Breaking Down The New Yorker's Slanted Robinhood Story

Part 1 in a series about what we see when journalism's gold standards go under the microscope.

May 13, 2021

The New Yorker has nearly 100 years of brand-building behind it, most notably in the form of “we take the facts very seriously”. While I’ve dug into the thinness of that reality before, I want to dissect a single story this time as the first in a series about the gold standards of journalism. [Part 2, about ProPublica, can be read here.]

Today we’re looking at a ~9,000 word New Yorker article that purports to shed light on whether fintech platform Robinhood is more innovative investment app or mobile casino — which is to say whether its net effect is likely towards “democratizing finance” or “encouraging risky behavior” (often presented as mutually exclusive).

In my conversations with journalists and editors over the last year, a frequent contention has been what a correctable mistake is. A wrong date clearly is. But what about a misleading quote? An improbable or inconsistent anecdote? An assumption borne from weak understanding? An action described with insufficient context?

It seems undeniable to me that you can give precise facts and spell all the names right while still fundamentally failing the reader. So I’ve picked seven excerpts from this article to give the sense of what I mean by that, followed by an afterword.

We reward corrections. See something wrong, misleading, or unfair? Say something. It helps. New readers can also learn more about our mission here.

The context you need for this first excerpt is that Robinhood restricted buying on a handful of stocks for a week or so starting this past January 28th, which caused lots of their users to feel like they were being played, perhaps due to Robinhood’s imagined collusion with shady Wall Street types.

Other brokerages placed temporary limits on trading in GameStop, but Robinhood’s ban was one of the most restrictive, and most of the online anger focussed on it. Someone posted on Reddit, “robinhoods whole shtick is that they allow the little guys to enter the market. guess they don’t actually give two shits about the little guy and i hope they lose a ton of users from this.”

What actually happened here is a bit technical and not super relevant for our purposes. (The curious can read more here if they like.) My concern is that we’re given severely negative framing in paragraphs 12 and 13, while the bit about how this isn’t quite what it looks like doesn’t appear until paragraph 55 (with about ~5,800 words in between!) — by which point many readers have certainly bounced or lost the thread.

Compounding this, their explanation there is also followed by two more tinfoil theory quotes (one from a random poker player??) along with an insane editorial take about how all this caused users to realize “they were not Robinhood’s primary customers, as they had believed; they were themselves the product, and they were being sold to firms such as Citadel Securities—Robinhood’s most lucrative customers”.

This all sounds very sinister, but is also rendered meaningless under the slightest interrogation. Brokers by definition are middlemen who have customers on both sides. And if we have to pick the side that matters more to Robinhood, it’s obviously retail customers. To suggest otherwise is either disingenuous or negligently ignorant.

(If you excite 10 million people into wanting to buy a commodified good or service, providers of said good or service will beat a path to your door. But no amount of contracts with said providers will make retail purchasers magically appear. They need to be won, and to be kept. While you can piss off the providers all you want and they’ll still bend to your aggregated purchasing power, pissing off the people providing that power to you will fundamentally break your business. This is 101 stuff.)

II.

Herdman in what follows is a young Robinhood user who was mad about the restrictions mentioned above despite not having had any intention of selling his positions at the time. His anger came from his perception that Robinhood’s actions had arrested the pace at which the stock was climbing toward his target sale price, causing it to nosedive instead. (His target was $1,000, or over 2x the stock’s peak.)

(While it’s true that Robinhood’s actions did stall the stock, it’s also true that the mass gambit being tried by investors like Herdman is what mechanically forced Robinhood and other brokers into said actions. This was just never explained well.)

Herdman thought of his own father, who, when Cody was a boy, had suffered a serious back injury on the job, and was given a hundred-and-twenty-five-thousand-dollar settlement, which he invested in the stock market and lost when the market collapsed in 2008. He had resorted to self-medicating with alcohol and painkillers. […] Herdman started typing [on Reddit]. “This is about more than just money, it’s about fucking these hedge fund managers until they understand what we’ve all gone through because of them,” he wrote. “I am holding to ensure my parents can live comfortable lives at the expense of the assholes that almost cost them their lives. This is for you, Dad.”

Did stock markets haemorrhage value in 2007-2008? Famously so. But the causes there were complex. Was excessive Wall Street greed a part of it? Sure! But so were a lot of other elements — including weak consumer financial education. (The author’s bio mentions two hedge funds before her turn to journalism, so she should know all this.)

Anyway, what does “lost” mean here in terms of that $125k? The natural reading is that it all went up in smoke. But the markets didn’t go to zero in 2008. The major US indices dropped to roughly half their pre-recession highs in the worst days of 2009 before bouncing back to 90%+ by early 2011. So what actually happened? If he got wiped out, he must have made some aggressive bets. But if that’s true, why blame hedge fund managers? Did they personally direct him away from a safe mutual fund?

(Curiously, it says much later on in the article that Herdman Sr. “refused to sell his stocks at depressed values”. So did he even lose anything? It’s wild to me that neither journalist nor editor noticed the massive discrepancy here.)

Anyway, the point: why is it ok to lob a grenade like this just because a source said it? If you’re going to ground part of your article’s frame on a negative quote about a group (even as unsympathetic a group as hedge fund managers), aren’t you responsible to verify the anecdote or at least put it in some minimum context?

We can also see a recursive problem here. Most of the people in this article (including the Herdmans) have very confused ideas of who is to blame for what. The article then presents this as a lamentable fact while contributing to that very problem!

III.

The majority of Robinhood’s revenue comes from trading volume: the more a user trades, the more the company makes. In this way, Barnes [the CEO of a competing app] explained, the interests of Robinhood’s users often conflict with those of the company. The average person builds wealth through long-term investing rather than by rapidly trading in and out of stocks. Barnes described Robinhood as “more financial entertainment than investment management or wealth building. What they’ve created is an incredibly fun, exciting, legal casino in your pocket.”

Some problems here:

Allowing a direct competitor to articulate Robinhood’s supposed interests???
Robinhood is both a slick investing app and a casino. Gamblers and passive investors are subsidizing each other. Each gets what they want, and Robinhood likes the safety of that balance. Their interests are all aligned just fine.
We don’t know Robinhood’s revenue breakdown. We know what the other categories are, but not how much each contributes, or how that mix is changing. (There’s little reason to believe trading fees will be their long-term cash cow.)
It’s unlikely that all that many Robinhood customers are “rapidly trading in and out of stocks”. Most are subject to the pattern day trading rule.

Using the numbers in the article (plus some from here), the mean-average Robinhood user is sitting on $2,692 in winnings, at an average expense of roughly $8.50 a month in trading fees (plus the $5 monthly subscription that ~3% of users are paying, plus any interest on margin loans). So far as casinos go, these winnings and fees are great!

Now, of course, we’re in a bunch of bull markets right now. These will turn eventually and many of those paper gains will never be realized. A player up big at 11pm might leave at 3am broke. But Robinhood’s long-term incentives are towards helping users generate wealth that sits in Robinhood accounts (as margin interest and interchange fees are where the real money is), so it’s not like this is an outcome they’re pushing toward.

IV.

The “practice” mentioned below is payment for order flow, or PFOF. These are the trading fees alluded to in the last section. The gist is that brokerages used to charge you a flat fee per trade (e.g. , $10), while Robinhood charges you $0 but gets a small PFOF kickback from the folks they sell your orders to.

Proponents of the practice including Robinhood, argue that PFOF results in customers getting better prices and faster execution. Some market experts disagree, and studies have been done showing that small investors would be better off without it.

Which experts disagree? What are their interests? Which studies? What did they find exactly? Why aren’t the studies linked?

Payment for order flow is something that it seems only one financial journalist understands well enough to explain. But the gist is that your Robinhood trade is likely to be relatively small and close to random, which are coveted traits in market terms. So the market is happy to offer a small rebate on said trades, which gets split between you (in the form of a price discount) and Robinhood (as a PFOF kickback).

This isn’t new mind you. Lots of brokerages charge PFOF and a per-trade fee. Robinhood’s intuition was that removing the per-trade fee would result in more customers making more trades, and that they could make more total money that way.

Now, it is true (or at least widely assumed) that Robinhood keeps more of this rebate than other brokerages. The norm (per the link above) is something like 80% in discount to you and 20% in kickback to your brokerage. Maybe Robinhood does 50-50? If the criticism here was just “brokerages should disclose these numbers clearly so that investors can shop intelligently”, sure, fine. But just say that narrowly then?

And as far as small investors go, Robinhood keeps from tenths of a cent to a high of 1.3 cents per share. If you’re only trading a few shares, your savings vs. the old model (~$10 per trade) are massive. It’s those trading quite high volumes who might end up worse off compared to a brokerage that charged a fixed fee but passed on more of that rebate.

But again there’s a recursive problem here. Robinhood was shy to explain the role of PFOF (even though it’s fine/great for most users) because of the misinformation and false perceptions there are. Does this reporting help or hurt that?

Vicki Bogan, a behavioral-finance expert at Cornell University, said, “The bigger game is the one that online brokers are playing with the retail investors. How much can they get the retail investors to trade even though it may not be beneficial for the investors to do so?” A Robinhood spokesperson said that attracting users who had previously been excluded from the financial system is a “profoundly positive change,” and that “suggesting otherwise represents an élitist, old way of thinking.”

Another way of understanding Robinhood’s product design is “come for the meme stocks, stay for the index funds”. Is the gambling part bad? Well compared to what? If a million twenty-somethings who usually spend their free money on White Claw and Taco Bell (both great tbh) decide to YOLO $250 into dogecoin instead, are they worse off for it? In the long-run probably not, so long as most of them stay on the platform. Because most investments do pay out, and over time we can intuitively expect people to migrate more of their wealth into safer bets.

If I’m a policy-maker, I’m keen for data to confirm and quantify this, but I’m at least open to the idea that Robinhood really is onboarding millions of first-time investors who were previously not directly benefiting from the massive historical gains of US stock markets. Some of these newcomers will lose money on dumb bets along the way, sure. But if the default alternative is leaving their spare cash in 0.1% savings accounts because there’s too much friction and too little fun involved in setting up a traditional brokerage account elsewhere, a few humbling losses along the way are fine!

(In fairness, the article did include a quote from that competing CEO, who self-interestedly praised Robinhood for encouraging “an entire population who wasn’t buying stocks to buy stocks”. But the article never takes this proposition seriously.)

VI.

This next excerpt is about a young man who, under the impression that his Robinhood bet had a max loss of $10k, encountered a freak combination of UX issues (a misleading display and a misleading email) that made him believe he’d lost ~$730k. He panicked, didn’t get a fast enough reply, and ended his life by early the next morning.

Kearns left the house, biked to a local train crossing, and jumped in front of an oncoming train. Later that day, his parents found a note from him: “If you’re reading this, I am dead. How was a 20 year old with no income able to get assigned almost a million dollars’ worth of leverage?” He added, “The amount of guilt I feel as I commit this is unbearable—I did not want to die.”

This is obviously tragic on a human level. But it’s also an extreme edge case that we have to parse sanely.

As the article admits, Robinhood hadn’t given him that much leverage. His real max loss was what he thought it was. But some bets settle in sequence, and he ended up seeing the debit before the offsetting credit. The amounts in question freaked him out, and he wasn’t able to get an explanation in time to calm his guilt-induced panic.

Robinhood’s response (which the article references but doesn’t link out to or summarize in item form) seemed entirely reasonable to me:

They raised the experience requirements to trade those types of options
They made it clearer how these options worked (to reduce the risk that anyone would overreact if they saw a misleading loss figure before the offset posted)
They made it easier for customers with certain options positions to get live / faster support
They made a token donation of $250k to a suicide prevention fund

The New Yorker’s framing resolves down to a typical preach-y “maybe this happened because tech companies have a ‘move fast and break things’ ethos and don’t care about providing a reasonable level of customer support”. But these are tradeoffs.

While iterating quickly does mean things are more likely to break, it also means things are far more likely to improve
Automating support saves a lot of money and makes $0 trading possible (which is fully open to those who just want to buy and hold safe investments)

I’m sorry for the Kearns family. I’m sure the folks at Robinhood felt terrible. But societies need to measure both sides of a tradeoff. Some people lose money (or think they have) and commit suicide. We’ve invested a lot of resources into responding to that (as we should!). But we haven’t made casinos illegal. And Robinhood isn’t just a casino. They’ve also helped a lot of people safely make money that’s changed their financial situations for the better. It’s reasonable to expect that these windfalls prevent some suicides and reduce human suffering in general. Regulators will probably insist that these things be quantified and evaluated in actuarial terms, and that’s great! They should be! And if the data says that Robinhood is actually causing net human harm, we should do something about it! But that isn’t the data we have today!

VII.

We get something close to a thesis in the second-to-last paragraph:

I was struck by the similarities between the lead-up to the financial crisis and the present moment, with millions of relatively inexperienced people jumping into the stock market, determined to take advantage of the wealth-creation machine.

There’s a way in which this is reasonable. There’s no doubt that stocks and crypto are frothy with enthusiasm right now. And if/when a significant reset comes, sophisticated investors are much more likely to get out faster. We should be explaining these risks.

Even so, 2021 isn’t all that similar to 2008:

Stocks and crypto are far more liquid than houses
A correction to either would be unlikely to cause a recession (which is what created the nasty spiral effects last time)
Some 85% of Robinhood users are facing max losses of their original investment
“Play money” investors are more likely to hold (if not to outright buy the dip), reducing the probability and impact of selloffs

This all in mind, should we ban young people from gambling on meme stocks or crypto just to save them from losing spare cash that they know is at risk?

Similarly, should we ban PFOF and reinstitute $10 trade fees just to make people trade less? If our intended outcome is based on a moralized sense that gambling should be heavily restricted, that’s fine. But if that’s our philosophy, let’s write an opinion column to that end, not a lopsided account purporting to be a feature article.

If I had kids, I’d tell them to put 80% into buy-and-forget index funds and the rest into whatever fun/risky stuff they want. But if they got as high as 50% I’d still be pleased, as that still feels way better than consumption. A dollar saved is a dollar earned, and even if they give half that dollar back on a bad bet they’re still up on the deal.

Robinhood is getting people into investing in general. That’s good, even if some of the investments are bad.

Afterword

Look, I’m not unbiased here. My clients are in tech (including fintech), and I’m generally on the side of technological progress. But Robinhood has been particularly aggressive and their CEO’s comms underwhelming. Good journalism should shine a light and hold power to account, etc. I’ve been happy to criticize Robinhood myself to that end. But I also make sure my readers get enough context to judge usefully. My perpetual concern in this newsletter is how many news outlets don’t do the same.

New York Magazine’s Ben Wallace published a Tech vs. Journalism rundown yesterday that I think made for good reading. He interviewed me for the story (for an angle that ended up not fitting), and I enjoyed my discussion with him. He seemed interested in a good-faith telling of a complex dispute where neither side is totally innocent. (Though you need to read his piece to the end to get the full effect.)

But where I’d go beyond Ben is in establishing exactly how slanted much of this reporting is. And there isn’t even a “breaking news happens fast” excuse for this one. The New Yorker had months to get the facts and framing right, and a gold standard editorial process that supposedly exists to ensure that this happens.

Why didn’t it work?

My sense is that:

It’s expensive to unpack how skewed these stories are, which insulates lots of bad journalism from full feedback loops.
Lots of their readers aren’t really expecting a rounded picture of how modern finance works, what fintech is adding specifically, what the real points of concern are, etc. Many just like juicy narratives that resolve down to “tech bad”.
Journalism has removed its public editors for a reason, and the onus is left on individual journalists and editors to decide how curious and fair they want to be — where the market rewards for being so are modest.

But we need better. And we need it bad. Imagine that this was written about a company where you worked, and that your loved ones read it uncritically? How would that complicate your life? What if you owned equity in the company? What if you were a potential recruit who now decides to go elsewhere after? How is this fair or good?

When I write about newspapers and journalists, I take a lot of precautions. I do a minimum of two drafts/reviews just on precision and fairness. I often invite in proofreaders to challenge my thinking. And I have an open corrections policy that rewards criticism, including extra payments for the equivalent of damages.

Why doesn’t journalism do the same?

While there’s at least one more major piece coming out this week, starting Monday I’ll also be mixing in a new format of shorter takes. These longer ones just stall too easily.

As ever, no one should feel pressured to do a paid subscription. If you believe strongly in this mission and $5/mo is immaterial to you, more funding means more admin help to increase output. But I don’t want money from those on a budget! I’ll try to keep most stories free.

The Save Journalism Committee

Discussion about this post