ANN ARBOR—Information continues to flow between departments at securities firms despite laws designed to prevent potential conflicts of interest and insider trading, according to a new University of Michigan Business School study.

The study casts doubt on the efficacy of so-called “Chinese walls,” intended to prevent securities firms from exchanging information between their market-making and brokerage activities and their corporate finance, investment banking and advisory functions, as mandated by law.

“Our study does not support the logic of the recent deregulation in financial services firms that assumes Chinese walls are effective and appropriate,” said H. Nejat Seyhun, professor of finance at the U-M Business School. “Recent deregulation has moved away from complete separation of various functions under separate corporate ownerships and allowed the same firm to engage in multi-service activities provided they are separated by a Chinese wall.

“But our evidence shows that these walls do not work effectively and that information flows between departments. Chinese walls need to be reinforced and measures are needed for increased monitoring and increased sanctions for violations.”

Using data of all insider transactions in publicly listed firms from 1975 to 2000, the study tests the effectiveness of Chinese walls when securities firms appoint inside representatives to their client firms (usually to serve on their board of directors). It also examines the changes in the bid-ask spread (the difference between quoted bid and ask prices, or what investors are willing to pay and what sellers are asking for in terms of stock price) of client firms when representatives of securities firms join their boards.

While Seyhun found that the presence of securities firms representatives on corporate boards does not discourage insider trading in client firms—in fact, such transactions, especially among smaller firms, are three-to-four times as frequent during this time—their presence does prevent insiders from trading profitably. Seyhun used four different methods of statistical analysis to determine the profitability of insider trading in various holding periods (ranging from one to 12 months) before, during and after securities firms representatives joined the boards of client firms.

He also compared the insiders’ performance during those periods with insiders at firms without securities representatives. He found that the profitability level of insiders at client firms with securities representatives was similar before the latter joined their boards (abnormal profits averaged 1.26 percent after one month and 4.96 percent after 12 months) to the profitability of insiders at firms that had no securities representatives on their boards (abnormal profits averaged .93 percent after one month and 4.32 percent after 12 months). But during the presence of securities representatives on firms’ boards, client insiders’ abnormal profits dropped dramatically, ranging from .18 percent after one month to 1.11 percent after 12 months, Seyhun says.

“If Chinese walls are effective and the inside representative of the securities firms do not convey any information back to the securities firm, then the unaffiliated corporate officers of the client firms should enjoy informational advantages and should be able to earn profits as before,” Seyhun said. “However, when the representatives of the securities firms are present on the board of directors, the ability of unaffiliated insiders to trade profitably is completely eliminated. After the departure of the representatives, client-firm insiders once again are able to trade profitably.”

Seyhun’s study also found that the presence of securities firms representatives on boards of directors reduces the bid-ask spread of client firms’ stocks. The dealer’s bid-ask spread provides compensation for inventory costs, order processing and risk-aversion, as well as losses to more informed traders. If losses to informed traders comprise a significant part of the bid-ask spread, then having a representative on the board of the client firm would reduce these costs, Seyhun says. To the extent the securities firms exploit their special connection with the client firms, inside information is reflected more quickly and fully in the stock prices, he adds. “Consequently, we would expect the information gap of the bid-ask spread to be smaller when the representatives of the securities firms are appointed to the board of directors,” Seyhun said. “Indeed, the arrival of the representatives of the securities firms reduces the bid-ask spread by about 50 percent, while the departure of the representatives restores the bid-ask spread back to its original level.

“Overall, our evidence suggests that Chinese walls are not effective. The presence of the securities firms representatives reduces informational asymmetries for the client firms, eliminates the ability of the client-firm insiders to trade profitably, and reduces the bid-ask spread, as well as the volatility of the stock returns in the client firms’ stocks.”