Yesterday’s House Financial Services Committee hearing about GameStop and Robinhood wasn’t great. Reuters has a good summary of one its few interesting bits, a scrap between the elected inquisitors and Robinhood CEO Vlad Tenev regarding whether his firm had to raise additional capital to continue operations during the GameStop saga; TechCrunch has reported on the matter since its inception, though learning a little bit more was useful.
Lawmakers also managed to extract an interesting, if expected data point: The company generates more than half of its revenues from payment for order flow (PFOF), a controversial practice in which Robinhood is paid by market makers for executing customer trades.
The Exchange explores startups, markets and money. Read it every morning on Extra Crunch, or get The Exchange newsletter every Saturday.
Those skeptical of PFOF contend that the setup effectively transforms users of neotrading services that monetize their order volume into the product being sold, leaving retail investors susceptible to poor trade execution pricing. Robinhood has gotten into trouble regarding trade pricing in the past. But those who don’t find PFOF to be an inherent issue contend that it allows for low-cost consumer access to the equities markets. That’s fair enough.
Regardless of where you land between — or even on — those two poles is immaterial. PFOF doesn’t appear to be in material danger of being regulated out of existence, and Robinhood’s use of the business model allowed it to generate huge growth in 2020. For perspective, Robinhood’s PFOF revenues rose from a little over $90 million in Q1 2020 to around $220 million in Q4.
How many users did it take to generate those PFOF sums? Tenev also told Congress in his written testimony that Robinhood has more than 13 million “customers,” though we lack clarity on precisely who counts as a customer. But those millions do not monetize equally. Some of those 13 million users are more lucrative than others.
To understand that, let’s start with working to learn what fraction of Robinhood users trade options. Here’s Tenev, via his testimony:
[A]s of the end of 2020, about 13% of Robinhood customers traded basic options contracts (e.g., puts and calls), and only about 2% traded multileg options. Less than 3% of funded accounts were margin-enabled.
This, combined with the fact that Tenev allowed that PFOF incomes comprise the majority of its revenue, comes to an interesting conclusion: A somewhat small fraction of Robinhood’s users are responsible for the vast bulk of its incomes. We can tell that is the case by recalling that when we examine PFOF data, Robinhood’s revenues from trades in S&P 500 stocks are modest, its incomes from trades involving non-S&P 500 stocks a bit larger, and its incomes from options’ order flow comprised the majority of the revenue reported in recent periods.
For example, in the months of October, November and December, TechCrunch calculates that Robinhood’s PFOF revenues were around 67%, 64% and 63% options-derived, respectively.
For reference, 13% of 13 million is 1.69 million. That’s the number of Robinhood users we estimate have traded options. The multileg options number is a far smaller 260,000 users.
Borrowing a term from the casino trade, these whales generate the bulk of the company’s revenue stream.
Don’t think that we are unfairly trying to accuse Robinhood of notable revenue concentration unfairly. It saw tremendous growth last year and the company’s results have proven attractive enough to attract a tower of funding.
All that’s fine, but while its user base is skewed between a majority of lower-value customers and a comparative handful of more profitable users, the company will have some issues to manage.
Consider the following, also from Tenev’s written testimony:
[T]he vast majority of our customers appear to be adopting buy and hold strategies to invest over extended periods. In fact, Robinhood customers purchasing fractional shares typically buy shares of blue chip companies. Only about 2% of customers qualify as pattern day traders.
This means that most users trade infrequently in less-profitable securities, from a PFOF perspective.
For example, PFOF incomes in December from Citadel to Robinhood were worth 20.126 cents per hundred S&P 500 shares (market orders), non-S&P stocks were worth 27.8964 per hundred shares (market orders), while options trades were worth 65.8218 cents per 100 shares (market orders). The more frequently traders use options to approach the stock market, the better Robinhood’s business will be.
We’ve written about Robinhood’s dependence on options-related incomes in the past, but we lacked confirmation that the company’s business model was more than 50% PFOF. Now that we have clarification of the fact — and I would hazard the PFOF’s portion of aggregate Robinhood revenue is more like 65% than 50% — we can better understand just how much Robinhood depends on its active options traders for its revenues.
A little more data will help make the point: Picking Citadel from amongst the venues where Robinhood derived PFOF in December of Q4, the consumer service generated $154,508.58 in PFOF for market orders of S&P 500 stocks, $4,780,624.90 in PFOF for market orders for non-S&P 500 stocks, and $9,360.52 from PFOF related to options market orders.
But Citadel also paid Robinhood $11,304,200.90 from Citadel for options PFOF for marketable limit orders, and $16,661,113.10 from the same source for PFOF nonmarketable limit orders in the same month. You can see how critical options incomes are — that data is just from December.
What’s fun about all of this is that we can now better understand Robinhood’s strength: A business model that allows it to generate hundreds of millions of dollars every quarter while not charging its consumer end users much. And its weakness being what we always bring up when we find revenue concentration at a company. Like TikTok at Fastly or Uber at Twilio. It can get sticky if that key customer, or user base, leaves.
If a company were able to snag just a portion of Robinhood’s options-trading user base, that company would need an influx of users of more pedestrian investing features to make up for the revenue loss.
Anyway, when do we get the damn S-1?
Comment