Imagine being on an airplane and the pilot makes the following announcement: “Ladies and gentlemen, due to increased competition, higher demands by our best customers, and complicated regulations, we can’t guarantee your safety. Thank you for choosing to fly with us!”
In the airline industry, everyone expects 99.99 percent reliability. They always have and always will. The securities industry once expected the same, too. Today, however, it has much lower expectations and accepts compromised performance.
Why the change? And why should the markets not be surprised by the latest evidence of trading running amok, such as the announcement earlier this month by BATS Global Markets Inc. that it made repeated, though minor, money-losing errors executing customer-trade orders? This isn’t all that shocking after Knight Capital Group Inc.’s erroneous trades almost bankrupted the company, and Nasdaq OMX Group Inc.’s mishandling of Facebook Inc.’s initial public offering undermined the trust that investors have in the IPO process.
The equation is really quite simple. Increased complexity, client concentration and demands for efficiency have led to something less than near-perfect reliability. Unless the industry is prepared to alter those inputs, we shouldn’t be surprised to see glitches, violations and breakdowns soar in the years ahead.
Increased complexity means that data systems aren’t just harder to code, but also to maintain. Computer code needs constant feeding and care. Think of your own personal computer. Have you ever loaded a new program, only to realize that an old one stops working as soon as the new one goes live? This isn’t uncommon in big systems, either. I remember the first software company I worked with. It had limited version control and just kept adding modules and features without a stable code base. The system, as it got bigger, eventually crashed.
Complexity also applies to oversight. Each new equity-order type requires new training of regulators and new systems to monitor trading. The proliferation of order types, each designed to fulfill unique investment strategies, introduces additional rules and procedures.
One way to address this might be to require that before regulators approve a new order type, they ensure that proper tracking systems are in place. Then the exchange seeking approval should have to pay for the added cost. The bottom line needs to be clear: Every new order type has the potential to degrade the integrity of our markets.
Client concentration -- the rising percentage of trading revenue derived from a subset of customers -- is an important driver of structural complexity. Because of consolidation, the exchanges already violate the 80-20 rule, meaning 80 percent of their business comes from 20 percent of their customers.
Many of the largest customers -- to a great extent, high-frequency traders -- are the ones demanding new order types. It is a standard quid pro quo: The traders send the exchanges large volumes of orders and want something in return, such as rebates. It also is understandable that the exchanges try to accommodate their demands -- exchanges are a business, after all, and want to make a profit. But it is wrong to think that giving priority to big customers doesn’t affect the quality of the market.
Finally, the drive for efficiency has had the paradoxical result of making markets less stable. As clients demand greater speed and execution of more complicated orders, they also want it to come at a lower price. Before technology changed the nature of trading, cost-cutting initiatives at U.S. stock exchanges were criticized for putting reliability at risk. That argument no longer holds. There is less fat to cut today, and newer systems are lean from the start. Lowering costs at this point means compromising reliability.
So how does a business, with limited growth potential, confronting increased competition, greater regulation and client demand for lower costs, find the resources to invest more capital in the enterprise? Most investment bankers would call that an untenable situation and recommend strategic alternatives.
Oddly enough, in all my discussions about market structure, not once do I recall an investor, trading firm or other participant demanding greater reliability. Dependability was something that exchanges once made a priority and that investors took for granted.
It isn’t clear that exchanges, traders or clients are ready to change the conversation. Complexity comes at a price, and reliability seems to be the cost that the market is being asked to bear. So let’s not be surprised by the BATS news or the other high-profile technology snafus. Rather, let’s only be surprised if these types of announcements don’t occur more often and become more costly.
(Amy Butte is the former chief financial officer of the New York Stock Exchange. She is a former equity-research analyst covering the financial-services industry. The opinions expressed are her own.)
Click on “Send Comment” in sidebar display to send a letter to the editor.
To contact the writer of this article: Amy Butte at firstname.lastname@example.org
To contact the editor responsible for this article: James Greiff at email@example.com