3 ms·

Rather than asking me the question, you should read the article that summarizes the SEC investigation or the investigation report itself. The article directly a

Rather than asking me the question, you should read the article that summarizes the SEC investigation or the investigation report itself. The article directly addresses the “isn’t PFOF common practice” and “maybe PFOF was beneficial to the customer” arguments and provides fairly detailed evidence that RH did this at a different scale and with more significant detrimental effects on its customers than common practice. Maybe you’ve read both and you disagree with the evidence presented. That’s ok by me: but it would be more efficient for you to state your arguments for why this case didn’t harm customers, rather than asking me to (poorly) litigate the SEC’s case. My only point here is a simple one: that opaque pricing schemes seem like a recipe for anti-competitive behavior.

tptacek 6y ago
I've read both and I'm not sure I know enough about your argument to challenge it directly, which is my critique of it. You're invoking these "isn't PFOF common practice" and "maybe PFOF was beneficial" questions as if they were controversies; of course, they are not. PFOF is practically universal, and it is universal because separating out non-toxic order flow is a win-win. If I had to summon a direct rebuttal to your argument, it would be this: I perceive you to be suggesting that the problem with Robinhood is that it uses PFOF, which means that customer outcomes are dependent on private deals Robinhood inks with companies like Citadel, which aren't transparent. The problem with that argument is that literally everybody, including, apparently, IBKR, inks private deals with companies like Citadel, which means you can't suggest that something is distinctively wrong with Robinhood for having done so. I perceive you to be suggesting that Robinhood would be more honest and transparent if they simply charged a commission that covered their expenses, rather than doing weird PFOF deals. Again: practically every brokerage does PFOF, because it would be dumb not to. Most brokerages charge commissions, but that doesn't make them transparent; they are still turning a private dial in their back office that trades off customer savings with brokerage profits based on PFOF deals. Robinhood, for what it's worth, is bad, and the badness does have something to do with "no commissions". But the problem isn't shady execution; it's that Robinhood is a gambling site, not an investment site.
- matthewdgreen 6y ago
  I think what you are trying to do is take an argument that has shades of gray and turn it into a black/white argument. This is a very CS thing to do, but I'm not sure it's really a very good way to look at this business situation. From what little I've read here it seems that there is a continuum of different business strategies in which brokerages use their customer base and clout to negotiate PFOF deals that simultaneously (1) produce revenue for the brokerage, and (2) produce better prices for their customers. Since exchanges have finite financial resources, there is some tension between these two goals, which means that at some point brokerages are essentially trading their own interests against their customers' interests. This creates a power and information imbalance that can probably harm customers if the exchange behaves in ways that maximize (1) at the cost of (2). (RH is alleged to have pushed this balance much farther than common industry practice.) But that does not mean all PFOF deals are definitionally bad. The question in this case is not "is PFOF always bad", but rather "is there some point at which brokerage-revenue-maximizing PFOF negotiations actually harm customers by overcharging them compared to a more transparent system." The answer to that question is simple: almost certainly there does exist such a point and when a brokerage reaches that point, customers probably won't be able to tell that the point has been reached. Moreover, brokerages with no commissions seem clearly incentivized to optimize towards or beyond that point. Maybe you could argue that RH hasn't reached that point in this specific case, but this does not seem to be your argument. The question then is: what do you do about it? Should all PFOF be disallowed? You seem to be making a compelling argument that this is not a good solution. Should there be some kind of standard for how much that brokerage/customer balance should be allowed to stretch? Plenty of business practices are regulated along lines like this, where the regulation doesn't outright ban the practice but instead provides some guidelines and standard practices to prevent abuse of weaker parties like customers. I'm not here to give an answer: I'm just pointing out that eventually if you allow brokerages to keep optimizing for revenue, customers will be harmed. Maybe they already have. At that point it's up to the regulators to do something, because that's their job.
  - tptacek 6y ago
    When you suggest that I'm collapsing this down to a binary issue, is that based on information you have about market structure? Have you done a bunch of research on this? Can you be more specific about it? Or is it just a thing you feel happens on message board threads? As I've said: I'm a little bit at a loss as to what your problem with Robinhood even is. They don't charge trading commissions. But they do take an 80/20 share of price improvement payments where other brokerages take 20/80. We can all agree that it's not OK to lie about doing that. But now we know†, and they've been fined. What next? Should they take 20/80? Why not 15/85? Is the right split just what the majority of the other discount brokerages take? Again: I loathe Robinhood. They are a force for evil. But when people come up with weird diagnoses for why they're bad, it leads them to make silly decisions about other discount brokerages, none of whom (a) seriously prioritize their customers (perhaps excepting Vanguard), or (b) operate the way people on message boards believe brokerages operate. † You might be able to argue that knowing the right Google query here, and also giving a shit††, would have enabled you to know this before the SEC announcement, because Robinhood's most important venue is also Schwab's, and you can just compare the per-share rebate they're getting. My confidence level here is not super high. †† (my subtext here is that people don't really give a shit)

tptacek 6y ago

I've read both and I'm not sure I know enough about your argument to challenge it directly, which is my critique of it. You're invoking these "isn't PFOF common practice" and "maybe PFOF was beneficial" questions as if they were controversies; of course, they are not. PFOF is practically universal, and it is universal because separating out non-toxic order flow is a win-win. If I had to summon a direct rebuttal to your argument, it would be this: I perceive you to be suggesting that the problem with Robinhood is that it uses PFOF, which means that customer outcomes are dependent on private deals Robinhood inks with companies like Citadel, which aren't transparent. The problem with that argument is that literally everybody, including, apparently, IBKR, inks private deals with companies like Citadel, which means you can't suggest that something is distinctively wrong with Robinhood for having done so. I perceive you to be suggesting that Robinhood would be more honest and transparent if they simply charged a commission that covered their expenses, rather than doing weird PFOF deals. Again: practically every brokerage does PFOF, because it would be dumb not to. Most brokerages charge commissions, but that doesn't make them transparent; they are still turning a private dial in their back office that trades off customer savings with brokerage profits based on PFOF deals. Robinhood, for what it's worth, is bad, and the badness does have something to do with "no commissions". But the problem isn't shady execution; it's that Robinhood is a gambling site, not an investment site.

matthewdgreen 6y ago
I think what you are trying to do is take an argument that has shades of gray and turn it into a black/white argument. This is a very CS thing to do, but I'm not sure it's really a very good way to look at this business situation. From what little I've read here it seems that there is a continuum of different business strategies in which brokerages use their customer base and clout to negotiate PFOF deals that simultaneously (1) produce revenue for the brokerage, and (2) produce better prices for their customers. Since exchanges have finite financial resources, there is some tension between these two goals, which means that at some point brokerages are essentially trading their own interests against their customers' interests. This creates a power and information imbalance that can probably harm customers if the exchange behaves in ways that maximize (1) at the cost of (2). (RH is alleged to have pushed this balance much farther than common industry practice.) But that does not mean all PFOF deals are definitionally bad. The question in this case is not "is PFOF always bad", but rather "is there some point at which brokerage-revenue-maximizing PFOF negotiations actually harm customers by overcharging them compared to a more transparent system." The answer to that question is simple: almost certainly there does exist such a point and when a brokerage reaches that point, customers probably won't be able to tell that the point has been reached. Moreover, brokerages with no commissions seem clearly incentivized to optimize towards or beyond that point. Maybe you could argue that RH hasn't reached that point in this specific case, but this does not seem to be your argument. The question then is: what do you do about it? Should all PFOF be disallowed? You seem to be making a compelling argument that this is not a good solution. Should there be some kind of standard for how much that brokerage/customer balance should be allowed to stretch? Plenty of business practices are regulated along lines like this, where the regulation doesn't outright ban the practice but instead provides some guidelines and standard practices to prevent abuse of weaker parties like customers. I'm not here to give an answer: I'm just pointing out that eventually if you allow brokerages to keep optimizing for revenue, customers will be harmed. Maybe they already have. At that point it's up to the regulators to do something, because that's their job.
- tptacek 6y ago
  When you suggest that I'm collapsing this down to a binary issue, is that based on information you have about market structure? Have you done a bunch of research on this? Can you be more specific about it? Or is it just a thing you feel happens on message board threads? As I've said: I'm a little bit at a loss as to what your problem with Robinhood even is. They don't charge trading commissions. But they do take an 80/20 share of price improvement payments where other brokerages take 20/80. We can all agree that it's not OK to lie about doing that. But now we know†, and they've been fined. What next? Should they take 20/80? Why not 15/85? Is the right split just what the majority of the other discount brokerages take? Again: I loathe Robinhood. They are a force for evil. But when people come up with weird diagnoses for why they're bad, it leads them to make silly decisions about other discount brokerages, none of whom (a) seriously prioritize their customers (perhaps excepting Vanguard), or (b) operate the way people on message boards believe brokerages operate. † You might be able to argue that knowing the right Google query here, and also giving a shit††, would have enabled you to know this before the SEC announcement, because Robinhood's most important venue is also Schwab's, and you can just compare the per-share rebate they're getting. My confidence level here is not super high. †† (my subtext here is that people don't really give a shit)