The Rules of the Venue Are the Market

A market is usually described as a place where buyers and sellers meet, which makes it sound like a natural phenomenon. It is a set of rules about who may meet whom, in what order, and on what information.

Those rules are written down. The governing statute sets the frame, the regulations under it fill in the detail, and the exchanges themselves file rulebooks that run to thousands of pages.

Read them and the neutral-sounding vocabulary starts to look like a series of decisions. Best execution has a definition. Priority has a definition. So does the moment at which an order is considered to exist.

Every one of those definitions advantages somebody. That is not a flaw in the drafting. It is what a rule is.

Order routing is the clearest example. An order can be sent to several venues, and the choice is governed by rules about price, speed and cost that trade off against one another. Whichever is weighted most heavily determines who systematically gets filled first.

The participants understand this precisely and invest accordingly. A firm that can act on a rule's consequence a fraction of a second sooner is not exploiting a loophole, it is doing what the rule rewards, which is the thing the rule should be judged on.

The public argument tends to be conducted in a different register. It asks whether a practice is fair, which is a moral question, when the tractable question is what behaviour the current rule pays for.

This matters more than it used to because the venue is now mostly software. A floor with people on it had frictions that were visible and slow, and it was obvious when they favoured somebody. A matching engine's frictions are in a specification.

The specifications are public. They are filed, published and open to comment through the ordinary rulemaking record, and anyone may read them. The difficulty is that reading them requires the expertise that only the regulated firms reliably employ.

So the comment file on a market structure rule is largely a conversation between the venues, the firms that trade on them, and a regulator, about a system the rest of the economy depends on and cannot follow.

The reform that would help most is not a change to any particular rule. It is a requirement that a proposed change state, in plain language, which participants it advantages and by what mechanism, and that the statement be published alongside the codified text.

That would not settle the arguments. It would move them into the open, where a claim that a rule is merely technical could be tested rather than asserted, which is the least a market that calls itself public should offer.

The order types are where the detail becomes decisive and the documentation becomes impenetrable.

A venue may offer dozens of ways to submit an order, each with different priority and different interaction with the book, described in filings written for specialists.

A participant who understands them is not cheating; they are reading the rules. A participant who does not is trading against people who have.

That is the clearest case in finance of a disclosure regime that satisfies its legal obligation while leaving almost everyone uninformed.

Fragmentation is what makes all of this consequential rather than merely technical.

An instrument trading across dozens of venues has no single price at any instant, so best execution becomes a judgement about which venue's price counted.

That judgement is made by the firm routing the order, which is frequently paid by the venue receiving it.