CooperNeff’s Bid For Blind’ Business Quant Shop Using Technology for Growth

Quantitative trader

CooperNeff Group is crossing the Street to make markets in baskets for institutions. Its plan is to use technology to differentiate itself from such entrenched competitors as Deutsche Bank Securities and Lehman Brothers.

CooperNeff, a subsidiary of French banking behemoth BNP Paribas, is one of the largest basket traders on Wall Street, but little known outside of a select circle. Nestled inside a suburban Philadelphia office park, it uses sophisticated software and mathematics to trade for its own account, accessing the market through sister company, BNP Securities.

It recently formed a broker dealer unit called Enhanced Portfolio Strategies (EPS) to bid, on a principal basis, for the multi-security basket trades of large institutions. In so doing it joins a handful of big investment banks which trade proprietarily and also bid for the portfolios of others.

Deutsche Bank and Lehman are reportedly the most aggressive program traders while Salomon Smith Barney, Goldman Sachs and Morgan Stanley Dean Witter make up the second tier. Merrill Lynch also offers the service.

Institutions typically buy and sell large baskets of securities when "rebalancing" portfolios to match changes in a stock index or shifting money between money managers. Seventy percent of the stocks are New York Stock Exchange names; the balance is in Nasdaq issues.

Approximately 70 percent of those trades are done on an agency basis, according to industry sources. The balance is bid blind (the market maker doesn't see the securities unless he wins) on a principal basis. That's the business CooperNeff intends to enter.

To run EPS, CooperNeff signed Mony Rueven and three others from renowned quant shop D.E. Shaw. Shaw folded Rueven's operation and sold certain assets to Knight Trading Group in February. A big player in the blind bidding game, Shaw's presence has been missed for at least a year, buy-side traders say. They hope CooperNeff will fill that void.

Senior managing director Robert Cavallaro oversees EPS. He says the decision to make markets was made out of a need to access liquidity above and beyond that which is available in the open market.

More trading is taking place "upstairs," leaving less liquidity on the NYSE and Nasdaq for CooperNeff. Large baskets of listed stocks are crossed after-hours and overseas while huge quantities of Nasdaq stocks are crossed internally at wholesalers like Knight.

"This order flow never makes it to an established exchange," Cavallaro said. "It's usually printed on a crossing session or after-hours. You can't get it just showing up on the floor of the New York Stock Exchange."

Market structure changes have also turned the environment increasingly hostile to large trades. The switch to trading in sixteenths, for example, has reduced the size of the typical trade as well as the size quoted by market makers and specialists, according to Cavallaro. That makes it much harder to get a large block done. Decimalization will only exacerbate the situation, Cavallaro believes.

By bidding for baskets in the range of $20 million to $500 million, CooperNeff will be able to access order flow from which it is currently excluded. The additional volume helps it to trade for its own account without impacting prices. "Doing more volume for the sake of having more market share would've cannibalized returns," Cavallaro said

On the flip side, institutional clients benefit from CooperNeff's volume, which ranges from 15 million to 30 million shares per day.

Sell-Side Challenge

But crashing the sellside of the basket-trading party is easier said than done. Firms like Goldman and Salomon have the clients as well as large block trading operations and distribution networks. "You need distribution," said Peter Forlenza, head of global portfolio trading at Salomon Smith Barney. "That means a Nasdaq desk that makes markets in a lot of stocks and a block desk. We rely on the expertise of our cash desk to move some of the large blocks we bid on."

Cavallaro concedes the point. "It's an area we are working really hard to develop," he said. "It's definitely important. You want to be able to call up a Fidelity or a Putnam to ask them if they want to trade 500,000 shares." He says CooperNeff is planning to build a distribution network similar to the one BNP Paribas runs in Europe. Some of the necessary ingredients came with the team from D.E. Shaw.

But it's banking on technology to get its foot in the door. To trade its portfolios, CooperNeff uses software to predict returns and assess risk. It will offer its risk management technology to customers. That will help them better understand the risks of their existing portfolios or those they plan to construct, according to Cavallaro. It will also guide them when preparing a basket for bidding.

CooperNeff builds its own risk models, having found off-the-shelf products lacking, Cavallaro said. "Our back-testing has proven that we generated a superior product," he said. (BARRA, Inc. of Berkeley, Calif., is a supplier of such software.)

Its risk models combine security- and sector-specific analysis. "When you look at a portfolio of stocks you have to understand not just the riskiness in the individual name, but the overall exposure of the portfolio to industries," Cavallaro explained.

An examination on a stock-by-stock basis of Internet and utility stocks will, for example, produce one risk profile. But research that takes into account that the two sectors move together would show a riskier portfolio.

Cavallaro notes sector rotation, or the movement of institutional funds in and out of different industry sectors, is higher than ever in today's stock market. That has created volatility which impacts execution quality.

"You'll see days when Nasdaq is down considerably and the Dow is up," Cavallaro said. "Investors are rotating money out of high growth into value or cyclicals. Sectors can move violently."

Changes in the economy also effect CooperNeff's models. "Five years ago the Internet was on no one's radar screen," Cavallaro said. "To adapt, the components important to your risk model must change. The evolution of your technology is crucial."

Cavallaro points out that a trader must also understand the risks inherent in executing lots of varying sizes-whether 10,000 or one million shares. "For each unique security there's a liquidity history and risk associated with having a position of a certain size," he said.

Such variables are often impacted by changes in market microstructure. Decimalization, for example, is expected to change the complexion of market liquidity. As the number of trading increments increases, spreads are likely to decrease so traders will be less willing to quote in size, Cavallaro predicts.

"Since we're doing ongoing maintenance and development in these areas we think it's a smart move to provide tools like this to institutions," Cavallaro said. The firm's risk management software comprises some 20 or 30 metrics such as momentum, value and sector analysis.

The plan is to make the technology available to large institutional clients such as pension and mutual funds in hopes of fostering long-term relationships. CooperNeff would like to avoid trading against savvy quantitative hedge funds that dart in and out of less-liquid stocks. Cavallaro sees relationship building as a five-year process.

CooperNeff's plan does have precedent. TLW Securities, initially a successful proprietary trading firm, now offers its TradeFactory basket trading system on a service bureau basis. Customers are mostly broker dealers either trading for themselves or their customers.

CooperNeff also plans to use the Internet to get closer to clients. Typically, a dealer bidding for a portfolio will send the client a CD-ROM with its pricing model. That gets installed on the client's computer. He inputs his portfolio data and the model produces a rough portrait of the risk inherent in the trade as well as an estimated bid price. Salomon's software is the leader in this area, according to industry insiders. Salomon's Forlenza says it's on 300 desktops worldwide.

The client then faxes print-outs to his brokers of choice that sketch risk characteristics, but do not divulge the actual securities. The bidders see sectors, liquidity characteristics and the size of the trade. From that information they determine their bids on a cents-per-share basis. Prices currently range between five cents and 30 cents.

CooperNeff wants to take the process two steps further. First, it plans to make its pricing software available over the Internet. Second, it plans to embed its positions in the analysis model. If CooperNeff is short and the client is long IBM, for example, the client may save money because the trade will offset a CooperNeff position. By offering the software over the Internet, the customer's analysis will always incorporate CooperNeff's latest positions.

"It's a great way to provide risk management technology to the customer because the risk model works best when it's fed with current data," Cavallaro said. CooperNeff does not hold positions longer than one month. Some are traded in a matter of minutes. The plan will also cut out any systems administrative work at the customer end.

Incorporating a bidder's positions into the model is not a new twist. Salomon once offered the service, but customers balked at having to divulge their positions, according to Forlenza. D.E Shaw, on the other hand, found some success with this modus operandi, sources say.

D.E. Shaw Missed

Those on the buyside also say D.E. Shaw is sorely missed and that a new player in the market is good news. "This is definitely a welcome development," said Greg Rogers, head trader at the $4 billion Philadelphia quant shop Aronson & Partners. "Bidding in the principal package market has not been as aggressive since Shaw left. It's great that those folks from Shaw are back in the game at CooperNeff. Since it's a competitive bidding process, the more people the better."

Forlenza agrees, but cites history. "There's always room for somebody new to the business who will offer good prices," he said. "But the key is consistency. A lot of people are in the business for six months then they lose money and get out." CooperNeff, of course, is thinking in terms of five-year plans.