Payment for Order Flow

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
willyd123
Posts: 141
Joined: Mon Feb 19, 2018 7:23 am

Payment for Order Flow

Post by willyd123 » Fri Oct 11, 2019 8:57 am

BHs -

I watched an interview with Kathleen Murphy of Fidelity Investments on CNBC this morning where she was discussing Fidelity's move to zero commissions. She mentioned that in addition to this move, Fidelity also pays higher interest on cash (a topic that is regularly posted about in this site) AND "best execution". In terms of best execution, what she said was that Fidelity "does NOT take payment for order flow on equity orders" and went on to say that many of Fidelity's competitors do take payment to the tune of hundreds of millions of dollars and to the detriment of their customers.

Is there an objective source that shows which of the brokerages actually does take payment for order flow or maybe more simply, which firm has the "best execution"?

arf30
Posts: 617
Joined: Sat Dec 28, 2013 11:55 am

Re: Payment for Order Flow

Post by arf30 » Fri Oct 11, 2019 9:07 am

This link shows which brokerages accept payment: https://www.stockbrokers.com/guides/order-execution

From what I've read it's generally not an issue for the average investor as the difference is usually only a few extra cents per trade and things like price improvement have a bigger impact.

afan
Posts: 4407
Joined: Sun Jul 25, 2010 4:01 pm

Re: Payment for Order Flow

Post by afan » Fri Oct 11, 2019 9:28 am

If you buy only highly liquid ETFs the spread will be one cent. Little opportunity for price improvement.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

Startled Cat
Posts: 479
Joined: Thu Apr 03, 2008 8:54 pm

Re: Payment for Order Flow

Post by Startled Cat » Fri Oct 11, 2019 10:20 am

There's not a lot of public information in this area, and it seems like a lot of the rankings of execution quality are really just guesstimates.

People seem to guess that a firm that does not accept payment for order flow is likely to give better execution, because it has no financial incentive to route to a particular market maker, and can choose the one that offers the best pricing for the customer. That makes some degree of sense, but it's difficult to quantify. It's not obvious to me that firms which decline PFOF actually succeed in getting better executions, or if they do, whether the effect is material.

My mental model of how all this works is that market makers want to capture the spread, as long as they are trading against uninformed (retail) orders. Therefore, they are willing to rebate part of the spread in order to get preferential access to that order flow. Retail customers are no worse off than if they traded directly on an exchange, because they would be paying the full spread in that case. The issue is really what portion of the rebate benefits the customer ("price improvement") and how much flows to the broker as a kickback ("payment for order flow"). Often it's some combination of both, and it seems at least conceivable that an order where the broker received payment for order flow would still get a competitive level of price improvement. (More knowledgeable forum participants, please step in and correct me if my outsider understanding is wrong).

The part that doesn't make sense to me is why bids and asks are rounded off to the nearest cent, if market makers are effectively willing to offer tighter spreads. It seems like a market failure that market makers end up offering these rebates and kickbacks to trade against orders instead of simply competing on tightness of spreads. My guess is that the regulation hasn't caught up with the industry realities, and all this complexity basically exists to work around the regulations. It probably works in the market makers' favor because it gives them an opportunity to discriminate between retail and institutional orders, but I would think it would be better for investors to simply allow

BuddyJet
Posts: 253
Joined: Mon Jun 24, 2019 8:56 pm

Re: Payment for Order Flow

Post by BuddyJet » Fri Oct 11, 2019 10:25 am

afan wrote:
Fri Oct 11, 2019 9:28 am
If you buy only highly liquid ETFs the spread will be one cent. Little opportunity for price improvement.
I have had my VTI trades improved by a half cent per share. While it didn’t amount to much, they do improve if at all possible.
People say nothing is impossible. I do nothing all day.

Topic Author
willyd123
Posts: 141
Joined: Mon Feb 19, 2018 7:23 am

Re: Payment for Order Flow

Post by willyd123 » Fri Oct 11, 2019 10:35 am

Thank you. Very helpful.

AlohaJoe
Posts: 4932
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Payment for Order Flow

Post by AlohaJoe » Fri Oct 11, 2019 10:42 am

arf30 wrote:
Fri Oct 11, 2019 9:07 am
From what I've read it's generally not an issue for the average investor as the difference is usually only a few extra cents per trade and things like price improvement have a bigger impact.
I think that afan is right that there's really a big win for price improvement. According to Fidelity's own tool if you make 50 trades a year (i.e. one a week) and you trade 100 shares each trade (which means for something like VTI which is $148 a share, that you are trading $15,000 a week!) then you will save $125 in a year.

So that is: trade $750,000 and save $125, which is 0.01%.

User avatar
Rainier
Posts: 1444
Joined: Thu Jun 14, 2012 5:59 am

Re: Payment for Order Flow

Post by Rainier » Fri Oct 11, 2019 6:39 pm

You have to trade hundreds of thousands of shares / dollars for this to matter.

Fidelity had this stupid graphic last week comparing my commissions paid ytd and price improvement. $20 of commission, 47 cents of price improvement.

Post Reply