The Idea
A strategic model shows how to buy or sell a large position most efficiently.
The Research
Optimal execution — how to buy or sell a large stock or position in the markets at the best price — is a classic problem of financial engineering. When an institutional investor or hedge fund buys or sells a large position, the transaction tends to move the markets. For example, if a mutual fund sells 100,000 shares of Google, the price of shares of Google will decline, and the fund will be forced to accept a lower price. (A dramatic, if extreme, example took place in January 2008, when Société Générale needed to liquidate €50 billion euros in futures positions racked up by a rogue trader. Liquidating the positions led to a €4.9 billion [US$7.5 billion] loss, the biggest in banking history.)
Previous researchers who have developed models for optimal execution generally have concluded that the best way to avoid moving the markets is to spread trades over time, such as throughout a day. In theory, this gives supply and demand a chance to balance out, minimizing the effects of large trades.
However, these models did not account for a significant problem. Suppose the mutual fund selling the 100,000 shares of Google spreads its trades over one day, selling a small chunk every five minutes. Very quickly, hedge funds and other traders on the watch for this sort of planned trading would notice and could take advantage by front running, or shorting the security ahead of the next planned sale.
Professor Ciamac Moallemi, working with Beomsoo Park and Benjamin Van Roy of Stanford University, explored whether there might be a way to balance the tension between the need to spread trading to get the best price and the risk of leaking information to competitors. Using game theory, they came up with a model that integrates strategy and statistics. In contrast to previous researchers’ models, which used algorithms to determine how to stagger trading, the model devised by Moallemi and his coresearchers has a strategic component that provides a set of tactics for sellers and competitors.
Practical Applications
Hedge funds, institutional investors
You can use this research to make trades most efficiently, applying this model in lieu of an algorithmic model to execute large positions in a more strategic manner less likely to tip off competitors.
READ THE RESEARCH
Ciamac Moallemi. "The Execution Game."
ABOUT THE RESEARCHER
Ciamac Moallemi
Ciamac C. Moallemi is the William Von Mueffling Professor of Business in the Decision, Risk, and Operations Division of the Graduate School of Business...