“Payment for order flow” just sounds bad. It describes a process where a market maker pays a broker to send it a retail order for shares, in return guaranteeing its execution at, or better than, the current best price. If the market maker is paying the broker for the order, how can it provide the best deal for the investor? The whole thing does sound a bit hokey. That is, until you understand the challenges of being a market maker and how equity orders are carried out. To determine price, equity markets are based on two basic order types. Not buyers and sellers, but liquidity providers — those placing limit orders, which transact only at a specified price or better — and liquidity takers, those that trade at posted prices. The firms that provide the prices seen, for the most part, are professional market makers. Their job is to continuously offer to buy and sell, but also to generate a profit. This is done through buying at the bid and hopefully selling at the offer, to capture a small spread plus any net fees. The problem with providing liquidity in markets that move in microseconds is that you either need to be very fast or very smart. Otherwise you will be picked off by faster, smarter and larger traders. That is why exchangesoffer liquidity-providing rebates as an incentive. US equities also have a minimum spread of one cent mandated by the Securities and Exchange Commission, the market regulator. This means that when spreads are trading at their lower limits, the mandated spread is actually too wide. Market makers would provide liquidity for less but they are capped by the SEC’s minimum spread. Because market makers do not want to be picked off, they quote tightly enough to trade (up to the mandated limit), but widely enough that they do not get hurt when prices move. If market makers could limit their exposure to fast, smart and large orders, they could price those orders better. That means developing a relationship with brokers that send a higher concentration of smaller orders, by paying for that order flow. By separating out those smaller retail orders that are less likely to impact supply and demand, the market maker can execute those orders at tighter spreads than offered on the exchange. This additional profit is passed on to the investor through improved prices and to the broker through payment for order flow. It enables retail brokers to subsidise lower, or zero, commissions. This makes sense. Why should an investor sending an order that has no effect on supply and demand pay the same price as those that do move the market? Retail brokers also receive guarantees from market makers that exchanges do not provide. The latter typically guarantee the order size even if the order is for more liquidity than is in the market at that price. They also “make good” any impact from poor execution. An alternative suggestion is that brokers should send their retail orders directly to the market. This, people argue, would enable retail orders to take advantage of hidden prices that do not show up on the market data feed. But it would also mean that brokers would need to route their orders without taking exchange costs into consideration, which many do not. Trying to find liquidity that is either not displayed or displayed only on direct feeds requires brokers to spend millions of dollars on people, technology and market data to build out a sophisticated trading desk. That would leave brokers locked in to using their own people and capabilities, instead of competitively reallocating their order flow to five or six market makers, not only willing to pay for their order flow but guaranteeing best execution. Recommended Markets Insight Equity investors set up to lose, even at zero commissions It is certainly not a good idea to only route orders to markets that pay the highest rebate or charge the lowest fee, but this is difficult to regulate. Firms are generally governed by best-execution policy, which in the US focuses on obtaining a price for the client order that is at, or better than, the best bid or offer in the market. This metric has traditionally only been used to measure a market order executed against exchange-posted prices. As long as the client order is executed at, or better than, the best price, it really does not matter if the order was executed at the cheapest or the most expensive exchange, as brokers typically pay the exchange fees. The net proceeds to the client are the same. Measuring best execution for limit orders is also challenging. It would mean tracking to see if a broker consistently posted limit orders at exchanges where the market moved against the client after the trade was executed. At what point in time do you measure? The next tick? A second later? There is no clear consensus on this and it complicates not only the business of brokerage but how the regulators enforce best-execution rules. While the world of payment for order flow sounds murky, it really is not — and the alternatives have problems. The system as is provides individual investors and their brokers with surely the best execution in a market dominated by faster, larger and more sophisticated investors. Larry Tabb is founder of the Tabb Group, a capital markets consultancy
There was a rule change inthe last year which resulted in better price executions. It not uncommon to receive price improvement even within a penny spread. Do you know the rule change?
Two simple changes: 1) No sub-pennying 2) Orders are matched every second instead of a millionth of a second = much closer to equal terms
What a romantic journalistic piece (of shit). Let's listen to the lobbyists to tell us how awesome payment for order flow is for the retail crowd. It's not and any sane person can see deduce so rather quickly.
The problem with PFOF is it provides a perverse incentive to MMs. They shouldn't, and I don't think my typical holding time of 1 month on any given option would really be influenced that much by a slightly looser bid/ask because I'm not ordering in bulk. All I, the trader, need to do is be cognizant of this fact and look for well priced options. I don't need the MM's tight quotes. In fact, this is absolutely true. Retails currently have limited price discovery. Watch the DOM on any instrument that is traded heavily (futures are a favorite) and see how despite the mark limit showing 200 by 200 for example, every time the offer gets lifted more orders appear. These are coming from dark pools and make it impossible to play order flow. You can't develop a sense of the players currently in the market at all. The problem is that when a market maker gives PFOF it's screwing up price discovery because the market maker can quote almost whatever they want - which may or may not be near the market price. In this sense, it makes it easy to see how citadel more-or-less frontruns markets by virtue of just being able to pay to be the patriarch of the market price. Price discovery happens through competition between market makers. Imagine you're trying to dump your inventory of oranges and an orange buyer comes in to the farmers market. You quote him a decent price, around your model of the orange market. The guy Citrus-Del across the street hands the guy a $100 bucks and says "i'll take everything you have". Where's the price discovery? The guy just bypassed the entire process of a competitive market! This shill piece only covered the downside of allowing market players to actually play the markets. Where were the criticism of PFOF? Side note: next time use paragraphs.
@gaussin - That was a cut and paste from TABB. FYI you can't see preferenced orders in options. Plus with some things trading on all 16 exchanges your broker doesn't have to play in a payment environment. Almost no genuine institution will participate in a PFOF environment - although they may participate in maker-taker.
Most exchange groups run a least three different models - not all. but most. A payment destination, a maker/taker destination which is usually the best liquidity and a time/price venue and they will all be directly connected via linkage. If you don't like payment route to one of the other two. Some venues overlap. Payment and price improvement are not mutually exclusive. What if you route to a nonpayment venue and NBBO is on a payment venue - in retail your order will ship and interact with payment. Some products trade on all 16 venues - other trade-in monopoly environments. Monopoly products are the remaining options with fees. MM firms make markets on multiple exchanges and about half the MM is three firms. Most large customers will elect to route to a non-payment venue and mark there order not to ship via linkage. Most smart routers will send your order to a PFOF venue. If they are not NBBO they will either match or ship.