Buy-Side Traders: Reassessing the Broker Relationship
In the world of electronic stock trading, traders like Floyd Coleman, cohead trader at AXA Rosenberg, an Orinda, Calif.-based global institutional investment manager, scan as many as 12 trading screens simultaneously - a fact that underscores just how complex it is to move large orders. With liquidity pools often fragmented, Coleman monitors a variety of electronic trading destinations, ranging from Archipelago, Bloomberg TradeBook and INET to crossing networks such as POSIT and Liquidnet.
The scenario illustrates how access to alternative trading technologies is changing the way the buy side executes orders. "We have lots of different ponds we're fishing in so that we can get all the fish that we need," explains Coleman of the quantitative money-management firm, which manages $60 billion in equities.
With a plethora of brokerage firms supplying algorithms and direct-market-access (DMA) tools to their institutional customers, the relationship between the buy and sell sides is undergoing dramatic change. Not only are buy-side traders growing more sophisticated in their use of algorithms - and transaction cost analysis tools - than ever before, they also are becoming more independent of the sell side. "We are able to take the broker out of the middle in many instances," relates Gus Sauter, chief investment officer at Vanguard, the Valley Forge, Pa.-based mutual fund giant that manages $360 billion in equities.
But technology is not the only factor driving the trend. Regulatory scrutiny of soft dollars and the push to unbundle research services from executions is putting pressure on the buy side to figure out how much it is paying for executions, according to Sang Lee, managing partner at Aite Group. "The buy-side traders are under a lot more pressure in terms of trying to fulfill their best-execution obligations," says Lee. "All of these things are coming together to fundamentally change the way the buy-side clients deal with their sell-side counterparts."
According to a TABB Group study, "Institutional Equity Trading in America 2005: A Buy-Side Perspective," released in May, buy-side traders are executing 20 percent of their orders electronically - via ECNs, DMA and crossing networks - and another 11 percent via algorithms. Additionally, the report reveals, buy-side firms currently route 17 percent less order flow via the phone than they did a year ago (the percentage of orders routed to brokers via the phone has dropped from 48 percent a year ago to 31 percent currently) and are projected to route just 20 percent of their order flow via phone by 2007. "Firms are migrating away from sales traders to no- and low-touch channels," comments Adam Sussman, a TABB Group consultant and the author of the report, which is based on interviews with 53 head traders at buy-side firms.
What Have You Done for Me Lately?
So, as buy-side institutions increasingly route their order flow to low- and no-touch electronic channels, do they still need the sell side?
"The buy side is increasingly getting the upper hand," says Sarah Diamond, global leader, financial markets, IBM Business Consulting Services. "The value that the sell side has been traditionally providing - access and insight - is at risk of diminishing," she contends. "Why should a money manager use a trading intermediary when they can effectively go to a direct counterparty themselves?"
Still, most buy-side firms continue to value their relationship with the sell side - at least in terms of handling large block transactions and finding counterparties in illiquid stocks. But others are questioning the future of the buy-side/sell-side relationship.
Some large buy-side firms even are taking on the scale characteristics of the sell side. "If I have 30, 40, 50 traders, I probably have the capacity of the sell side. So what is my need for the sell side?" asked Madison Gulley, executive vice president, director of global equity trading, Fort Lauderdale-based Franklin Templeton Investments, speaking at the TradeTech institutional equity trading conference in April.
In the past, Gulley told attendees, the partnership between the buy and sell sides was "much more subjective" and dominated by research and discounted trading. The buy side passed on the risk and accountability for executions to the sell side. "We just outsourced all of that. The economics of the relationship was questionable," he continued, citing a lack of transparency and adding that Rule 28e (the safe harbor rule that regulates soft dollars) "was not well understood." Today, with the buy side adopting more "metrics, [taking on] venue selection and assuming more risk back on ourselves, the sell side must add distinctive value to stay in the game," Gulley asserted.
With access to algorithmic trading tools, pre- and post-trade analytics, and direct-market-access pipes, the buy side is more self-sufficient. "We are taking greater control of our order flow," says Adam Stewart, head of equity trading at Atlanta-based Trusco Capital, which manages $20 billion in equities out of $70 billion in total assets. Rather than interacting over the phone, "We access markets directly, either though algorithms or DMA providers, and we're working those trades throughout the day," explains Stewart. The big difference is that "Historically, you would have seen a sales trader doing that for you."
Not All Algorithms Are Created Equal
One of the most significant factors in the shift in the buy side's relationship with the sell side is the proliferation of algorithms. According to the TABB Group report, algorithms currently capture 11 percent of buy-side order flow, up from 7 percent a year ago, and likely will capture as much as 17 percent by 2007. That amounts to a 34 percent compounded annual growth rate, which is the largest increase for any of the electronic trading methods.
The buy side also is becoming more discriminating in its adoption of algorithms, reports the TABB Group. Instead of selecting algorithms on a trial-and-error basis, the buy side increasingly is using pre- and post-trade analysis tools, as well as its own experience, to select which orders to send through which flavor of algorithm.
"We don't want to see a VWAP strategy," says AXA's Coleman, referring to volume weighted average price, the most popular benchmark used in algorithmic trading. Since there is a lot of similarity in 90 percent of the broker algorithms, Coleman says, he wants the brokers to offer something useful that is customized. "You may start with it out of the box, but everyone's trading style is different; there are idiosyncrasies," he explains. "It should be customized to take everybody's idiosyncrasies into account."Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio