A Worthwhile Tradeoff?
According to Frino, his study and Weaver's report call into question the efficiency of the current market structure. While proponents of the status quo say dark pools provide better liquidity and price discovery for hard-to-trade securities, Frino says it's clear that too much of this sort of trading has a harmful impact on the marketplace.
"I'm not averse to dark pools -- there's a role for them," Frino explains. "You can't physically trade a large block in the lit market because there's just insufficient liquidity. You've got to go dark. But I don't see a role for allowing trades that could be done on the lit market to be fragmented in a dark pool."
Joe Gawronski, president and COO of institutional agency broker Rosenblatt Securities, agrees that too much dark liquidity could have a negative impact on lit markets. "In concept I would think that would be the case -- even potentially in larger markets like our own," Gawronski says. "Here, about a third of the trading is done off-exchange. In markets such as Australia and Canada, you're talking about smaller amounts off-exchange, but they are much smaller markets, so the pool of price discovery is smaller. Both this study and the earlier Weaver report suggest a negative impact on spreads from increased non-displayed activity."
But while Frino's study contends that dark pools could also harm price discovery, Gawronski cautions that not much research has been done in that area and dark pools' effect on how prices are determined is still unclear. "Think of the concept of an actual marketplace, like a bazaar in the Middle East or an online aggregator like eBay today," he explains. "People come together and the pricing gets discovered properly when there are a multitude of buyers and sellers interacting. If you have a bunch of different markets all over the place that aren't connected in any way, price discovery is more difficult. That's the benefit of a centralized marketplace."
And that's essentially how the U.S. markets have been structured following the 2005 adoption of Reg NMS, Gawronski adds. "Our displayed markets -- and to some extent dark pools -- are connected through regulation and technology, which mitigates the dangers of price discovery deterioration. That said, that connectivity is far from perfect and, for the less-liquid stocks, for which there actually is more off-exchange activity generally speaking, the issue is likely exacerbated."
A Dissenting View
Renee Colyer, the chief executive of global market research consulting firm Forefactor, questions whether the Canadian marketplace actually fits into Frino's model because of the nation's "upstairs market," where traders communicate directly with each other to negotiate prices. "I honestly don't believe that you can study the impact of dark trading accurately in Canada or any other market that has an upstairs trading market," Colyer argues. "In Canada, a significant percentage of our volume is traded upstairs, especially the blocks."
"Upstairs" trades, which are dark in the sense that they aren't open to the public, account for 20 percent to 30 percent of Canada's trading volume. Only 3 percent of Canadian trading activity occurs in electronic dark pools, according to Forefactor data.
Nevertheless, Frino says, emerging marketplaces such as Australia's will be better off if they restrict the amount of liquidity that's allowed to trade in the dark. That already appears to be the case in Canada, whose dark pool regulations likely played a role in Goldman Sachs' decision to shutter the Sigma X Canada dark pool earlier this year, though dark pools there never attracted much volume to begin with.
"Maybe the fundamental question is whether liquidity actually is improved as a whole if you allow a portion of your order flow that could be done on market to fragment, and I think the answer to that is no," Frino insists. "Weaver's research says no, and my research says no."