With commission rates falling, brokerage firms are looking for "other ways" to generate revenue from order flow, and internalization is one of those ways, he added.
Michael Plunkett, president, North America, at New York-based Instinet LLC, said that while brokerage firms internalizing orders might take the other side of a trade to profit from it, "I don't think internalization in and of itself is bad for institutional investors or plan sponsors."
Industry players agree that internalization can be good for investors by allowing them to trade big blocks of stock with little or no market impact. But conflicts arise if the broker uses information about that trade to profit from it.
"When the average trade size is close to 300 shares on the New York Stock Exchange, it's getting tougher and tougher for traders to really trade without leaving an imprint in the marketplace," explained Sang Lee, managing partner at Aite Group, a Boston-based financial services consulting firm.
"Internalization could resolve some of those issues, but at the same time there are negative connotations associated with it," he said. "Are (the brokers) meeting best-execution obligations by internalizing orders? If you see the final price of an execution is not close to the NBBO (national best bid and offer) ... I would have issues."
One reason the price of an internalized order might end up away from the NBBO is if the order is sent to a broker's proprietary trading desk, according to Joseph Wald, chief executive officer of EdgeTrade Inc., New York, an agency brokerage firm that provides trading algorithms and direct market access systems.
"If you're sending your order to a broker, what you want to do is get best execution, get your stock (trade) done," he said. "But a broker-dealer with a (proprietary trading) desk may hold onto that order. This kind of fishing for lower market impact may really be creating a missed opportunity to trade and that's far more expensive for a buy-side trader."