Hacker News Reader

Comment by anonu | Hacker News Reader

Typical internalizer trade is:

1. RH user wants to buy 1000 shares of XYZ. Offer price is $10.00

2. RH forwards the full order to their execution venue partner. They get paid (assuming SeekingAlpha story is true) $260/$1mm traded, or $2.60.

3. Executor takes the order and immediately sends 900 shares to the market, lifting the offers. Now best offer is $10.10

4. Executor facilitates the tail of the order, 100 shares, at $10.10. So they are short -100 shares here - or they can give a de minimis price improvement (like 10.0985) to meet their contractual obligations.

5. Temporary market impact attenuates and offer mean reverts to $10.05

6. Executor still has risk on their books - but their paper mark-to-market profit is $5.00.

7. Risk can be mitigated by waiting for crossing order or waiting for more favorable offers to close out the short.

So roughly speaking they paid $2.60 to get the order, but made $5 on it...

View on HN

I'm having a hard time believing this is actually a common case. The flow you've laid out almost certainly breaks SEC rules - Credit Suisse's (now closed) retail execution unit was fined for this exact trade less than a month ago (https://www.sec.gov/news/press-release/2018-224). If this is how internalizers make money, they should all be in existential crisis right now. As far as I can tell, they are not.

The SEC cracks down hard on ATSs. But it's rarely for what they do. It's for what they didn't do ... Which is be transparent and disclose. This is the case in the link and something I've observed with all these SEC actions brought against ATSs.

The flow I explained in one way these guys make money. It's certainly exaggerated. At the one place I worked we ran a $0.5bn book of this stuff. You don't always get chances to get out. So risk management is key. Your book turns over every few days.

Sure, most SEC complaints are for lack of transparency, because they don't have laws against a lot of the stuff they want to stop. But they don't go around prosecuting for lack of transparency of good things. If they fined someone for it, it's probably because they don't approve of the thing they did, which is a good sign they're going to fine others that do it.

And it's not like a lot of internalizers are going around with prospectuses saying "we're gonna purposefully move markets to give your customers worse prices", so the SEC isn't going to have much trouble fining them if they decide to (assuming this is actually happening all the time).

That's how you're saying the internalizer made money on it. Obviously, they make money on the order flow. The question is how they cost you anything. What could you have done differently to capture the $2.40 in hypothetical profits here?

He's saying if you spent a few weeks getting hooked up to the exchange directly, you possibly could have gotten your order filled for $10 a share

So, first off, I am not saying internalizers are all evil and that pfof is bad. That $2.40 your retail trade earned the internalizer is the price you pay for convenience and immediacy. You let someone else handle the order and they do what they want with it - within certain guidelines.

You could have saved the hypothetical $2.40 had you traded differently, albeit slower, and caused less market impact with your order. Of course, trading slower comes with its own slippage risk or opportunity cost that the market moves out of your favor. But it could also move in your favor... However, if you let a greedy HFT use your order to impact the market - it will always be out of your favor.

For a small retail guy, you are powerless unless there is a RobinHood 2.0 that decides to offer comm-free trades + does not sell your orders + makes money doing something else ... stock loan maybe?

EDIT: Also the HFTs hate limit orders because it reduces the amount they can internalize. They also generally have a contractual obligation to always execute your order. So you can always use a limit at or inside the offer if youre buying, and potentially save on that cost.

I don't understand the outline of your argument. If what you're saying is true, then you can non-hypothetically save that $2.40 by... trading slower and with less impact on the market.

This is why I asked for specificity. Yes, we agree, internalizers pay for order flow because they can make money on it. The question is: can you make that money on your own order flow, or is the allocation of orders to internalizers Pareto-optimal?

Internalizers want the entire order. If RH splits a order across multiple internalizers they will not be happy. So RH allocates in a Pareto-optimal way. As the order creator, you cannot earn that $2.40, you can only save by getting a better price.

Upthread, you said "It's never free. You're always paying somewhere." But it turns out to be pretty hard to even describe how Robinhood customers are "paying" for their order routing.

There is no market out there that lets you trade slower in return for providing a smaller margin for market makers.

It is possible to build one, but it would look so different from existing markets that it would be an uphill battle to get people to adopt it.

Not sure I follow exactly what you're proposing. But it sounds like this was IEX's approach with their self-imposed speed bump.

IEX's speed bump in the end simply allowed their special internal dpeg order type to pull slightly back when they saw adverse movements incoming, for many normal trades it's essentially an exchange that acts like it's a little farther away (longer latency) than it actually is with some dark order types which move weird.

This trade is theoretically possible, but customers don't frequently trade enough volume to move all the lit markets that much, and it's extremely dubious from a legal perspective anyways.

Reply on news.ycombinator.com