Monday, July 8, 2013

Reuters to Stop Sneak Peek of Consumer-Confidence Data

By Tim Dulaney, PhD and Tim Husson, PhD

Thomson Reuters will no longer be offering investors an early look at the results of the University of Michigan consumer-confidence survey, the NY Times reported yesterday and the Wall Street Journal is reporting this morning.  The move was prompted by an investigation by the NY Attorney General and would likely be temporary while the investigation is ongoing.

We talked about this story a bit last week, emphasizing the in depth analysis conducted by Nanex, LLC on high frequency trading data following the early release of consumer-confidence data.  Nanex has demonstrated that even a few seconds' worth of lead time can be enough for high frequency traders to significantly affect major financial markets.

A part of the story we haven't really touched on is the economics of buying and releasing the data.  According to the NY Times, Thomson Reuters pays more than a million dollars annually to the University of Michigan for the exclusive rights to distribute the data early.  About a dozen high-frequency trading firms pay somewhere around $6,000 a month for the privilege of receiving the data at 9:54:58 AM -- a full two seconds before other Reuters clients at 9:55 AM.  This means that a significant portion of the costs to obtain this data is covered by the revenue generated from the high-frequency trading firms wanting that crucial two second advantage.

Nanex's work suggests that markets may move in response to the consumer confidence information before even Thomson Reuters' clients can to react at 9:55 AM.  By the time the rest of the world receives this information at 10:00 AM, a great deal of price discovery has likely already occurred.  It is no wonder that high frequency trading firms pay a premium for a mere two second advantage.

The regulation of high-frequency trading is something that will affect each and every investor since "more than half of all American stock trades are now executed by firms that rely on computer algorithms to execute thousands of orders a second."  We'll continue to follow this story as it develops.

No comments:

Post a Comment

Please keep comments appropriate. Malicious comments or solicitations will be removed.