Deep Dive: Robinhood's Revenue Model & Payment For Order Flow
Hi all, Julie here.
Very excited to announce the launch of Fintech Today’s first podcast next week—an interview series with people in the fintech industry we admire.
We have some incredible guests lined up. Ian interviewed SoFi CEO Anthony Noto, who dives into a plethora of topics—moving from Wall Street to Silicon Valley, fun tidbits from Twitter’s IPO, and a ton about SoFi; everything from getting started, to the Galileo acquisition and newly inked Samsung partnership. I also spoke to Cadre’s Ryan Williams on what it’s like being a black fintech founder.
Sign up for our waitlist here and we’ll let you know when the first episode drops next week.
The first 30 minutes will be freely available, but the full hour long interview is only available for FTT+ subscribers—we'll be sending you all a private RSS feed to access the full interviews. Setup should be easy and we'll be here to help out (Ian's making a step by step guide as we speak.)
On to this week's column.
My head hurts, and it’s because I decided to once again go down the rabbit hole that is the battle of Robinhood and payment for order flow (PFOF). As we all know, Robinhood championed itself and the brokerage fighting for the small guys, trying to give all of us a fighting chance at making money from the stock market. What they didn’t tell users until they were forced to is how much money they were making by selling their order flow to the likes of Citadel and others (something other brokerages do as well, but they don’t market themselves as God’s gift to lower to middle class Americans wanting to trade).
I wrote about this a few times at Bloomberg, but given the absurd rise in retail trading as brokerages all go to $0 commissions, the tragic suicide of one of Robinhood’s users, and new disclosure rules, I wanted to visit it again. Quick note before I do that, this guy right here is arguably the best hire Robinhood has ever made based on all of this (brownie points if you can guess who it is before clicking the link).
The TLDR: there isn’t just one answer. There are a lot of factors such as the type of people using Robinhood, what they trade (aka: lots of options contracts!) and how Robinhood charges for orders that come into play. Here are what some people that are smarter than me had to say, both on and off the record:
Anonymous: "The users are casual traders that are relatively predictable in terms of buy/sell movement. It’s harder to tell on more advanced platforms who knows what they’re doing & who doesn’t, but with Robinhood it’s generally accepted that everyone isn’t doing it as their day job. They also de facto can’t be sophisticated, the Robinhood interface isn’t built for traders. You can’t trade directly from the chart in Robinhood.”
Anonymous 2: ”I used to be on the options market maker side but am now on the other side with a new online brokerage. We've been thinking a lot about best execution in a world of PFOF the past few weeks. Robinhood closes their customers positions in the last hour of expirations. That all gets automated as a ton of market-to-close orders hitting the PFOF desks. The big PFOF desks are at a huge advantage on that sort of order flow because it's mostly small (very low risk) and they are very fast, with a ton of exchanges to route to. They are able to print all those trades much faster than smaller market makers could react. The way that system works in general is each exchange has a bunch of firms making markets. In general, the algos on the market making side have a bid/ask that they show (usually matching the NBBO bid/ask) but then a no show bid/ask which is based on their algo's pricing of fair value for that option, and how much edge they would need to improve on order flow prints. In general that system exists so market makers don't get run over by big orders. But in the case of most orders coming in from Robinhood it's quite small and very little risk. But the smaller market making firms have a hard time being quick enough to see it because they lack the resources of the large PFOF desks. So the large PFOF desks are willing to pay a lot for that order flow because it's very risk free and a lot of it gets automated. So Robinhood order flow is worth more than say large orders from Interactive Brokers.”
Joe Saluzzi, Partner at institutional agency broker Themis Trading: Joe points out that “the method that Robinhood uses to calculate PFOF is different. They get a piece of the spread compared to other brokers that get a flat fee. Most high priced stocks have higher spreads so since Robinhood calculates payment based on fixed % of the spread,they might receive more per share on these stocks.”
Benn Eifert on Twitter made some really good points: “I assume you've seen the options trading interface, where they show only the offer price and not the bid/ask/mid :) The more specific answer of course is that the profitability of order flow determines its pricing power. The product is the user, and high volume of non-toxic (uninformed) order flow in high spread options trades is highly profitable.”
Anonymous 3: “The product encourages ill-informed options trading. The options RH traders engage in are less liquid and have higher spreads = more lucrative for market-makers.”
Alex Steiner, early stage fintech investor Anthemis: In a recent post, which you should read, Alex points out that “Notably, Fidelity is one broker that still has not started selling order flow and receives no revenue from PFOF. This revenue stream has been particularly top-of-mind recently, as Robinhood’s recently-filed 606 Form, filed with the SEC in May, shows nearly $100 million in Q1 2020 revenue from PFOF. Net PFOF payments made to Robinhood increased from $19.4 million in January to over $45 million in March.”
Patrick McKenzie, a Stripe employee with an awesome blog, wrote: “Why Robinhood earns more than the discount brokerages for order flow is readily apparent: their product encourages options trading. This is bad for customers, principally because the supermajority of Robinhood customers do not understand options trading and the risk/reward calculus for them is even more borked than it is for the vast majority of people who engage in it. The options trades they engage in are also less liquid, and so have higher spreads, and so are more lucrative for market-makers, but the thing you should be most concerned with is their product decisions which encourage options trading, not with the price of liquidity on those options trades.
Julie here again. Next week we’ll have part 2 of this, where we explore if this hurts customers and how this practice could shift as everyone goes to no commission trading.