Instinet Extends Average-Price Trade System to Europe, Hong Kong

Instinet Inc., a broker that owns more than a dozen alternative venues, will expand a platform using so-called volume-weighted average price, in which orders are matched from buyers and sellers who are willing to wait for the market to generate a price over time.

The New York-based company owned by Nomura Holdings Inc. (8604) in Tokyo is introducing its VWAP Cross to equity securities in Europe and Hong Kong, according to Jonathan Kellner, Instinet’s president for U.S. business. The matching system, used in the U.S. in its current form since 2003, pairs buy and sell orders at the average price over a specified period.

Traders use VWAP crosses to try to ensure that some of their buy or sell orders sent before the market opens will be filled at the day’s average price. Those transactions don’t affect the stock during the day, unlike orders sent to exchanges that others can see or that move prices. A drawback is that a trader gives up the possibility of beating the VWAP by buying a lot of shares at lower prices as the market starts to rise.

“A product like this is helped by the expansion of algorithmic trading and that’s been happening in those parts of the world in recent years,” Justin Schack, managing director for market structure analysis at New York-based Rosenblatt Securities Inc., said in a phone interview. “This trend should continue over the next several years.”

Morgan Stanley, ITG

Systems that match orders at the volume-weighted average price, including those run by Morgan Stanley and Investment Technology Group Inc. in New York, are a way for asset managers to limit the effect of volatility on the prices they get. A broker selling at an average level over time avoids second-to- second swings. While the VWAP gauge, used since the late 1980s as a way for institutions to assess brokers’ trading performance, has been supplemented by newer measures, it remains popular among pensions and mutual funds.

Brokers that provide clients with algorithms or computer- driven strategies that chop larger orders into smaller pieces that are sent to exchanges or alternative venues based on a stock’s volume and trading characteristics usually have VWAP tactics. About 15 percent of Instinet’s volume passes through algorithms that seek to achieve a stock’s VWAP, Kellner said.

“When we find there’s going to be meaningful liquidity, we launch the cross” in a new region, Kellner said. “There seems to be a fair amount of interest. There’s a desire for that kind of ability to lay off risk and guarantee VWAP.”

Even if traders don’t use VWAP as their primary execution benchmark, they may use algorithms that seek to meet the gauge for more-liquid stocks so they can concentrate on equities that are harder to buy or sell because they’re less liquid or more volatile, Schack said.

Closing Price

Other benchmarks include the previous day’s closing price and what’s called implementation shortfall, or the price of a company’s shares when the trader receives the order. The shortfall refers to the difference between that level and the average price received for a transaction when it’s completed.

“VWAP’s primary virtue is everybody can understand it,” Wayne Wagner, former founder of Plexus Group Inc., a transaction cost analysis firm that JPMorgan Chase & Co. bought in 2002 and sold to ITG in 2006, said in a telephone interview. “For evaluation purposes, anyone doing the trading is probably using more of an implementation shortfall approach. Plan sponsors like VWAP because it communicates to them and managers can explain it to their clients.”

Instinet’s U.S. VWAP system runs three sessions before the market opens, matching orders for the VWAP over the trading day. Another session focuses on the average price from 11 a.m. until 4 p.m. New York time. The European VWAP session is set for 7:45 a.m. London time and matches orders based on the full-day average prices. The Hong Kong cross will occur at 8:50 a.m. local time.

Japan, Korea

Instinet also runs similar matches based on companies’ VWAP gauge in Japan, Korea and Canada. Its initial system for matching customers’ orders based on the VWAP for the trading day, called MarketMatch, was introduced in the early 1990s, according to James Ross, chief operating officer at Stamford, Connecticut-based AX Trading Group LLC, which operates an alternative venue. He previously worked at Instinet.

The introduction of the matching systems comes as investors in Europe and Asia seek trades away from the main exchange or through venues that allow them to buy or sell without moving the stock price in the public market. European dark pools, or private venues that pair orders without showing them in advance, totaled 4.6 percent of equities value traded in January, up from 2.9 percent a year earlier, Rosenblatt said in a Feb. 27 report.

Dark Pools

Most dark pools match orders within the spread created by the best available bid and offer or at the midpoint of those levels. VWAP is a moving price that incorporates the effect of trading in the market and is typically easier to achieve than buying shares at the offer price when the trader gets the order or selling at the best available bid.

“The biggest hurdle is people have to get around the fact that they can’t cancel their orders during the day,” Kellner said. “But VWAP is a concept everybody gets.”

Instinet’s CBX dark pool in the U.S. averaged 30.8 million shares daily in January while the VWAP Cross traded 4.5 million a day, according to a Feb. 28 report from Rosenblatt Securities. The largest dark pool in the U.S. is Credit Suisse Group AG’s Crossfinder, which averaged 132.5 million shares a day in January, or 1.9 percent of the average daily equities volume.

Transactions in Instinet’s U.S. VWAP Cross averaged 1,092 shares in January, behind only Liquidnet Holdings Inc.’s main dark pool and ITG’s Posit system. Among the dark markets Rosenblatt tracks, Crossfinder has the smallest average trade size of 178 shares. Goldman Sachs Group Inc.’s Sigma X pool averaged 241 per trade in January and Bids Trading LP’s system had an average of 304 shares.

“There’s a challenge to find liquidity,” Kellner said. “People are definitely looking at their execution costs. The certainty of execution also helps.”

To contact the reporter on this story: Nina Mehta in New York at nmehta24@bloomberg.net.

To contact the editor responsible for this story: Nick Baker at nbaker7@bloomberg.net.

Press spacebar to pause and continue. Press esc to stop.

Bloomberg reserves the right to remove comments but is under no obligation to do so, or to explain individual moderation decisions.

Please enable JavaScript to view the comments powered by Disqus.