Trading firms that apply quantitative strategies are devising strategies to create friction-less trading, thereby reducing transactions costs and boosting alpha. One of these strategies is high-frequency trading.

“To be an effective stat arb shop, you need to think about your cost of trading,” said Mani Mahjouri, chief strategist and chief investment officer at Tradeworx. “Being good at high frequency trading facilitates that part of the process. About four years ago, we made the decision to get into high frequency. We try to maintain a top notch execution platform. Liquidity is like a commodity, with a supply and a demand curve.”

Mahjouri runs the trading business at Tradeworx, including its high-frequency market making business, which does about 1% of U.S. equity market volume in liquidity-providing strategies. Tradeworx also has a statistical arbitrage hedge fund, which has been in operation for 12 years.

“What we’ve discovered is that as time passes, more people are discovering these alpha strategies,” said Mahjouri. “As markets get more liquid, we’ve seen that for any particular alpha, the spread tends to narrow. The scope of the opportunity gets smaller as more people are accessing that alpha, bringing the market closer to equilibrium in an efficient market sense.”

The Securities and Exchange Commission on Tuesday published Part II of its Equity Market Structure literature review, summarizing and discussing papers that address high frequency trading (“HFT”). The papers analyze non-public datasets in which market activity can be attributed to trading accounts that have been identified as engaging in HFT.

A forthcoming Part III of the literature review will address a series of papers that do not have access to datasets in which market activity could be attributed to HFT accounts, but rather use various measures calculable from publicly available market data to proxy for HFT. Such HFT proxies include high message rates, bursts of order cancellations and modifications, high order-to-trade ratios, small trade sizes, and increases in trading speed. These proxies generally are associated with the broader phenomena of algorithmic trading and computer-assisted trading in all their forms.

“While HFT clearly is a large subset of algorithmic trading and computer-assisted trading, the HFT Dataset papers, as well as some recent market events and enforcement proceedings, indicate that other types of automated trading are significant and may be quite difficult to distinguish from HFT in the absence of trading account data that can be used to distinguish different types of market participants,” the SEC said in the literature review.

It always costs money to extract money, whether it’s the bid/ask spread or incurring some information disadvantage by posting a passive order. “Generally, if you want to acquire a position at a given time, you have to give something up in return,” said Mahjouri.

Regulation NMS (National Market Structure), which prevents a trading venue from ignoring, or “trading through,” a better price posted on another trading venue, provided that the quote is accessible electronically, has made the market better for all participants, particularly the end users, but it has also led to a tremendous amount of complexity in the trading paradigm.

“A huge part of our competency is in market microstructure so we’re able to efficiently extract alphas at a lower cost than typical quant funds,” Mahjouri said. “We methodically test our ideas over large data sets, and fine tune our parameters through rigorous back testing. We are able to come up with a set of indicators that stand up under empirical scrutiny.”