Max Bowie: Will the Early Heads-Up Get the Thumbs-Down?

Thomson Reuters recently landed in hot water with the New York Attorney General’s (AGs) office after it was revealed that the vendor routinely gave an early release of the University of Michigan’s monthly consumer sentiment surveys to clients willing to pay extra for it. The AG wasn’t concerned about the five-minute head-start Thomson Reuters clients get when U-Mich makes the data available to ordinary investors on its own website, but rather focused on a two-second early release to high-frequency traders willing to pay extra over Thomson Reuters’ regular client base.

Though the vendor has suspended the practice in agreement with the AG’s office while its investigation continues, the case raises two questions: Is it fair to release information separately to different classes of customers, and if a vendor has paid for the right to exclusively distribute a dataset, does it have the right to decide who gets it, and when?

To the second point, the answer is surely yes: Thomson Reuters no doubt pays a hefty price for the right to distribute the data, as well as Markit’s purchasing managers index (PMI) data—and indeed, has since agreed a deal to exclusively distribute market research firm Ipsos’ monthly Primary Consumer Sentiment Index data through its Datastream products.

The first question is harder to answer, depending on the data and how it can impact investors. There have long been established practices for the equal release of certain types of data—and its distributors are working hard to make their services more equal than others. For example, economic releases from government agencies are kept under lockdown until a specific time, prompting a race in recent years to transmit that information fastest after release, and to translate it into actionable data.

Logic dictates that market forces will apply: If those accessing the data with a two-second delay see less value as a result, they’ll either stump up the extra cash, or look for another indicator, and over time, activity will congregate around that two-second early release, which becomes the default delivery.

There have long been established practices for the equal release of certain types of data—and its distributors are working hard to make their services more equal than others.

Do clients know what they’re missing, and whether they have the option of a faster service? If so, why aren’t the authorities questioning network providers that offer a mix of microwave and fiber data distribution technologies. Aren’t they equally making information available faster to those willing to pay more for the most cutting-edge network? A similar phenomenon exists in the auto industry, where companies sell more powerful, turbocharged engines to those car-buyers willing to spend more. As a service provider, are you obliged to make your very best service available to all clients equally, or merely offer them the option to upgrade?

Brief Boost
The analogy of a turbocharged car is appropriate: Around town, the turbo will give you a brief boost, but the driver would still be constrained by the laws of the road—i.e., speed limits. Yes, your Dodge Viper is faster away from the lights than your heavy-hauling RAM truck (representing block trades), or your Chevy Aveo (retail orders), but let’s be honest, these don’t perform the same function, and aren’t in competition for the same results. To get full advantage, these faster vehicles need to head to a dedicated racetrack where they can compete against one another without disrupting the flow of traffic, or causing an accident.

So perhaps the answer to concerns about the impact of high-frequency traders on other investors isn’t to black-flag them outright, but designate their own race—i.e. create marketplaces segregated from retail flow. Proposed Canadian startup exchange Aequitas says it can do that by using order-type tags assigned to each trade to exclude HFT flow from one of its market models. Other exchanges might consider prescribing how long a security must be held before it can be sold, and migrating those who want to trade faster to a sub-market that reports netted aggregate positions, say, every second to provide transparency into HFT flow without it disrupting the rest of the market.

And if that came to pass, would we care so much about participants in one market getting an advantage over others who—in every sense—are competing in a different market?