoo. 

The members of Deplatform Hate told me they believed in the need to chase these extremists wherever they went and get them de-platformed over and over again. 

With most successful action, they told me, the target’s audience size dropped off and rarely recovered, meaning fewer people were exposed to their hateful rhetoric. Still, the work could often feel like a never-ending game of whack-a-mole. 

“Deplatforming by itself is not going to rid the world of hate,” Exoo says. “It’s a tourniquet on an open wound. It’s not a solution in and of itself.”

Exoo allowed that January 6th “was absolutely a failure” on the part of anyone who was trying to stop far-right organizing and violence, which included Deplatform Hate. “If we do our jobs right, the world doesn’t know about it,” he said. 

“The bomb doesn’t go off. The Capitol doesn’t get stormed.” 

But to a much larger extent, he added, “it was a failure for a lot of people who hold power and who just didn’t take far-right organizing seriously.”

I thought back to something Molly Conger had said about her encounter with Robert Bowers. As guilty as she had felt about not paying more attention to Bowers, the experience had also left her more resolute that what she did, that what Deplatform Hate does, was needed more than ever. 

“To me, it’s proof that some of these shit posters are murderers,” she told me. “We have to believe it’s possible to turn the tide.”

The Zillow Fantasy Could End in Rude Awakening

Zillow surfing may be the new great American pastime, but all the world’s lookie-loos haven’t ensured much in the way of profit thus far

By Laura Forman

Zillow become the punchline of a ‘Saturday Night Live’ skit earlier this month./ PHOTO: ROSALIND O'CONNOR/NBC



Buying Zillow Group’s ZG 1.17% stock today is like buying a home at the height of the market. 

Literally.

The online real-estate company posted stellar fourth quarter results on Wednesday, showcasing 63% sequential revenue growth of its automated homeflipping business. 

Meanwhile, revenue for its mortgages segment nearly tripled year-over-year, while its legacy Premier Agent business is still going strong, up 35% on the same basis.

Zillow surfing, hardly new, is having a moment thanks to the Covid-19 pandemic. 

It even became the punchline of a “Saturday Night Live” skit earlier this month, likening it to a 30-year old’s adult fantasy. 

That fantasy appears to have long legs: Zillow is predicting home sales this year will increase 21% with home prices rising double digits.


But with Zillow’s shares up 270% over the last 12 months, including a nearly 18% gain on Thursday following its latest results, a cold shower may be in order. 

Now 15 years old, Zillow has long been at the forefront of real estate’s digital transformation. 

Its recent moves, including its shift into iBuying and its acquisition of virtual home touring company ShowingTime, are sure to solidify its position as the future of online real estate. 

But consumer demand doesn’t automatically mean profit, which has very much been a work in progress for Zillow.

You wouldn’t necessarily know that from reading Zillow’s shareholder letters. 

On Wednesday, Zillow posted quarterly unit economics for its Homes segment, which houses iBuying. 

That data showed Zillow making nearly $22,000 on average per home it bought and subsequently sold before interest expense. 

Even after interest expense, Zillow’s math shows it made positive returns on homes sold for the first quarter ever.

Enter Mike DelPrete, a scholar in residence on real-estate technology at the University of Colorado at Boulder, who on Thursday published a blog post noting iBuyers were “turning obfuscation of profit into an art form.” 

In it, he likens Zillow’s unit economics chart to “an immaculate conception version of iBuying, where transactions magically occur without employees, customers materialize out of thin air, and technology is freely available for all.”


There are further expenses to account for, in other words. 

Lots of them. 

Mr. DelPrete’s math shows that, after all expenses, Zillow in fact lost $72,000 per home. 

That would explain why Zillow also said its Homes business lost money, even on the basis of adjusted earnings before interest, taxes, depreciation and amortization, in the fourth quarter, while it showed positive returns on a per home basis.

The good news is that the company is making progress toward profitability. 

Mr. DelPrete estimates Zillow lost nearly $129,000 in the third quarter per home on the same basis. 

Yet the majority of the improvement appears to be driven by factors largely beyond Zillow’s control. 

Zillow said operational improvements such as renovations, holding and selling costs accounted for roughly a third of the improvement in returns before interest expense from the third to the fourth quarter. 

The majority, then, came from factors such as better-than-expected home price appreciation and faster sales velocity.

Profit in Zillow’s home business increasingly matters to its overall bottom line as the unit expands. 

As of the fourth quarter, revenue for that business is nearly equal to that of its Premier Agent business, which currently makes up the bulk of its high margin internet, media and technology segment, driving overall profitability for the company today.

What, then, will happen if Zillow’s economists are overly optimistic about the future state of the U.S. market? 

Fannie Mae’s January forecast shows new and existing home prices growing by an average of less than 8% in 2021 with total home sales rising by less than 4%—well below Zillow’s estimates.

Encouragingly, Zillow reported total visits to its site were up 19% year-over-year in 2020, suggesting there is still a lot of pent-up desire in the real-estate market. 

The question for investors in 2021 becomes just how much consumers’ fantasies will play out in real life.