Dan Iancu: Tapping a Moral Philosopher to Solve a Money Manager’s Dilemma

Written

Dan Iancu: Tapping a Moral Philosopher to Solve a Money Manager’s Dilemma

A Stanford scholar examines how to juggle needs of diverse clients amid uncertainty.
Woman sitting on a museum bench looking a sculptures
A scholar bridges philosophy with portfolio management. | Reuters/Yannis Behrakis

On one level, it is a highly technical solution to a narrow-bore problem in modern portfolio management. On another level, it’s a problem that has preoccupied social philosophers for centuries: How do we balance fairness to individuals with the goal of doing what’s best for society as a whole?

Perhaps it’s no surprise, then, that Dan Iancu of Stanford Graduate School of Business has tapped into the theories of a renowned philosopher of social justice — the late John Rawls.

Iancu, an assistant professor of operations, information, and technology, studies complex problems that involve juggling the needs of multiple players amid conditions of uncertainty and risk.

Much of this falls under the rubric of “dynamic optimization.” How can companies minimize risk in supply chains with many levels of contractors and subcontractors? How do you avoid perverse incentives in financial covenants for retailers who borrow money to finance their inventories?

But in a new paper, Iancu and Nikolaos Trichakis at Harvard Business School tackle a practical ethical dilemma that many money managers face when carrying out trades for multiple portfolios with different goals.

Though investors aren’t usually aware of it, portfolio managers frequently bundle the trades from multiple clients to improve efficiency.

The problem is how to allocate the transaction costs. Suppose that one client wants to immediately buy 10,000 shares of Wal-Mart, while a second client wants to buy only 10 shares and is willing to be patient. If that big order moves the market and drives up the price of Wal-Mart shares, the small investor may have to pay a higher price than if her purchase had been carried out at a more leisurely pace. The second investor could lose the price benefit that comes from being patient.

What is the fair thing to do? The standard approach right now is to bundle the trades together and then allocate the costs based on each client’s share of the total trade. It seems fair because the portfolio manager is aiming to maximize profits for the group of investors as a whole. In practice, however, it may be unnecessarily costly to individual clients.

To figure out a solution, Iancu and Trichakis devised a model of portfolio management that allows the manager to choose between a blend of two different ethical approaches.

The first approach is to maximize the good of the whole group. In social terms, it’s what utilitarian philosophers would describe as seeking the greatest good for the greatest number of people. For a portfolio manager, it means maximizing the total returns for the whole group.

The problem is that maximizing the group’s total “happiness” may not be fair to the individuals. Imagine two clients, a big investor and a small one, who both are interested in buying shares of Target and Wal-Mart.

The quickest way to maximize total happiness might well be to carry out all the orders on the spot, because that would make the one big investor extremely happy. But that could easily be unfair to the small investor, who could face a higher stock price as a result of being lumped in with the whale-size buyer.

The alternative approach is to focus on fairness.

This is where John Rawls comes in. As the author of major works on social and political ethics, most famously A Theory of Justice, Rawls argued that the best approach was to focus on delivering the most possible happiness to those who were worst off.

For the two clients who both want shares of Wal-Mart and Target, a “fairness” approach could be to have the small investor buy Target while the big investor buys Wal-Mart. That would prevent overlap between the two and protect the small investor from being big-footed.

In short, Iancu and Trichakis devised a model for portfolio management that allows a manager to blend the “greatest good” approach and the “fairness’’ approach. They then ran computer simulations with hypothetical portfolios to test the extreme versions of each approach against the other.

The big surprise: The Rawlsian approach was not only fairer to the small investor, but also almost as efficient and beneficial for the big investor. In the case of the Wal-Mart and Target investors, the small investor was protected and the disadvantages to the big investor were very small.

Iancu and Trichakis don’t have a simple intuitive explanation for these results. They also caution that they aren’t advocating a pure Rawlsian approach. Their model is open to all manner of combinations between maximizing fairness and maximizing social welfare.

But economists usually assume there is a natural conflict between maximizing fairness and maximizing total returns. It turns out, says Iancu, that at least in this particular financial problem, the trade-off isn’t that large after all.

Dan Iancu is an assistant professor of operations, information, and technology at Stanford Graduate School of Business.

For media inquiries, visit the Newsroom.
Explore More

Insights

People working in a call center
August 9, 2017
Written

How Digital Nervous Systems Can Raise Your Organizational IQ

As technology accelerates the flow of information, businesses have a new role model: computer networks.

Insights

A hospital waiting room
August 7, 2017
Written

Why Hospital ER Wait Times Are Often Wrong

Research shows how to improve a system that doesn’t work.

Insights

The End of Privacy, Part 1: Data Scientists Know All Your Data | Art by Tricia Seibold with photo by iStock/Bim
May 9, 2017
Video

Michal Kosinski: The End of Privacy

In part one of a three part series, a Stanford GSB professor discusses a controversial algorithm that knows more about you than your best friend.