Dec 2, 2021 - Economy

A new case against payment for order flow

Illustration of a cowboy riding the wall street bull statue.

Illustration: Megan Robinson/Axios

SEC chair Gary Gensler is looking at the controversial practice of payment for order flow, or PFOF, to determine whether it should be banned. Fresh data from retail brokerage Public, shared first with Axios, adds weight to the case that it should be stopped.

Why it matters: PFOF, a formerly obscure market-structure backwater, got a bad reputation during the meme-stock craziness earlier this year, causing a lot of pointed questions to be asked of the SEC by members of Congress. Now the regulator is considering whether it makes sense, or should be outlawed on the grounds that it represents a conflict of interest for brokers.

Threat level: At risk is about $4 billion per year in PFOF flows from high-frequency traders to brokerages, as well as about $14 billion more in profits for the HFTs, per Larry Tabb, Bloomberg's head of market structure research.

  • What's not at risk is the Robinhood business model, where most of the brokerage app's profits come from options and crypto rather than stock trading.

The big picture: Brokerages are obliged to deliver "best execution" for their clients. When Public moved away from the PFOF model in February and started trading directly on exchanges, it was unclear whether it would manage to significantly improve the prices its customers received.

  • Now the verdict is in. Public COO Stephen Sikes tells Axios that "we’ve found we can deliver better pricing."

By the numbers: Public's preferred measure of execution quality is something called EFQ —measured on a scale from 0 to 100. Lower is better: At 100, all trades get filled at the quoted bid-offer spread, which is also known as NBBO. At 0, all trades get filled at the midpoint between the bid and the offer.

  • Public's EFQ is now 33, which compares to about 45 for Robinhood, per its SEC filings. Public estimates that its EFQ when it used the PFOF system was between 40 and 60.
  • Over the next year or so, says Sikes, the EFQ for Public should come down to very close to 0, as more investors, exchanges, and alternative trading systems compete to trade directly with Public's clients at the market midpoint.

How it works: Retail investors are coveted as stock-trading counterparties, since they are unlikely to have better information than the institutions on the other side of the trade.

  • There is therefore no shortage of investors who would love to be able to trade with retail at the midpoint of NBBO. Two exchanges — NYSE and IEX — have opened retail windows designed for exactly that kind of trading, and Sikes sees that part of his business growing fast.
  • As the execution quality of on-exchange trading continues to improve, it becomes much harder for other brokerages to claim that their PFOF system represents "best execution."

Be smart: For individual investors, the amount of money saved on execution costs is generally going to be tiny — maybe a few pennies per share. Public also retains all the interest income it gets from lending stocks out to short-sellers — plus it asks investors for tips, which can easily exceed the money they gain in the form of better execution.

The bottom line: If PFOF gets banned, zero-commission brokerages aren't going to go away. But the brokers' incentive to encourage as much trading as possible might be reduced.

Go deeper