PFOF Isn’t The Boogeyman

A quick dive into the concept of Payment For Order Flow, how recent press has made it the hot topic, and why getting rid of it might not be in the best interest of neither the brokerage nor the consumer.

Since The Great GameSqueeze Debacle of 2021 (which I’m trademarking) that occurred a few weeks ago, there have been quite a few fear-stoked misconceptions about a variety of aspects of the brokerage finfrastructure stack that powers consumer trading as we know it today. It was a week that saw everything from Elon Musk discussing trading with the Robinhood CEO on Clubhouse, to allegations that Robinhood removed symbols from its platform so that Citadel could bail out a hedge fund called Melvin Capital from a perilous short position that it had taken in the $GME symbol. If it sounds a bit ridiculous to you, well, that’s because it is totally bonkers.

One of the more prominent theories that appears to have arisen is a broad-stroke theory that was stirred up by everyone from Fintech platforms like Public, to a variety of Financial Advisors that have portrayed and industry-standard concept, called Payment For Order Flow (PFOF), as a bad thing.

Public has gone so far as to remove PFOF altogether from its revenue model, stating, “The main criticism of PFOF is that it creates misalignment between the financial aims of the brokerage and the interests of its clients.” In this short essay I’ll try to shed some light on why I believe the issue is more due to transparency versus misalignment, evaluate Public’s decision to stop receiving PFOF altogether, view how other platforms like Wealthfront have navigated this touchy concept, and offer what a potential alternative could have looked like.

“The main criticism of PFOF is that it creates misalignment between the financial aims of the brokerage and the interests of its clients.” — The Public.com Blog

Let’s get the confusing stuff out of the way first;

Per the SEC’s definition, Payment For Order Flow (PFOF) is “a method of transferring some of the trading profits from those firms engaged in market making to the brokers that route customer orders to specialists for execution.”

Why does PFOF exist? Without getting too much in the weeds, it has to do with the brokerage chain of command. Any time a consumer hits ‘Submit’ on a trade, that trade may eventually end up in front of someone called a Designated Market Maker. Market Makers (or MM for short) are paid to do exactly what their job title says; they make markets in stock symbols. They do this by taking positions on the opposite sides of the public. Market Makers don’t take the side of every trade however, as they really do not like adverse selection risk, and as such, they try to avoid it like the plague. Therefore, when a Market Maker takes a position, it’s usually with a novice investor or brokerage as the contraparty versus say, a hedge fund. They want these novice trades so bad that they’ll actually pay the brokerage platforms to be able to trade against and make markets with them — the payment for those orders is PFOF.

So now that we know about PFOF, you may be asking yourself; how much does PFOF pay? You’re looking at roughly fifteen mills, or fifteen hundredths of a penny per share ($0.0015) at the extremely high end that is paid to the broker. In addition to the payment to the broker for the order, an end customer will see a price improvement due to something called “Execution Quality” or EQ — this is an extremely complicated concept that is tremendously difficult to calculate, and I will not dive into in this essay. The important piece to know is that when your broker sends an order to a Market Maker, you are often receiving compensation for that order in the form of a price improvement.

In some cases, brokerages will cut deals with certain market makers to earn a percentage of the ‘spread,’ which is the difference in price between the bid and ask. This can lead to a reduction in EQ and may leave the brokerage susceptible to discipline for not following NBBO standards laid out by Regulation NMS, like Robinhood was charged with by FINRA and the SEC in 2019 and 2020 (respectively).

PFOF — does it happen in other parts of financial services? Of course. When it comes to transactions and money movement, everyone has their hand in the pot.

For example; rebates and transactional fees do not just occur in brokerage. Let’s think about the merchant payments and processing space for a second. Every time you buy a coffee with your credit card, the merchant (let’s call them Bucks of Star Coffee as an homage to my friend Ahmed Siddiqui’s book, “Anatomy of the Swipe”) is charged processor fees, merchant acquirer fees, and network fees. All in all, these various fees can end up being as high as 4–8% all-in depending on the network, method of entry (keyed in vs. swiped), the terminal used to process the transaction and the rails the payment is processed across. While the client may not see the fees, the merchant in this instance, Bucks of Star Coffee, is definitely taking them into account when deciding how to price a cup of their delicious mocha. In addition, the issuing bank also earns a piece of the transaction each time the card is used. This earned income is called an interchange fee, and is typically around .3% for a debit card swipe, and 1.8% for a credit card.

This allows the bank to capture a bit of revenue for the services that they are offering, much like PFOF allows a brokerage to do the same. While ugly and unseemly in nature, these fees help to keep the financial services (and in turn the coffee) flowing.

This can’t be the first time PFOF has been removed from the equation, right?

Contrary to what you may have heard, Public actually isn’t the first brokerage to stop accepting PFOF and go direct to exchanges. Back in April of 2017, Wealthfront brought their brokerage operations in house, which allowed them to negotiate directly with the exchanges. What they did next was really interesting. Instead of saying “no” to payments for order flow, they redirected those payments back to their clients as an even larger price improvement than they would get via normal PFOF activities. So not only are their customers receiving NBBO pricing when Wealthfront goes to make a purchase or rebalance on behalf a customer’s account, the client is also receiving an additional discount which is that payment being directed into the customer’s transaction, and positively affecting their returns.

On the contrary, Public has only indicated that they will be turning off their PFOF activities and negotiating directly with exchanges via their clearing broker, Apex. There has been no literature (at least from what I’ve read) that Public will be directing the PFOF back to the customer in the form of price improvements. Another alternative would have been for Public to do something altruistic with PFOF proceeds, like donating them the benefit of a non-profit.

As far as I can tell, Apex (or their executing brokers) will continue to reap the financial rewards that come from selling order flow, while customers will not see the price improvements associated with PFOF activities. On top of that their customers are now being asked subsidize costs by tipping on each order they execute on the platform, which will affect those customers’ bottom line. That is not to say that Public’s tipping monetization model won’t be successful, as a $1 tip on a share of Tesla will yield ~600x in revenue versus comparable PFOF yield, which will be exponentially higher in a fractional order setting where slices of a share can be sold and tipped on.

I’m simply asking if we’re just picking the lesser of two evils here, and then requesting that the customer pick up the difference.

It begs the question; why just shut it off altogether?

While Payment For Order Flow does not necessarily create a misalignment between customer and brokerage incentives, it can be a source of quite a bit of confusion. This is especially true if your PR team drops the ball and sends your CEO out to the wolves completely unprepared with a reasoning for ‘why’ the issue occurred in the first place, without first getting in front of the mayhem that was unfolding.

In my opinion, the consumer brokerage platforms we see today are marketing powerhouses first, and technology platforms second. That is to say, that mostly everything that these brokerage platforms offer has been done before in some form or another. Believe it or not, Robinhood was not the first brokerage to offer zero commission trading; it was attempted previously in 2006 by a company you’ve likely never heard of called Zecco (which was acquired by Tradeking and then again by Ally).

The consumer brokerage platforms we see today are marketing powerhouses first, and technology platforms second.

What is new this time around, however, is the speed at which information travels. This has caused platforms like Robinhood, Public and the stocks they enable trading in to go viral to no end. With platforms like Reddit and Twitter, information, opinions and yes, scrutiny can be leveled at a speed and scale which was previously reserved for regulators and Jim Cramer of Mad Money fame. While it has suffered some bumps and bruises along the way, Robinhood has largely been able to recover (pending a congressional hearing on February 18th) and will continue to recover from the PR and trading debacle that transpired at the top of the month. The difference is however, that along with other platforms who still engage in PFOF activities, Robinhood will still have a revenue stream that brought them $600m in revenue in 2020 — a large amount of which will allow them to continue providing their $0 commission services to their user base.

NB: Alexander Gillette is an ardent supporter of the democratization of financial services that this wave of brokerage and roboadvisory platforms has brought to the public domain. While some of this essay may seem critical, he has nothing but respect for the individuals who drive the innovation at the platforms named in this essay. :)