Welcome to the Capital Note, a newsletter about business, finance, and economics. On the menu today: Robinhood’s arbitrage shop, its potential IPO, the end of Bezos’s tenure as Amazon CEO, and a 1920 short squeeze. To sign up for the Capital Note, follow this link.
Robinhood’s Arbitrage Shop
Robinhood’s business model is based on hiding costs. In the pre-Robinhood world, retail traders paid a fixed commission between $5 and $7 per trade. Now that practically all retail brokerages have eliminated commission, retail traders compensate brokerages not with commissions but by overpaying for securities, usually a penny or less per share higher than the best price available. The visible, upfront commission cost is now hidden from customers, most of whom assume they are trading for free.
This process is facilitated by market makers such as Citadel Securities, whose high-frequency-trading desks profit from the spreads between bid and ask prices in stocks and options. When Robinhood sends a purchase order to Citadel Securities, the market maker’s traders sell stocks to Robinhood’s customers at a slightly higher cost than they purchase them for (if they’re doing their job correctly). Citadel’s market-making wing makes a few pennies per trade, but at scale this can be an extremely lucrative business, especially in less-liquid options markets where spreads are larger.
On balance, the price per trade for small-money investors is lower in the Robinhood world, because a fixed commission per trade imposes inordinate costs on small trades. Robinhood makes up for its lower prices by tapping into a larger pool of customers that trades often, because it pays no upfront transaction costs. The Wall Street Journal calculated that a $1,000 trade yields Robinhood roughly 26 cents, nearly 60 times the revenue of traditional brokerages such as Charles Schwab and TD Ameritrade.
This business model means there is practically no built-in limit to Robinhood’s profitability. Like a social-networking platform, it just needs to increase its number of daily-active users, because the marginal cost of an additional customer — a little more server space and a little more funding — is almost zero. And the sign-up process for a brokerage account — in which a new customer must input his bank information, social security number, driver’s license, etc. — creates a sticky user base with a high lifetime value.
So Robinhood is much more like a social network than a financial institution. Just as Facebook leverages its massive user base to sell ads, Robinhood leverages its user base to sell order flow. This is essentially risk-free arbitrage, because the risk embedded in those trades is immediately offloaded to market makers. If Robinhood executed the orders itself, it would have to hold stocks and options on its balance sheet to transact with its customers. Instead, Robinhood lets the army of PhDs at Citadel Securities handle the plumbing. The high-frequency traders minimize risk by executing orders as quickly as possible and hedging their market exposure.
Excluding margin lending and other lines of business, Robinhood’s trading revenue amounts to the difference between what it pays for customers’ orders in server space, advertising, labor costs, and maintenance, and what it sells those orders for. This model means that scale should provide nearly costless benefits to Robinhood, because the difficulties of increased trading volume are borne by market makers. All Robinhood has to do is keep its app up and running (which isn’t exactly easy, but is certainly easier than maintaining the app while also running a market-making operation).
That’s the idea, but what last week revealed is that, much like its customers, Robinhood faces its own hidden costs. Although market makers execute Robinhood’s order flow, the brokerage is still a counterparty to its customers’ transactions. When Robinhood users buy stocks, market makers find someone to sell it to them. Those transactions end up going to the National Securities Clearing Corporation, a clearinghouse where member brokerages settle trades between their customers. If Robinhood customers were net buyers from Fidelity customers, Robinhood transfers funds to and receives stocks from Fidelity two days after the trades were executed.
Clearinghouses are useful because, among other things, they reduce transaction costs and counterparty risk. Rather than having to exchange cash and securities for every single transaction, brokerage firms only transfer the difference between their customers’ purchases and sales, a process called “netting.” If Robinhood users purchase $1 billion in stocks on a given day, the brokerage will typically only have to send a small fraction of that amount in cash to other institutions. And thanks to deposit requirements, clearinghouse members don’t have to worry about the brokerages they do business with failing to meet their obligations. If a brokerage fails, their counterparties know that they will be entitled to its clearinghouse deposits.
Naturally, when a brokerage is at greater risk (e.g., when a critical mass of its customers pile into a meme stock overvalued by approximately 700 percent), its clearinghouse-deposit requirements increase. According to Robinhood’s CEO, the NSCC asked Robinhood to put up $3 billion in cash last week, at least ten times as much as its typical deposit. Unwilling or unable to risk its own cash, Robinhood halted trading in GameStop, AMC, and other stocks favored by retail investors.
On one hand, the trading halt was the one-off result of an insane week in financial markets. Robinhood can be reasonably certain that this kind of thing won’t happen every week. On the other hand, it raises serious questions about Robinhood’s business model. Because volume is the be-all and end-all for Robinhood, its singular focus has been removing trading frictions by charging nothing in cash, letting customers trade before their deposits have gone through, and offering margin to all customers by default.
But in its push to remove trading frictions, Robinhood seems to have incurred its own hidden costs in the form of increased balance-sheet risk. Even if another GameStop–type rally does not occur, the brokerage will need to be more cognizant of the costs of scale. And there is no free lunch, because managing risks by, say, reducing margin lending or being pickier about which customers it allows onto its platform, will eat away at revenue.
Around the Web
Robinhood considering an IPO:
A swift public listing would let the company capitalize on its rapid growth of the past year — and provide a potential windfall for investors who stepped in to support the firm during its crises at hand.
Indeed, people close to Robinhood said the startup plans on moving ahead with an IPO. And alternatively, one person said, the brokerage could pursue a debut via direct listing or a deal with a publicly traded “blank-check” firm, known as a special purpose acquisition company.
I like to imagine Redditors shorting a newly public Robinhood into oblivion and taking it over.
Mr Bezos is indeed following the Microsoft model. In the third quarter, he will switch — as Bill Gates did — from day-to-day running to a position of broader strategic leadership. With little sign of a horse race, Andy Jassy, the current head of Amazon’s cloud computing division, will step up to be the new chief executive. Amazon insisted that Mr Bezos, as executive chairman, would only be involved in what it described as “one-way door” decisions, from which there is no turning back.
Short squeezes are nothing new. From a 1998 issue of Financial History magazine:
Allan A. Ryan was the son of the famous financier William Fortune Ryan. The elder Ryan arrived in the United States penniless in 1886 but was worth $50 million by 1905. The younger Ryan was a bullish trader who profited immensely from the rising market of 1919. One of his largest investments was in the Stutz Motor Car Company of America, Inc., in which he held controlling interest.
Ryan became president of Stutz Motors in 1916. At the beginning of January 1920, Stutz sold for about $100 a share and had risen throughout the month. On Feb. 2, the stock rocketed from $120 to $134 when an organized group of bear raiders began selling the stock short.
Ryan decided to fight the bears by buying all of the Stutz share that came up for sale. He raised the funds needed for his plan by borrowing from banks and wealthy acquaintances, such as Charles Schwab. The price of Stutz reached a low of $113 in early March buy began to rise thereafter, as Ryan’s purchasing strategy began to affect the stock’s price. By mid-month, excited bears were borrowing Stutz stock from their brokers and selling it into the market (to Ryan) in anticipation of its ultimate decline. Ryan began to “feed the bears” by lending shares to brokers, who in term lent them to their clients. In late March, many Stutz shareholders were eager to cash in on the inflated price, and so they sold, unbeknownst to them, to Ryan. By the end of the month when the price reached $391, Ryan owned or held contracts for 110,000 shares. He had cornered the market and the shorts were squeezed.
To sign up for the Capital Note, follow this link.