4 ms·

I think we can knock them, or at least I personally would not sleep very well if I worked there. RH is making it incredibly easy (and addictive) for naive inves

I think we can knock them, or at least I personally would not sleep very well if I worked there. RH is making it incredibly easy (and addictive) for naive investors to access complex financial instruments and is selling their order flow at a high premium to the most advanced trading firms on the planet. That does not unsettle you?

tptacek 6y ago
No, it doesn't: most retail brokerages "sell" their order flow to execution firms that specialize in actually filling orders. That's approximately how it's supposed to work. This is the problem with Taibbi's technical writing. I'm not a domain expert in trading --- my knowledge of it comes from several years spent doing security assessments of order routers and exchange systems, and talking to lots of people in the field and reading books. But Taibbi simply doesn't know what he's talking about.
- kasey_junk 6y ago
  Robinhood was receiving a premium for their flow compared to the other brokers last I looked. That at least gives me pause in that it means the market feels like being on the other side of those trades is more valuable for some reason & Robinhood has a disincentive to change it if that reason is “Robinhood’s UX makes bad decisions easy.”
  - tptacek 6y ago
    That's a super interesting point! Way subtler than what Taibbi is saying. I heard something, I can't remember from where, about the economics of promoting meme stocks --- which Robinhood is guilty of --- but that might just have involved shorting loan fees.
    - zarkov99 6y ago
      Taibbi made the same point in the article: > These firms pay more significantly more for Robinhood’s order flow than they do for the order flow of other firms: an average of 17% more, according to a Bloomberg analysis.
      - tptacek 6y ago
        Taibbi in this article mostly pushes the conspiracy theory that market makers are trading manipulatively and screwing retail investors on execution, when the opposite thing is mostly true: retail investors get mostly better prices from market makers, who pay for the order flow because it's safer to trade, and thus cheaper, and the savings are being chopped up between the market maker and the brokerage. Taibbi has had years to figure this out, but seems utterly uninterested in how any of it works, despite any number of entertainingly readable Matt Levine posts that will explain it in a matter of paragraphs.
        rmrfstar 6y ago
        > retail investors get mostly better prices from market makers When broker-dealers (BD) say this, check their foot notes. The precise statement is something like "80% of market orders are filled with a price improvement relative the the NBBO at the time of execution." That is a super narrow definition of "better". It says nothing about limit order fill rates, improvement rates, or improvement magnitudes. More importantly, it says nothing about the path of the NBBO midpoint. "Better" needs to be measured relative to the counterfactual where orders are not sent to the BD. The BDs that fill Robinhood orders are all associated with large systematic hedge funds. The hedge funds make directional bets and hold positions overnight, activities that move the NBBO midpoint. That is a seriously suspicious conflict of interest. If the BD is benefiting from hedge fund research about short-term price movements (probably legal if disclosed to hedge fund LPs), it will harm Robinhood clients. If the hedge fund is benefiting from BD research about retail flows (probably illegal), it will harm Robinhood clients. The data only has to cross the corporate boundary once to be harmful. The worst thing about this clear conflict is that the BD and hedge fund are the only ones presently capable of measuring that harm. Until the SEC undertakes a systematic analysis, it is inappropriate to call this a "conspiracy theory". The SEC doesn't even have the data it needs to study this question. Weird.

tptacek 6y ago

No, it doesn't: most retail brokerages "sell" their order flow to execution firms that specialize in actually filling orders. That's approximately how it's supposed to work. This is the problem with Taibbi's technical writing. I'm not a domain expert in trading --- my knowledge of it comes from several years spent doing security assessments of order routers and exchange systems, and talking to lots of people in the field and reading books. But Taibbi simply doesn't know what he's talking about.

kasey_junk 6y ago
Robinhood was receiving a premium for their flow compared to the other brokers last I looked. That at least gives me pause in that it means the market feels like being on the other side of those trades is more valuable for some reason & Robinhood has a disincentive to change it if that reason is “Robinhood’s UX makes bad decisions easy.”
- tptacek 6y ago
  That's a super interesting point! Way subtler than what Taibbi is saying. I heard something, I can't remember from where, about the economics of promoting meme stocks --- which Robinhood is guilty of --- but that might just have involved shorting loan fees.
  - zarkov99 6y ago
    Taibbi made the same point in the article: > These firms pay more significantly more for Robinhood’s order flow than they do for the order flow of other firms: an average of 17% more, according to a Bloomberg analysis.
    - tptacek 6y ago
      Taibbi in this article mostly pushes the conspiracy theory that market makers are trading manipulatively and screwing retail investors on execution, when the opposite thing is mostly true: retail investors get mostly better prices from market makers, who pay for the order flow because it's safer to trade, and thus cheaper, and the savings are being chopped up between the market maker and the brokerage. Taibbi has had years to figure this out, but seems utterly uninterested in how any of it works, despite any number of entertainingly readable Matt Levine posts that will explain it in a matter of paragraphs.
      - rmrfstar 6y ago
        > retail investors get mostly better prices from market makers When broker-dealers (BD) say this, check their foot notes. The precise statement is something like "80% of market orders are filled with a price improvement relative the the NBBO at the time of execution." That is a super narrow definition of "better". It says nothing about limit order fill rates, improvement rates, or improvement magnitudes. More importantly, it says nothing about the path of the NBBO midpoint. "Better" needs to be measured relative to the counterfactual where orders are not sent to the BD. The BDs that fill Robinhood orders are all associated with large systematic hedge funds. The hedge funds make directional bets and hold positions overnight, activities that move the NBBO midpoint. That is a seriously suspicious conflict of interest. If the BD is benefiting from hedge fund research about short-term price movements (probably legal if disclosed to hedge fund LPs), it will harm Robinhood clients. If the hedge fund is benefiting from BD research about retail flows (probably illegal), it will harm Robinhood clients. The data only has to cross the corporate boundary once to be harmful. The worst thing about this clear conflict is that the BD and hedge fund are the only ones presently capable of measuring that harm. Until the SEC undertakes a systematic analysis, it is inappropriate to call this a "conspiracy theory". The SEC doesn't even have the data it needs to study this question. Weird.