Order Execution II: Price improvement
We continue our series on order execution. In this post, we take a turn and discuss some arguments in favor of market makers.
A hierarchy of insiders
Recall that in Part I, the situation regarding market makers seemed pretty bleak: We saw a system where brokers and market makers colluded to stiff retail customers. The only thing protecting retailer traders was essentially competition among brokers and market makers (as well as some transparency regulations barely enforced by the SEC). Such protection requires a well-informed retail customer base, which is dubious.
However, there exists an argument that the market maker infrastructure can actually benefit retail investors.
The argument is founded on the premise that there are essentially three tiers of traders, each with less insider knowledge than the next:
- Retail traders,
- Market makers,
- Hedge funds, banks, and insiders.
The argument is that as a result, there should essentially be two exchanges: An actual exchange, in which the market makers and insiders participate, and a pseudo-exchange, in which retailers and market makers participate. This offers retailers a thin layer of protection from the insiders.
Price improvement
In (theoretical) practice, this would work as follows:
Market makers profit from bid-ask spreads and the law of large numbers, rather than insider information. As a result, on exchanges where institutional insiders participate, where orders are large, uneven, and contaminated by “cheaters” with insider knowledge, it behooves them to tread cautiously, which translates into quoting wider spreads. I am reminded of this unsettling scene from Schindler’s List:
In contrast, on “exchanges” (meaning, streams of orders from brokerages) where only retailers participate, there is low risk of contamination by insider information. Further, the law of large numbers takes effect very efficiently due to many independent small participants. Market makers can thus reduce their bid-ask spreads while maintaining profitability, while retailers enjoy slightly better executions. A portion of the price improvement is passed on to the retailer; this is known unimaginatively as price improvement. The remainder is passed on to the broker as the typical source of payment for order flow.
Still, even assuming this system benefits retail traders, the allocation between price improvement and payment for order flow is at the discretion of the broker. As a result, my advice from Part I stands: It’s important to research which broker you use, since some may have better price improvement policies than others. This is particularly important if you place large trades.